CN118262203A - Contraband detection method, device, computer equipment and storage medium - Google Patents
Contraband detection method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN118262203A CN118262203A CN202211701081.9A CN202211701081A CN118262203A CN 118262203 A CN118262203 A CN 118262203A CN 202211701081 A CN202211701081 A CN 202211701081A CN 118262203 A CN118262203 A CN 118262203A
- Authority
- CN
- China
- Prior art keywords
- feature
- detected
- images
- feature extraction
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 130
- 230000004927 fusion Effects 0.000 claims abstract description 129
- 238000000605 extraction Methods 0.000 claims description 138
- 238000010586 diagram Methods 0.000 claims description 32
- 238000000034 method Methods 0.000 claims description 27
- 230000004913 activation Effects 0.000 claims description 24
- 238000011176 pooling Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 15
- 238000013507 mapping Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 239000000463 material Substances 0.000 abstract description 5
- 238000007689 inspection Methods 0.000 description 9
- 238000012549 training Methods 0.000 description 8
- 230000003213 activating effect Effects 0.000 description 3
- 230000006854 communication Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000010438 heat treatment Methods 0.000 description 3
- 230000007175 bidirectional communication Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000000843 powder Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a contraband detection method, a device, computer equipment and a storage medium, wherein the contraband detection method comprises the following steps: acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types; respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected; performing feature fusion on the first feature images to obtain second feature images of the object to be detected; and determining a contraband detection result of the article to be detected based on the plurality of second feature maps. According to the embodiment of the application, the contraband detection result of the article to be detected is determined based on the plurality of second feature maps, so that the cost of manpower and material resources can be reduced, and the detection efficiency of the contraband and the accuracy of the contraband detection result are improved.
Description
Technical Field
The application relates to the technical field of computer vision, in particular to a method and a device for detecting contraband, computer equipment and a storage medium.
Background
In recent years, with the rapid development of transportation and logistics industry, the phenomenon of transporting contraband by using express delivery and logistics entrainment is endless, and detection of contraband based on X-ray is an important component of transportation and logistics. The traditional contraband detection method is that an X-ray image of an object to be detected is obtained through a security inspection machine, and then the X-ray image of the object to be detected is observed manually by a professional security inspection person at the security inspection machine to confirm whether the object to be detected contains the contraband.
The traditional contraband detection method needs to consume great manpower and material resources, the working efficiency is reduced along with the increase of the fatigue degree of security inspection personnel, the condition of missing inspection and false inspection is easy to occur, and the detection result is inaccurate because the objects in the X-ray images are mutually overlapped in a semitransparent state.
Disclosure of Invention
The embodiment of the application provides a contraband detection method, a device, computer equipment and a storage medium, which can reduce the cost of manpower and material resources and improve the detection efficiency of the contraband and the accuracy of a detection result of the contraband.
In one aspect, the present application provides a method for detecting contraband, the method comprising:
Acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types;
Respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
performing feature fusion on the first feature images to obtain second feature images of the object to be detected;
And determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
In some embodiments of the present application, the feature fusion of the first feature maps to obtain second feature maps of the object to be detected includes:
performing feature stitching on the plurality of first feature graphs to obtain a stitching feature graph of the object to be detected, wherein the stitching feature graph comprises features of a plurality of channels;
weighting the characteristics of each channel in the plurality of channels to obtain a channel attention map of the object to be detected;
and carrying out feature extraction on the channel attention map to obtain a plurality of second feature maps of the object to be detected.
In some embodiments of the present application, the contraband detection method is applied to a contraband detection model, the contraband detection model includes a first feature extraction module, a plurality of spatial attention modules and a feature fusion module, the feature extraction is performed on the channel attention map to obtain a plurality of second feature maps of the object to be detected, including:
Inputting the channel attention map into the first feature extraction module, and carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales;
respectively inputting the plurality of third feature images into a plurality of spatial attention modules, and respectively processing the plurality of third feature images through the plurality of spatial attention modules to obtain a plurality of fourth feature images;
inputting the plurality of fourth feature images into the feature fusion module, and carrying out feature fusion on the plurality of fourth feature images through the feature fusion module to obtain a plurality of second feature images of the object to be detected.
In some embodiments of the present application, the first feature extraction module includes a first feature extraction unit, a second feature extraction unit, a third feature extraction unit, and a fourth feature extraction unit that are sequentially cascaded, where the first feature extraction unit, the second feature extraction unit, the third feature extraction unit, and the fourth feature extraction unit correspond to the plurality of spatial attention modules, the channel attention map is an input of the first feature extraction unit, an output of the first feature extraction unit is an input of the spatial attention module corresponding to the first feature extraction unit, and an input of the second feature extraction unit, an output of the second feature extraction unit is an input of the spatial attention module corresponding to the second feature extraction unit, and an input of the third feature extraction unit, an output of the third feature extraction unit is an input of the spatial attention module corresponding to the third feature extraction unit, and an input of the fourth feature extraction unit, and an output of the fourth feature extraction unit is an input of the spatial attention module corresponding to the fourth feature extraction unit.
In some embodiments of the present application, the feature fusion module includes a first feature fusion unit, a second feature fusion unit, a third feature fusion unit, and a fourth feature fusion unit that are sequentially cascaded, where the first feature fusion unit, the second feature fusion unit, the third feature fusion unit, and the fourth feature fusion unit correspond to the plurality of spatial attention modules, the input item of the first feature fusion unit is an output item of the spatial attention module corresponding to the first feature fusion unit, the input item of the second feature fusion unit is an output item of the spatial attention module corresponding to the second feature fusion unit, and an output item of the first feature fusion unit, the input item of the third feature fusion unit is an output item of the spatial attention module corresponding to the third feature fusion unit, and the output item of the second feature fusion unit, and the input item of the fourth feature fusion unit is an output item of the spatial attention module corresponding to the fourth feature fusion unit, and an output item of the third feature fusion unit.
