CN112711999A - Image recognition method, device, equipment and computer storage medium - Google Patents
Image recognition method, device, equipment and computer storage medium Download PDFInfo
- Publication number
- CN112711999A CN112711999A CN202011554905.5A CN202011554905A CN112711999A CN 112711999 A CN112711999 A CN 112711999A CN 202011554905 A CN202011554905 A CN 202011554905A CN 112711999 A CN112711999 A CN 112711999A
- Authority
- CN
- China
- Prior art keywords
- image
- recognized
- preset
- image recognition
- recognition
- 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
- 238000000034 method Methods 0.000 title claims abstract description 66
- 230000006870 function Effects 0.000 claims description 45
- 238000012545 processing Methods 0.000 claims description 42
- 230000009466 transformation Effects 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 230000003321 amplification Effects 0.000 claims description 8
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 8
- 238000013434 data augmentation Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 description 36
- 230000008569 process Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 9
- 230000003416 augmentation Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000007792 addition Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/148—Wavelet transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image identification method, an image identification device, image identification equipment and a computer storage medium. The image recognition method comprises the following steps: acquiring a first image to be recognized, wherein the first image to be recognized comprises a target image to be recognized; according to a first preset wavelet function, denoising a first image to be recognized to obtain a second image to be recognized comprising a target image to be recognized; and acquiring an image recognition result of the second image to be recognized according to the preset image recognition model to obtain the target image. According to the image identification method, the interference noise in the original image can be effectively removed, the useful information in the original image is reserved, and the recognizable image characteristics in the image can be acquired.
Description
Technical Field
The present invention relates to the field of image processing, and in particular, to an image recognition method, apparatus, device, and computer storage medium.
Background
With the rapid development of image recognition and image processing technologies, the image is acquired in real time, and advanced recognition and semantic understanding of target identification in the image are more and more widely applied. For example, in an automatic driving scene, whether the camera sensor can perform effective advanced recognition and semantic understanding on traffic signs on both sides of a road or not is important, and automatic driving in a high-speed scene is important.
Before the image is identified and semantically understood, the quality of the noise reduction preprocessing performed on the image has a great influence on the subsequent image identification and semantically understood effects, so when the image is subjected to the noise reduction processing, how to keep the integrity of useful information in the original image as much as possible and effectively remove useless information in the original image get more and more attention.
Disclosure of Invention
The embodiment of the invention provides an image identification method, an image identification device, image identification equipment and a computer storage medium. The method can effectively remove the interference noise in the original image, retain useful information in the original image and is beneficial to acquiring recognizable image characteristics in the image.
In a first aspect, an embodiment of the present invention provides an image recognition method, where the method includes:
acquiring a first image to be recognized, wherein the first image to be recognized comprises a target image to be recognized;
according to a first preset wavelet function, denoising a first image to be recognized to obtain a second image to be recognized comprising a target image to be recognized;
and acquiring an image recognition result of the second image to be recognized according to the preset image recognition model to obtain the target image.
In some implementations of the first aspect, denoising the first to-be-recognized image according to a first preset wavelet function to obtain a second to-be-recognized image including a target to-be-recognized image, includes:
acquiring an approximate component and a detail component of a first image to be identified according to a first preset wavelet function, wherein the detail component comprises a high-frequency component; acquiring an approximation coefficient of the approximation component and a detail coefficient corresponding to the high-frequency component;
performing Gaussian filtering on the high-frequency components to obtain filtered high-frequency components;
and performing wavelet inverse transformation calculation according to the approximate component, the approximate coefficient, the filtered high-frequency component and the detail coefficient to obtain a second image to be identified.
In some implementation manners of the first aspect, obtaining an image recognition result of the second image to be recognized according to a preset image recognition model includes:
performing wavelet transformation processing on the second image to be identified according to a second preset wavelet function to obtain characteristic information of the second image to be identified;
and performing inverse wavelet transform processing on the characteristic information according to a second preset wavelet function to obtain an image identification result and obtain a target image.
In some implementations of the first aspect, acquiring the first to-be-recognized image includes:
acquiring a third image to be identified;
and carrying out preset data augmentation transformation on the third image to be recognized to obtain the first image to be recognized.
In some realizations of the first aspect, after the performing the preset data augmentation transformation on the third image to be recognized, the method further includes:
and carrying out color enhancement on the third image to be recognized after the amplification transformation to obtain a first image to be recognized.
In some realizations of the first aspect, before obtaining the image recognition result of the second image to be recognized according to the preset image recognition model, the method further includes:
acquiring a preset image recognition training sample, wherein the preset image recognition training sample comprises a preset image obtained after amplification and transformation according to preset data, and the preset image comprises a preset image to be recognized;
and training the preset image recognition network according to the preset image recognition training sample to obtain a preset image recognition model.
In some realizations of the first aspect, the first preset wavelet function is a symmetric wavelet basis function.
In a second aspect, an embodiment of the present invention provides an image recognition apparatus, including:
the acquisition module is used for acquiring a first image to be recognized, and the first image to be recognized comprises a target image to be recognized;
the de-noising module is used for de-noising the first image to be recognized according to a first preset wavelet function to obtain a second image to be recognized comprising a target image to be recognized;
and the recognition module is used for acquiring an image recognition result of the second image to be recognized according to the preset image recognition model to obtain the target image.
In a third aspect, the present invention provides an image recognition apparatus comprising: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, may implement the image recognition method of the first aspect or any of the realizable forms of the first aspect.
In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer program instructions are stored, which, when executed by a processor, implement the image recognition method of the first aspect or any of the realizable manners of the first aspect.
The embodiment of the invention provides an image identification method, which comprises the steps of carrying out noise reduction treatment on an image to be identified according to a first preset wavelet function, wherein the first image to be identified comprises a target image to be identified, so that main characteristic information in the image to be identified can be effectively reserved, and particularly, detail texture characteristics in the image can be extracted and identified. Moreover, the image area of the image to be recognized after the noise reduction processing is effectively reduced compared with the image area before the noise reduction processing, and the main characteristic information in the image to be recognized is combined, so that the recognition speed and the image recognition precision can be effectively improved when the image to be recognized after the noise reduction processing is performed according to the preset image recognition model, and an accurate image recognition result is obtained.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of an image recognition method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of another image recognition method provided by the embodiment of the invention;
fig. 3 is a schematic structural diagram of an image recognition apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an image recognition apparatus according to an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
With the rapid development of image recognition and image processing technologies, the image is acquired in real time, and advanced recognition and semantic understanding of target identification in the image are more and more widely applied. For example, in an automatic driving scene, whether the camera sensor can perform effective advanced recognition and semantic understanding on traffic signs on both sides of a road or not is important, and automatic driving in a high-speed scene is important.
Before the image is identified and semantically understood, the quality of the noise reduction preprocessing of the image has great influence on the subsequent image identification and semantically understood effects. On one hand, the existing noise reduction processing method obtains a new noise reduction method that a central pixel value represents an adjacent pixel from a traditional spatial domain pixel characteristic noise reduction algorithm, such as arithmetic mean filtering, median filtering, Gaussian filtering, bilateral filtering and derivative deformation thereof, by analyzing direct relation between the central pixel and other adjacent pixels in a gray scale space in a window with a certain size. However, the traditional free pixel feature noise reduction algorithm performs certain fusion and selection on direct association between a central pixel and other adjacent pixels in a window with a certain size based on an image, and has no obvious effect of improving the noise reduction effect.
On the other hand, the image is first subjected to space-frequency domain transformation, the image is converted from a space domain to a transformation domain, then the noise is divided into high, medium and low frequency noise from the frequency, and a space-domain pixel characteristic noise reduction algorithm and a transformation domain noise reduction algorithm are combined, so that a high peak signal-to-noise ratio can be obtained, for example, a three-dimensional Block Matching algorithm (Block Matching 3D, BM3D), but the time complexity of the method is high.
Therefore, when performing noise reduction processing on an image, more and more attention is paid to how to maintain the integrity of useful information in the original image as much as possible and effectively remove useless information in the original image.
In view of the above, an embodiment of the present invention provides an image recognition method, where after a first to-be-recognized image including a target to-be-recognized image is obtained, a denoising process is performed on the first to-be-recognized image according to a first preset wavelet function, and since the first to-be-recognized image can be converted into a frequency domain by the first preset wavelet function, after the denoising process is performed from the angle of the frequency domain, main feature information in the to-be-recognized image can be effectively retained, so as to achieve a denoising effect, and then, when the denoised to-be-recognized image is performed according to a preset image recognition model, the recognition speed and the image recognition accuracy can be effectively improved, and an accurate image recognition result is obtained.
The following describes an image recognition method according to an embodiment of the present invention with reference to the drawings.
Fig. 1 is a flowchart illustrating an image recognition method according to an embodiment of the present invention. As shown in fig. 1, the method may include S110-S130.
And S110, acquiring a first image to be recognized.