In some embodiments of the present application, each of the spatial attention modules includes a pooling layer, an activation layer, and a fusion layer, where the inputting the plurality of third feature maps into the plurality of spatial attention modules respectively, and processing the plurality of third feature maps through the plurality of spatial attention modules respectively, to obtain a plurality of fourth feature maps includes:
inputting the third feature map into the pooling layer, and performing feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map;
inputting the mapped third feature map into the activation layer, and acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, wherein the spatial attention map is used for representing the importance degree of pixel points at different positions of the third feature map;
And inputting the spatial attention force diagram into the fusion layer, and multiplying the spatial attention force diagram with the third characteristic diagram through the fusion layer to obtain a fourth characteristic diagram.
In some embodiments of the present application, the contraband detection model further includes a plurality of second feature extraction modules, and the feature extraction is performed on the plurality of images to be detected, so as to obtain a plurality of first feature graphs of the object to be detected, including:
Inputting a plurality of images to be detected into a plurality of second feature extraction modules respectively, and carrying out feature extraction on the plurality of images to be detected through the plurality of second feature extraction modules respectively to obtain a plurality of first feature images of the objects to be detected.
In another aspect, the present application provides a contraband detection apparatus comprising:
The image acquisition unit is used for acquiring a plurality of images to be detected of the object to be detected, wherein the images to be detected are images with the same image content and different image types;
the feature extraction unit is used for respectively extracting features of the plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
The feature fusion unit is used for carrying out feature fusion on the plurality of first feature images to obtain a plurality of second feature images of the object to be detected;
and the contraband detection unit is used for determining the contraband detection result of the article to be detected based on the plurality of second feature graphs.
In some embodiments of the present application, the contraband detection method is applied to a contraband detection model comprising a plurality of second feature extraction modules, the feature extraction unit being specifically configured to:
Inputting a plurality of images to be detected into a plurality of second feature extraction modules respectively, and carrying out feature extraction on the plurality of images to be detected through the plurality of second feature extraction modules respectively to obtain a plurality of first feature images of the objects to be detected.
In some embodiments of the application, the feature fusion unit is specifically for:
performing feature stitching on the plurality of first feature graphs to obtain a stitching feature graph of the object to be detected, wherein the stitching feature graph comprises features of a plurality of channels;
weighting the characteristics of each channel in the plurality of channels to obtain a channel attention map of the object to be detected;
and carrying out feature extraction on the channel attention map to obtain a plurality of second feature maps of the object to be detected.
In some embodiments of the present application, the contraband detection model includes a first feature extraction module, a plurality of spatial attention modules, and a feature fusion module, where the feature fusion unit is specifically configured to:
Inputting the channel attention map into the first feature extraction module, and carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales;
respectively inputting the plurality of third feature images into a plurality of spatial attention modules, and respectively processing the plurality of third feature images through the plurality of spatial attention modules to obtain a plurality of fourth feature images;
inputting the plurality of fourth feature images into the feature fusion module, and carrying out feature fusion on the plurality of fourth feature images through the feature fusion module to obtain a plurality of second feature images of the object to be detected.
In some embodiments of the present application, each spatial attention module includes a pooling layer, an activation layer, and a fusion layer, and the feature fusion unit is specifically further configured to:
inputting the third feature map into the pooling layer, and performing feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map;
inputting the mapped third feature map into the activation layer, and acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, wherein the spatial attention map is used for representing the importance degree of pixel points at different positions of the third feature map;
And inputting the spatial attention force diagram into the fusion layer, and multiplying the spatial attention force diagram with the third characteristic diagram through the fusion layer to obtain a fourth characteristic diagram.
In another aspect, the present application also provides a computer apparatus, including:
One or more processors;
a memory; and
One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the contraband detection method of any of the first aspects.
In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program to be loaded by a processor for performing the steps of the contraband detection method of any of the first aspects.
The method and the device for detecting the contraband based on the images to be detected of different types can solve the problem that the detection result is inaccurate due to the fact that articles in the X-ray images are overlapped with each other in a semitransparent state, and because the plurality of second feature images are fused with the features of the images to be detected of different types, the detection result of the contraband is determined based on the plurality of second feature images, the accuracy of the detection result of the contraband can be improved, the detection result is independent of security check staff, the cost of manpower and material resources is reduced, the detection efficiency is high, and the detection result is accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a contraband detection system according to an embodiment of the present application;
FIG. 2 is a flow chart of one embodiment of a method for contraband detection provided in an embodiment of the present application;
fig. 3 is a schematic flow chart of a specific embodiment of a method for detecting contraband provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of a specific structure of a contraband detection model provided in an embodiment of the present application;
FIG. 5 is a schematic diagram of a specific flow chart of feature map processing performed by the spatial attention module according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an embodiment of a contraband detection apparatus provided in an embodiment of the present application;
FIG. 7 is a schematic diagram of an embodiment of a computer device provided in an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application.
In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on the drawings are merely for convenience in describing the present application and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present application. Furthermore, the terms "first," "second," "third," "fourth" and the like are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first", "second", "third", "fourth" may include one or more of the described features, either explicitly or implicitly. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the present application, the term "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "exemplary" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
It should be noted that, because the method of the embodiment of the present application is executed in the computer device, the processing objects of each computer device exist in the form of data or information, for example, time, which is essentially time information, it can be understood that in the subsequent embodiment, if the size, the number, the position, etc. are all corresponding data, so that the computer device can process the data, which is not described herein in detail.
The embodiment of the application provides a contraband detection method, a contraband detection device, computer equipment and a storage medium, and the method, the device and the computer equipment and the storage medium are respectively described in detail below.
Referring to fig. 1, fig. 1 is a schematic diagram of a contraband detection system according to an embodiment of the present application, where the contraband detection system may include a computer device 100, and a contraband detection apparatus, such as the computer device in fig. 1, is integrated in the computer device 100.
The computer device 100 in the embodiment of the application is mainly used for acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types; respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected; performing feature fusion on the first feature images to obtain second feature images of the object to be detected; based on a plurality of the second feature graphs, the contraband detection results of the articles to be detected are determined, so that the cost of manpower and material resources can be reduced, and the detection efficiency of the contraband and the accuracy of the contraband detection results are improved.
In the embodiment of the present application, the computer device 100 may be an independent server, or may be a server network or a server cluster formed by servers, for example, the computer device 100 described in the embodiment of the present application includes, but is not limited to, a computer, a network host, a single network server, a plurality of network server sets, or a cloud server formed by a plurality of servers. Wherein the Cloud server is composed of a large number of computers or web servers based on Cloud Computing (Cloud Computing).
It is to be understood that the computer device 100 used in embodiments of the present application may be a device that includes both receive and transmit hardware, i.e., a device having receive and transmit hardware capable of performing bi-directional communications over a bi-directional communication link. Such a device may include: a cellular or other communication device having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. The computer device 100 may be a desktop terminal or a mobile terminal, and the computer device 100 may be one of a mobile phone, a tablet computer, a notebook computer, and the like.
It will be appreciated by those skilled in the art that the application environment shown in fig. 1 is merely an application scenario with the present application, and is not limited to the application scenario with the present application, and that other application environments may include more or fewer computer devices than those shown in fig. 1, for example, only 1 computer device is shown in fig. 1, and that the contraband detection system may further include one or more other services, which are not limited herein.
In addition, as shown in fig. 1, the contraband detection system may further include a memory 200 for storing data, such as various images to be detected, for example, RGB images, gray scale images, edge images, high energy X-ray images, low energy X-ray images, etc., such as contraband detection results, for example, coordinates of contraband in the images to be detected, types of contraband, etc.
It should be noted that, the schematic view of the scenario of the contraband detection system shown in fig. 1 is only an example, and the contraband detection system and the scenario described in the embodiments of the present application are for more clearly describing the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application, and those skilled in the art can know that, with the evolution of the contraband detection system and the appearance of the new service scenario, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.
Firstly, in the embodiment of the present application, a contraband detection method is provided, an execution subject of the contraband detection method is a contraband detection apparatus, the contraband detection apparatus is applied to a computer device, and the contraband detection method includes: acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types; respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected; performing feature fusion on the first feature images to obtain second feature images of the object to be detected; and determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
As shown in fig. 2, which is a schematic flow chart of an embodiment of a method for detecting contraband according to the present application, the method for detecting contraband may include the following steps 301 to 304, which are specifically as follows:
301. And acquiring a plurality of images to be detected of the object to be detected, wherein the images to be detected are images with the same image content and different image types.
The article to be detected is an article that needs to be detected whether the article contains contraband, and the article to be detected includes but is not limited to packages, baggage and the like. The various images to be detected are images obtained by image acquisition of the object to be detected through the image acquisition device, the various images to be detected are images with the same image content and different image types, and the various images to be detected comprise, but are not limited to, RGB images, gray scale images, high-energy X-ray images, low-energy X-ray images, LAB images, HSV images, edge images, laplacian filtered images and the like. For example, when the various images to be detected include an RGB image, a high-energy X-ray image, a low-energy X-ray image, and other types of images, the object to be detected may be placed on a security inspection machine, the RGB image, the high-energy X-ray image, and the low-energy X-ray image of the object to be detected are collected by the security inspection machine, and then the RGB image is subjected to image processing to obtain the other types of images of the object to be detected.
For example, when the plurality of images to be detected include RGB images, gray-scale images, edge images, high-energy X-ray images and low-energy X-ray images, the RGB images, the high-energy X-ray images and the low-energy X-ray images of the object to be detected can be collected by the security inspection machine, then the RGB images are converted into gray-scale images, and the RGB images are processed by an edge detection algorithm (such as a canny edge detection algorithm), so as to obtain the gray-scale images and the edge images of the object to be detected.
Considering that the articles in the X-ray image overlap each other in a semitransparent state and the different contraband articles have different identities in different types of images, for example, the identification of scissors in an edge image is higher, and the identification of blue-green heating paste in a color image such as RGB or LAB is higher. When the contraband detection is carried out on the article to be detected, a plurality of images to be detected of the article to be detected are firstly obtained, the contraband detection is carried out on the article to be detected based on the plurality of images to be detected, the problem that the contraband detection result is inaccurate due to the fact that articles in the X-ray images are overlapped with each other in a semitransparent state and the different contraband is different in identification degree under different types of images can be solved, and the accuracy of the contraband detection result is improved.
302. And respectively extracting the characteristics of the plurality of images to be detected to obtain a plurality of first characteristic diagrams of the objects to be detected.
Considering that the feature dimensions of each image to be detected are not identical, for example, the feature dimension of an RGB image is h×w×3, the feature dimension of a gray image is h×w×1, where H and W represent the height and width of the image, respectively, and 3 and 1 represent the number of channels of the image. After obtaining multiple images to be detected of the object to be detected, the embodiment respectively performs feature extraction on the multiple images to be detected to obtain multiple first feature images of the object to be detected.