In embodiment S110 of the present invention, the first image to be recognized includes a target image to be recognized. For example, the target image to be recognized may be a traffic sign on a road during the driving of the vehicle on the road.
In some embodiments, during the driving process of the vehicle, the position, the direction, and the size of the target image to be recognized may affect the recognition accuracy, and therefore, during the training process of the preset image recognition model, the preset data augmentation transformation is performed on the training set, for example, data augmentation processing such as translation, rotation, scaling, and occlusion addition is performed on the training sample in the training set, so as to improve the generalization capability of the preset image recognition model, and enable the preset image recognition model to cope with the complex image to be recognized. For example, when the shooting distance is long, the traffic sign (target image to be recognized) in the original image is small,
therefore, acquiring the first image to be recognized may specifically include the following steps: and acquiring a third image to be recognized, and performing preset data amplification transformation on the third image to be recognized to obtain a first image to be recognized. Therefore, the identification precision of the first image to be identified is improved. The third image to be recognized may be, for example, an original image obtained by a camera or a sensor, that is, the third image to be recognized. For example, during the driving process of a vehicle on a road, an image to be recognized including a traffic sign on the road can be acquired through the vehicle-mounted camera.
In some embodiments, in order to facilitate extracting the color of the target image to be recognized, color enhancement may be performed on the third image to be recognized after the augmentation transformation, so as to obtain the first image to be recognized. For example, the color enhancement may be to convert the third image to be recognized after the augmentation transformation into an HSV format, or may also be to convert the third image to be recognized after the augmentation transformation into a YUV format, so that the color in the traffic sign is easier to extract.
After the first image to be recognized is obtained, in order to improve the robustness and classification accuracy of the recognition of the target image to be recognized, S120 may be performed next.
And S120, performing noise reduction processing on the first image to be recognized according to the first preset wavelet function to obtain a second image to be recognized comprising the target image to be recognized.
The first image to be recognized is subjected to noise reduction processing through the first preset wavelet transform, the first image to be recognized can be converted into a frequency domain, and after the noise reduction processing is carried out from the angle of the frequency domain, main characteristic information in the image to be recognized can be effectively reserved, so that the noise reduction effect is achieved.
In some embodiments, the noise reduction processing on the first image to be recognized may include the following steps: firstly, acquiring an approximate component and a detail component of a first image to be identified according to a first preset wavelet function, wherein the detail component comprises a high-frequency component; acquiring an approximation coefficient of the approximation component and a detail coefficient corresponding to the high-frequency component; secondly, Gaussian filtering is carried out on the high-frequency component to obtain a filtered high-frequency component; and finally, performing wavelet inverse transformation calculation according to the approximate component, the approximate coefficient, the filtered high-frequency component and the detail coefficient to obtain a second image to be identified.
In order to avoid phase distortion of the first image to be identified in the denoising process and achieve a better denoising effect, the first preset wavelet function may be a symmetrical wavelet basis function.
In some embodiments, the first preset wavelet function may be a Dmeyer wavelet, based on which the first to-be-identified image is decomposed into an approximation component and a detail component. The approximate component is mainly a low-frequency signal, and the detail component may include a plurality of high-frequency components, wherein the noise component is mainly concentrated in the high-frequency components. And performing inverse transformation on the reconstructed data by using the approximate component and the detail component after the noise reduction treatment, so as to reconstruct the first image to be recognized and obtain a second image to be recognized.
In the embodiment of the invention, the data volume of the second image to be recognized relative to the first image to be recognized is greatly reduced, namely the image area is effectively reduced, and the second image to be recognized is obtained to include more image characteristics of the target image because only the high-frequency components concentrated by the noise components are subjected to noise reduction processing.
After the second image to be recognized is obtained, S130 may be performed next.
130. And acquiring an image recognition result of the second image to be recognized according to the preset image recognition model to obtain the target image.
In S130 of the embodiment of the present invention, in order to resist noise propagation in the deep network and effectively retain the basic target structure in the feature map of the image to be recognized, the preset image recognition model performs downsampling on the first image to be recognized by using the second preset wavelet function, so as to implement feature extraction on the target image to be recognized in the first image to be recognized. Specifically, firstly, performing wavelet transformation processing on a second image to be identified according to a second preset wavelet function to obtain characteristic information of the second image to be identified; and then, performing inverse wavelet transform processing on the characteristic information according to a second preset wavelet function to obtain an image identification result and obtain a target image.