In a specific embodiment, the contraband detection method is applied to a contraband detection model, where the contraband detection model includes a plurality of second feature extraction modules, as shown in fig. 3, and in step 302, feature extraction is performed on a plurality of images to be detected, so as to obtain a plurality of first feature graphs of the object to be detected, which may include the following step 401, specifically as follows:
401. Inputting a plurality of images to be detected into a plurality of second feature extraction modules respectively, and carrying out feature extraction on the plurality of images to be detected through the plurality of second feature extraction modules respectively to obtain a plurality of first feature images of the objects to be detected.
The contraband detection method is applied to a contraband detection model, wherein the contraband detection model is a pre-trained network model for detecting contraband of an object to be detected, the contraband detection model is obtained by training a preset network model based on a pre-acquired training sample set, and the preset network model can be a deep learning model or a machine learning model, such as a convolutional neural network (Convolutional Neural Networks, CNN), a deconvolution neural network (De-Convolutional Networks, DN) and the like.
The pre-acquired training sample set comprises a plurality of groups of training images corresponding to a plurality of predicted articles and a real detection result of each predicted article, each group of training images comprises a plurality of types of predicted images of each predicted article, and the plurality of types of predicted images are images with the same image content and different image types. Accordingly, the training process of the contraband detection model includes: inputting a plurality of groups of training images into a preset network model, outputting a prediction detection result of each predicted article through the preset network model, determining a loss value according to the prediction detection result, a real detection result and a loss function of the preset network model, correcting model parameters of the preset network model according to a preset parameter learning rate when the loss value does not meet preset conditions, and continuously executing the steps of inputting the plurality of groups of training images into the preset network model, and outputting the prediction detection result of each predicted article through the preset network model until the loss value meets the preset conditions. The loss value meeting the preset condition may be that the loss value is smaller than a preset first threshold, or that a difference value between two obtained loss values is smaller than a preset second threshold.
As shown in fig. 4, the contraband detection model includes a plurality of second feature extraction modules, where the plurality of second feature extraction modules are respectively used for extracting features of a plurality of images to be detected. Correspondingly, the steps of respectively extracting the characteristics of the various images to be detected are as follows: and respectively inputting the multiple images to be detected into multiple second feature extraction modules, and respectively carrying out feature extraction on the multiple images to be detected through the multiple second feature extraction modules to obtain multiple first feature images of the objects to be detected.
For example, a plurality of images to be detected are respectively input into a plurality of second feature extraction modules, and a plurality of first feature graphs with dimensions of H, W and 1 are output through the plurality of second feature extraction modules.
303. And carrying out feature fusion on the plurality of first feature images to obtain a plurality of second feature images of the object to be detected.
After the multiple first feature maps of the object to be detected are obtained, feature fusion is performed on the multiple first feature maps to obtain multiple second feature maps of the object to be detected, and because the multiple second feature maps are feature maps obtained by fusing image features of multiple images to be detected, contraband detection is performed based on the multiple second feature maps, and accuracy of a contraband detection result can be improved.
In a specific embodiment, with continued reference to fig. 3, the feature fusion of the first feature maps in step 303 to obtain the second feature maps of the object to be detected may include the following steps 402 to 404, which are specifically as follows:
402. Performing feature stitching on the plurality of first feature graphs to obtain a stitching feature graph of the object to be detected, wherein the stitching feature graph comprises features of a plurality of channels;
403. Weighting the characteristics of each channel in the plurality of channels to obtain a channel attention map of the object to be detected;
404. And carrying out feature extraction on the channel attention map to obtain a plurality of second feature maps of the object to be detected.
In contraband detection, the basis of discrimination of different contraband is different, for example, in scissors imaging, the edge contour of an object is taken as the main basis of discrimination, the output characteristics of the edge image are given a larger weight, and in the determination of powder and heating paste, both are granular, but the heating paste is mainly blue-green and the powder is mainly orange-yellow, and color images such as RGB or LAB are given a larger weight.
In order to obtain salient features related to contraband in the first feature maps, in this embodiment, after obtaining the first feature maps of the to-be-detected article, feature stitching is performed on the first feature maps to obtain a stitched feature map of the to-be-detected article, where the stitched feature map includes features of multiple channels, then weighting is performed on features of each channel in the multiple channels of the stitched feature map to obtain a channel attention map of the to-be-detected article, and finally feature extraction is performed on the channel attention map to obtain the second feature maps of the to-be-detected article. The characteristics of each channel in the plurality of channels of the spliced characteristic map are weighted, so that the characteristics of some unimportant channels can be weakened, the characteristics of other important channels can be enhanced, and the accuracy of the contraband detection result can be improved.
With continued reference to fig. 4, the contraband detection model further includes a feature stitching module, where the plurality of first feature graphs are input terms of the feature stitching module, and output terms of the feature stitching module are stitched feature graphs. Correspondingly, the characteristic splicing is carried out on the plurality of first characteristic diagrams, and the step of obtaining the spliced characteristic diagram of the object to be detected comprises the following specific steps: inputting the plurality of first feature images into a feature splicing module, and carrying out feature splicing on the plurality of first feature images through the feature splicing module to obtain spliced feature images of the to-be-detected object.
With continued reference to fig. 4, the contraband detection model further includes a channel attention module, the spliced feature map is an input item of the channel attention module, the channel attention map is an output item of the channel attention module, and accordingly, the steps of weighting the features of each channel in the plurality of channels to obtain the channel attention map of the object to be detected are specifically as follows: and inputting the spliced characteristic diagram into a channel attention module, and weighting the characteristics of each channel in the plurality of channels by the channel attention module to obtain the channel attention diagram of the object to be detected.
In a specific implementation manner, when the channel attention module performs weighting processing on the characteristics of each channel in the multiple channels, the global averaging may be performed on the characteristics of each channel in the multiple channels to obtain global characteristics of each channel, then weight information of each channel characteristic is determined according to the global characteristics of each channel, and the weight information of each channel characteristic is multiplied by the characteristics of the corresponding channel in the spliced feature map to obtain a channel attention map.
In a specific embodiment, the forbidden article detection model includes a first feature extraction module, a plurality of spatial attention modules, and a feature fusion module, and in step 404, the feature extraction is performed on the channel attention map to obtain a plurality of second feature maps of the to-be-detected article, which may include the following steps 501 to 503, specifically as follows:
501. Inputting the channel attention map into the first feature extraction module, and carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales;
502. Respectively inputting the plurality of third feature images into a plurality of spatial attention modules, and respectively processing the plurality of third feature images through the plurality of spatial attention modules to obtain a plurality of fourth feature images;
503. Inputting the plurality of fourth feature images into the feature fusion module, and carrying out feature fusion on the plurality of fourth feature images through the feature fusion module to obtain a plurality of second feature images of the object to be detected.
The contraband detection model further comprises a first feature extraction module, a plurality of spatial attention modules and a feature fusion module, the channel attention is intended to be an input item of the first feature extraction module, an output item of the first feature extraction module is an input item of the plurality of spatial attention modules, an output item of the spatial attention module is an input item of the feature fusion module, and an output item of the feature fusion module is a plurality of second feature graphs.
Correspondingly, the steps for carrying out feature extraction on the channel attention map are specifically as follows: inputting the channel attention map into a first feature extraction module, carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales, then respectively inputting the plurality of third feature maps into a plurality of spatial attention modules, respectively processing the plurality of third feature maps through the plurality of spatial attention modules to obtain a plurality of fourth feature maps, finally inputting the plurality of fourth feature maps into a feature fusion module, and carrying out feature fusion on the plurality of fourth feature maps through a feature fusion module to obtain a plurality of second feature maps of the object to be detected.
With continued reference to fig. 4, the first feature extraction module includes a first feature extraction unit, a second feature extraction unit, a third feature extraction unit, and a fourth feature extraction unit that are sequentially cascaded, where the first feature extraction unit, the second feature extraction unit, the third feature extraction unit, and the fourth feature extraction unit correspond to the plurality of spatial attention modules, the channel attention is intended to be an input item of the first feature extraction unit, an output item of the first feature extraction unit is an input item of the spatial attention module corresponding to the first feature extraction unit and an input item of the second feature extraction unit, an output item of the second feature extraction unit is an input item of the spatial attention module corresponding to the second feature extraction unit, and an input item of the third feature extraction unit, an output item of the third feature extraction unit is an input item of the spatial attention module corresponding to the third feature extraction unit, and an input item of the fourth feature extraction unit, and an output item of the fourth feature extraction unit is an input item of the spatial attention module corresponding to the fourth feature extraction unit. Based on the first feature extraction module and the plurality of spatial attention modules, the channel attention force diagram is subjected to feature extraction, more information related to the contraband can be obtained, and the accuracy of the contraband detection result is improved.
With continued reference to fig. 4, the feature fusion module includes a first feature fusion unit, a second feature fusion unit, a third feature fusion unit, and a fourth feature fusion unit that are sequentially cascaded, where the first feature fusion unit, the second feature fusion unit, the third feature fusion unit, and the fourth feature fusion unit correspond to the plurality of spatial attention modules, respectively, an input item of the first feature fusion unit is an output item of the spatial attention module corresponding to the first feature fusion unit, an input item of the second feature fusion unit is an output item of the spatial attention module corresponding to the second feature fusion unit, and an output item of the first feature fusion unit, an input item of the third feature fusion unit is an output item of the spatial attention module corresponding to the third feature fusion unit, and an input item of the fourth feature fusion unit is an output item of the spatial attention module corresponding to the fourth feature fusion unit, and an output item of the third feature fusion unit.
In a specific embodiment, each spatial attention module includes a pooling layer, an activation layer and a fusion layer, in step 502, a plurality of third feature maps are input into a plurality of spatial attention modules, and the plurality of third feature maps are processed by the plurality of spatial attention modules to obtain a plurality of fourth feature maps, which may include the following steps 601 to 603, specifically as follows:
601. inputting the third feature map into the pooling layer, and performing feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map;
602. Inputting the mapped third feature map into the activation layer, and acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, wherein the spatial attention map is used for representing the importance degree of pixel points at different positions of the third feature map;
603. And inputting the spatial attention force diagram into the fusion layer, and multiplying the spatial attention force diagram with the third characteristic diagram through the fusion layer to obtain a fourth characteristic diagram.
Each spatial attention module comprises a pooling layer, an activating layer and a fusion layer, wherein the third feature map is an input item of the pooling layer, an output item of the pooling layer is an input item of the activating layer, and an output item of the activating layer is an input item of the fusion layer. Correspondingly, the step of processing the third feature map specifically includes: and inputting the third feature map into a pooling layer, carrying out feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map, inputting the mapped third feature map into an activation layer, acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, inputting the spatial attention map into a fusion layer, and carrying out multiplication processing on the spatial attention map and the third feature map through the fusion layer to obtain a fourth feature map. Different weights are given to different pixel positions of the third feature map through the spatial attention module, so that the network can pay attention to the region of the contraband better, and the accuracy of the detection result of the contraband is improved.