In some embodiments, the first preset wavelet function and the second preset wavelet function may be the same or different. Illustratively, according to a preset image recognition model, performing a lower adoption process on the first band to-be-recognized image by using a Dmeyer wavelet transform to obtain feature information of a second to-be-recognized image, wherein the step size is 1. Accordingly, after the wavelet transform is used, the acquisition of the reconstructed image requires the inverse wavelet transform, i.e., the up-sampling process, to be performed on the feature information of the second image to be recognized in combination with the Dmeyer wavelet. By using the second preset wavelet transform, it is possible to resist the propagation of noise in the depth network and to help maintain the basic structure of the target image.
According to the image identification method provided by the embodiment of the invention, the image to be identified is subjected to noise reduction treatment according to the first preset wavelet function, wherein the first image to be identified comprises the target image to be identified, so that the main characteristic information in the image to be identified can be effectively reserved, and particularly, the detail texture characteristics in the image can be extracted and identified. Moreover, the image area of the image to be recognized after the noise reduction processing is effectively reduced compared with the image area before the noise reduction processing, and the main characteristic information in the image to be recognized is combined, so that the recognition speed and the image recognition precision can be effectively improved when the image to be recognized after the noise reduction processing is performed according to the preset image recognition model, and an accurate image recognition result is obtained.
In some embodiments, before obtaining the image recognition result of the second image to be recognized according to the preset image recognition model, a trained preset image recognition model needs to be obtained first. The acquiring of the preset image recognition model may include the steps of: firstly, acquiring a preset image recognition training sample, wherein the preset image recognition training sample comprises a preset image obtained after amplification and transformation according to preset data, and the preset image comprises a preset image to be recognized; and then, training the preset image recognition network according to the preset image recognition training sample to obtain a preset image recognition model. In the embodiment of the invention, because the wavelet transform is used for replacing the convolution operation, the generation of model parameters can be greatly reduced, and the training speed of the preset image recognition model is improved.
In order to better describe the image recognition processing flow provided by the embodiment of the present invention, in combination with another schematic flow chart of the image recognition method provided by the embodiment of the present invention shown in fig. 2, the image recognition processing method may include the following steps:
s201, acquiring an image to be identified.
The image to be recognized comprises a target image to be recognized, the image to be recognized also comprises noise generated in the shooting or propagation process of the image to be recognized, or the image to be recognized is influenced by a natural environment, so that the target recognition image is shielded by natural factors which cannot be avoided, such as leaves, fog and the like, in the image to be recognized.
And S202, presetting data amplification processing.
And carrying out preset data amplification processing on the image to be recognized so as to reduce the influence of the distance between the target image to be recognized and the camera on the image recognition precision.
And S203, color enhancement processing.
In this step, in order to more easily extract colors in images to be recognized of different targets, color enhancement processing may be performed on the images to be recognized after the preset data enhancement processing, for example, the images to be recognized after the preset data enhancement processing are converted into HSV format or YUV format.
And S204, performing wavelet transformation.
And performing wavelet transformation processing on the image to be identified after the color enhancement processing according to a preset first wavelet function to obtain an approximate component and a detail component of the image to be identified and a corresponding approximate coefficient and a corresponding detail coefficient.
And S205, denoising.
And filtering and denoising the detail components according to a Gaussian filter function to obtain the detail components subjected to denoising treatment.
And S206, reconstructing the image.
And according to the approximate component, the detail component after the noise reduction treatment, the corresponding approximate coefficient and the detail coefficient, performing inverse transformation according to a first preset wavelet function to obtain a reconstructed image to be identified.
And S207, image recognition.
And carrying out feature recognition on the reconstructed image to be recognized according to a preset image recognition model to obtain a target image.
In the embodiment of the invention, because the image features of the target image to be recognized are mainly in the approximate components (low-frequency components), the high-frequency components of the target image to be recognized are subjected to noise reduction processing according to wavelet transformation, so that the image features of the target image to be recognized can be completely reserved. And then, the feature recognition is carried out on the reconstructed image to be recognized by adopting a preset image recognition model, so that the detail texture features of the target image can be better extracted and recognized, and the recognition precision of the target image is improved.
Fig. 3 is a schematic structural diagram of an image recognition apparatus according to an embodiment of the present invention, and as shown in fig. 3, the image recognition apparatus 300 may include: an acquisition module 310, a noise reduction module 320, and an identification module 330.
The acquiring module 310 is configured to acquire a first image to be recognized, where the first image to be recognized includes a target image to be recognized;
the denoising module 320 is configured to perform denoising processing on the first to-be-identified image according to