As shown in fig. 5, the pooling layer may use a convolution kernel of 1×1, perform feature mapping R h×w×c→Rh×w on a third feature map with dimensions of h×w×c by using the pooling layer, compress the dimensions of the third feature map to h×w, then apply the mapped third feature map to a sigmoid function phi (x) by using the activation layer to obtain weight coefficients of different positions of the third feature map to construct a spatial attention map, and finally perform multiplication operation on the spatial attention map and the third feature map by using the fusion layerRepresenting a spatial attention seeking to multiply the third feature map, a fourth feature map is obtained. Wherein phi (x) can be expressed as
304. And determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
The contraband detection result is a contraband detection result of the article to be detected, which is determined based on the plurality of second feature maps, and the contraband detection result comprises, but is not limited to, the types of the contraband, the pixel coordinates of the contraband on a plurality of images to be detected, and the like. After the multiple second feature maps of the object to be detected are obtained, the contraband detection result of the object to be detected is determined based on the multiple second feature maps, and the contraband detection accuracy can be improved due to the fact that the multiple second feature maps are obtained by fusing the features of multiple images to be detected and the contraband detection is performed based on the multiple second feature maps.
In a specific implementation manner, with continued reference to fig. 4, the contraband detection model further includes an output module, where the plurality of second feature maps are input items of the output module, and the contraband detection result is an output item of the output module. Correspondingly, based on a plurality of second feature graphs, the step of determining the contraband detection result of the object to be detected specifically comprises the following steps: and inputting the plurality of second feature maps into an output module, and outputting a contraband detection result of the article to be detected through the output module.
In order to better implement the contraband detection method according to the embodiment of the present application, on the basis of the contraband detection method, the embodiment of the present application further provides a contraband detection apparatus, as shown in fig. 6, where the contraband detection apparatus 700 includes:
an image acquisition unit 701, configured to acquire a plurality of to-be-detected images of an object to be detected, where the plurality of to-be-detected images are images with the same image content and different image types;
A feature extraction unit 702, configured to perform feature extraction on a plurality of to-be-detected images, to obtain a plurality of first feature graphs of the to-be-detected objects;
A feature fusion unit 703, configured to perform feature fusion on the plurality of first feature maps to obtain a plurality of second feature maps of the object to be detected;
And a contraband detection unit 704, configured to determine a contraband detection result of the item to be detected based on a plurality of the second feature maps.
In the embodiment of the application, the contraband detection is carried out based on various images to be detected in different types, so that the problem of inaccurate detection results caused by the mutual superposition of articles in the X-ray images in a semitransparent state can be solved.
In some embodiments of the present application, the contraband detection method is applied to a contraband detection model, the contraband detection model includes a plurality of second feature extraction modules, and the feature extraction unit 702 is specifically configured to:
Inputting a plurality of images to be detected into a plurality of second feature extraction modules respectively, and carrying out feature extraction on the plurality of images to be detected through the plurality of second feature extraction modules respectively to obtain a plurality of first feature images of the objects to be detected.
In some embodiments of the present application, the feature fusion unit 703 is specifically configured to:
performing feature stitching on the plurality of first feature graphs to obtain a stitching feature graph of the object to be detected, wherein the stitching feature graph comprises features of a plurality of channels;
weighting the characteristics of each channel in the plurality of channels to obtain a channel attention map of the object to be detected;
and carrying out feature extraction on the channel attention map to obtain a plurality of second feature maps of the object to be detected.
In some embodiments of the present application, the contraband detection model includes a first feature extraction module, a plurality of spatial attention modules, and a feature fusion module, where the feature fusion unit 703 is specifically further configured to:
Inputting the channel attention map into the first feature extraction module, and carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales;
respectively inputting the plurality of third feature images into a plurality of spatial attention modules, and respectively processing the plurality of third feature images through the plurality of spatial attention modules to obtain a plurality of fourth feature images;
inputting the plurality of fourth feature images into the feature fusion module, and carrying out feature fusion on the plurality of fourth feature images through the feature fusion module to obtain a plurality of second feature images of the object to be detected.
In some embodiments of the present application, each of the spatial attention modules includes a pooling layer, an activation layer, and a fusion layer, where the inputting the plurality of third feature maps into the plurality of spatial attention modules respectively, and the feature fusion unit 703 is specifically further configured to:
inputting the third feature map into the pooling layer, and performing feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map;
inputting the mapped third feature map into the activation layer, and acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, wherein the spatial attention map is used for representing the importance degree of pixel points at different positions of the third feature map;
And inputting the spatial attention force diagram into the fusion layer, and multiplying the spatial attention force diagram with the third characteristic diagram through the fusion layer to obtain a fourth characteristic diagram.
The embodiment of the application also provides a computer device which integrates any of the contraband detection apparatuses provided by the embodiment of the application, and the computer device comprises:
One or more processors;
a memory; and
One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to perform the steps of the contraband detection method described in any of the embodiments of the contraband detection method described above.
The embodiment of the application also provides computer equipment which integrates any contraband detection device provided by the embodiment of the application. As shown in fig. 7, a schematic structural diagram of a computer device according to an embodiment of the present application is shown, specifically:
the computer device may include one or more processing cores 'processors 801, one or more computer-readable storage media's memory 802, power supply 803, and input unit 804, among other components. Those skilled in the art will appreciate that the computer device structure shown in FIG. 7 is not limiting of the computer device and may include more or fewer components than shown, or may be combined with certain components, or a different arrangement of components. Wherein:
The processor 801 is a control center of the computer device, connects various parts of the entire computer device using various interfaces and lines, and performs various functions of the computer device and processes data by running or executing software programs and/or modules stored in the memory 802, and calling data stored in the memory 802, thereby performing overall monitoring of the computer device. Optionally, the processor 801 may include one or more processing cores; preferably, the processor 801 may integrate an application processor that primarily handles operating systems, user interfaces, applications, etc., with a modem processor that primarily handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 801.
The memory 802 may be used to store software programs and modules, and the processor 801 executes various functional applications and data processing by executing the software programs and modules stored in the memory 802. The memory 802 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the computer device, etc. In addition, memory 802 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.
The computer device also includes a power supply 803 for powering the various components, preferably, the power supply 803 can be logically coupled to the processor 801 via a power management system such that functions such as managing charge, discharge, and power consumption can be performed by the power management system. The power supply 803 may also include one or more of any components, such as a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.
The computer device may further comprise an input unit 804, which input unit 804 may be used for receiving input digital or character information and for generating keyboard, mouse, joystick, optical or trackball signal inputs in connection with user settings and function control.
Although not shown, the computer device may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 801 in the computer device loads executable files corresponding to the processes of one or more application programs into the memory 802 according to the following instructions, and the processor 801 executes the application programs stored in the memory 802, so as to implement various functions, as follows:
Acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types;
Respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
performing feature fusion on the first feature images to obtain second feature images of the object to be detected;
And determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.
To this end, an embodiment of the present application provides a computer-readable storage medium, which may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like. On which a computer program is stored, which computer program is loaded by a processor for performing the steps of any of the contraband detection methods provided by the embodiments of the present application. For example, the loading of the computer program by the processor may perform the steps of:
Acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types;
Respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
performing feature fusion on the first feature images to obtain second feature images of the object to be detected;
And determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and the portions of one embodiment that are not described in detail in the foregoing embodiments may be referred to in the foregoing detailed description of other embodiments, which are not described herein again.
In the implementation, each unit or structure may be implemented as an independent entity, or may be implemented as the same entity or several entities in any combination, and the implementation of each unit or structure may be referred to the foregoing method embodiments and will not be repeated herein.
The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.
The above description of the method, the device, the computer equipment and the storage medium for detecting contraband provided by the embodiment of the present application applies specific examples to illustrate the principle and the implementation of the present application, and the description of the above embodiment is only used to help understand the method and the core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, the present description should not be construed as limiting the present application.
Claims (10)
1. A method of contraband detection, the method comprising:
Acquiring a plurality of images to be detected of an object to be detected, wherein the images to be detected are images with the same image content and different image types;
Respectively extracting features of a plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
performing feature fusion on the first feature images to obtain second feature images of the object to be detected;
And determining a contraband detection result of the article to be detected based on the plurality of second feature maps.
2. The method for detecting contraband according to claim 1, wherein the feature fusion of the plurality of first feature maps to obtain a plurality of second feature maps of the object to be detected includes:
performing feature stitching on the plurality of first feature graphs to obtain a stitching feature graph of the object to be detected, wherein the stitching feature graph comprises features of a plurality of channels;
weighting the characteristics of each channel in the plurality of channels to obtain a channel attention map of the object to be detected;
and carrying out feature extraction on the channel attention map to obtain a plurality of second feature maps of the object to be detected.
3. The method for detecting contraband according to claim 2, wherein the method for detecting contraband is applied to a contraband detection model, the contraband detection model includes a first feature extraction module, a plurality of spatial attention modules and a feature fusion module, the feature extraction is performed on the channel attention map, and a plurality of second feature graphs of the object to be detected are obtained, and the method includes:
Inputting the channel attention map into the first feature extraction module, and carrying out feature extraction on the channel attention map through the first feature extraction module to obtain a plurality of third feature maps with different scales;
respectively inputting the plurality of third feature images into a plurality of spatial attention modules, and respectively processing the plurality of third feature images through the plurality of spatial attention modules to obtain a plurality of fourth feature images;
inputting the plurality of fourth feature images into the feature fusion module, and carrying out feature fusion on the plurality of fourth feature images through the feature fusion module to obtain a plurality of second feature images of the object to be detected.
4. The contraband detection method according to claim 3, wherein the first feature extraction module includes a first feature extraction unit, a second feature extraction unit, a third feature extraction unit, and a fourth feature extraction unit that are sequentially cascaded, the first feature extraction unit, the second feature extraction unit, the third feature extraction unit, and the fourth feature extraction unit respectively correspond to the plurality of spatial attention modules, the channel attention attempt is an input item of the first feature extraction unit, an output item of the first feature extraction unit is an input item of a spatial attention module corresponding to the first feature extraction unit, and an input item of the second feature extraction unit, an output item of the second feature extraction unit is an input item of a spatial attention module corresponding to the second feature extraction unit, and an input item of the third feature extraction unit, an output item of the third feature extraction unit is an input item of a spatial attention module corresponding to the third feature extraction unit, and an input item of the fourth feature extraction unit, and an output item of the fourth feature extraction unit is an input item of a spatial attention module corresponding to the fourth feature extraction unit.
5. The contraband detection method according to claim 3, wherein the feature fusion module comprises a first feature fusion unit, a second feature fusion unit, a third feature fusion unit and a fourth feature fusion unit which are sequentially cascaded, the first feature fusion unit, the second feature fusion unit, the third feature fusion unit and the fourth feature fusion unit are respectively corresponding to the plurality of spatial attention modules, an input item of the first feature fusion unit is an output item of the spatial attention module corresponding to the first feature fusion unit, an input item of the second feature fusion unit is an output item of the spatial attention module corresponding to the second feature fusion unit and an output item of the first feature fusion unit, an input item of the third feature fusion unit is an output item of the spatial attention module corresponding to the third feature fusion unit and an output item of the second feature fusion unit, and an input item of the fourth feature fusion unit is an output item of the spatial attention module corresponding to the fourth feature fusion unit and an output item of the third feature fusion unit.
6. The method according to any one of claims 3 to 5, wherein each of the spatial attention modules includes a pooling layer, an activation layer and a fusion layer, the inputting the plurality of third feature maps into the plurality of spatial attention modules respectively, and the processing the plurality of third feature maps through the plurality of spatial attention modules respectively to obtain a plurality of fourth feature maps includes:
inputting the third feature map into the pooling layer, and performing feature mapping on the third feature map through the pooling layer to obtain a mapped third feature map;
inputting the mapped third feature map into the activation layer, and acting the mapped third feature map on an activation function through the activation layer to generate a spatial attention map of the third feature map, wherein the spatial attention map is used for representing the importance degree of pixel points at different positions of the third feature map;
And inputting the spatial attention force diagram into the fusion layer, and multiplying the spatial attention force diagram with the third characteristic diagram through the fusion layer to obtain a fourth characteristic diagram.
7. The method for detecting contraband of claim 3, wherein the contraband detection model further comprises a plurality of second feature extraction modules, the feature extraction is performed on the plurality of images to be detected respectively, and a plurality of first feature graphs of the object to be detected are obtained, including:
Inputting a plurality of images to be detected into a plurality of second feature extraction modules respectively, and carrying out feature extraction on the plurality of images to be detected through the plurality of second feature extraction modules respectively to obtain a plurality of first feature images of the objects to be detected.
8. A contraband detection apparatus, characterized in that the contraband detection apparatus comprises:
The image acquisition unit is used for acquiring a plurality of images to be detected of the object to be detected, wherein the images to be detected are images with the same image content and different image types;
the feature extraction unit is used for respectively extracting features of the plurality of images to be detected to obtain a plurality of first feature images of the objects to be detected;
The feature fusion unit is used for carrying out feature fusion on the plurality of first feature images to obtain a plurality of second feature images of the object to be detected;
and the contraband detection unit is used for determining the contraband detection result of the article to be detected based on the plurality of second feature graphs.
9. A computer device, the computer device comprising:
One or more processors;
a memory; and
One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the contraband detection method of any one of claims 1 to 7.
10. A computer readable storage medium, having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the contraband detection method of any of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211701081.9A CN118262203A (en) | 2022-12-28 | 2022-12-28 | Contraband detection method, device, computer equipment and storage medium |
| PCT/CN2023/133011 WO2024139854A1 (en) | 2022-12-28 | 2023-11-21 | Inspection method and apparatus, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211701081.9A CN118262203A (en) | 2022-12-28 | 2022-12-28 | Contraband detection method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118262203A true CN118262203A (en) | 2024-06-28 |
Family
ID=91611861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211701081.9A Pending CN118262203A (en) | 2022-12-28 | 2022-12-28 | Contraband detection method, device, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN118262203A (en) |
| WO (1) | WO2024139854A1 (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106290423B (en) * | 2016-10-18 | 2024-04-05 | 同方威视技术股份有限公司 | Method, device and system for scanning imaging |
| WO2020023762A1 (en) * | 2018-07-26 | 2020-01-30 | Walmart Apollo, Llc | System and method for produce detection and classification |
| CN113449556B (en) * | 2020-03-26 | 2024-06-04 | 顺丰科技有限公司 | Contraband detection method and device, edge computing equipment and storage medium |
| CN114626443B (en) * | 2022-02-25 | 2024-05-03 | 华南理工大学 | Object rapid detection method based on conditional branching and expert system |
| CN114882446B (en) * | 2022-07-08 | 2022-10-04 | 安徽启新明智科技有限公司 | Image association method, device, equipment and medium |
-
2022
- 2022-12-28 CN CN202211701081.9A patent/CN118262203A/en active Pending
-
2023
- 2023-11-21 WO PCT/CN2023/133011 patent/WO2024139854A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024139854A1 (en) | 2024-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11798132B2 (en) | Image inpainting method and apparatus, computer device, and storage medium | |
| CN110807362B (en) | Image detection method, device and computer readable storage medium | |
| CN110378235A (en) | A kind of fuzzy facial image recognition method, device and terminal device | |
| CN108229418B (en) | Human body key point detection method and apparatus, electronic device, storage medium, and program | |
| CN107688824A (en) | Picture match method and terminal device | |
| CN113887447B (en) | Training and reasoning method and device for density estimation and classification prediction model of dense population targets | |
| CN112528908B (en) | Living body detection method, living body detection device, electronic equipment and storage medium | |
| CN107679475A (en) | Shops's monitoring evaluation method, device and storage medium | |
| CN112464803A (en) | Image comparison method and device | |
| CN108205680A (en) | Image characteristics extraction integrated circuit, method, terminal | |
| CN110415237A (en) | Skin blemishes detection method, detection device, terminal device and readable storage medium storing program for executing | |
| CN109671055A (en) | Pulmonary nodule detection method and device | |
| CN112668675B (en) | Image processing method and device, computer equipment and storage medium | |
| CN118262203A (en) | Contraband detection method, device, computer equipment and storage medium | |
| CN113610460A (en) | Article packaging method, apparatus, electronic device and computer readable medium | |
| CN113822144A (en) | Target detection method and device, computer equipment and storage medium | |
| CN113449559B (en) | Table identification method and device, computer equipment and storage medium | |
| CN107798292A (en) | Object identifying method, computer program, storage medium and electronic equipment | |
| CN115112661B (en) | Defect detection method, device, computer equipment and storage medium | |
| CN115298705A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN116977271A (en) | Defect detection method, model training method, device and electronic equipment | |
| CN114118114A (en) | Image detection method, device and storage medium thereof | |
| CN112990131A (en) | Method, device, equipment and medium for acquiring working gear of voltage change-over switch | |
| CN111428729A (en) | Target detection method and device | |
| CN118279223A (en) | Method and device for determining position of package and related equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |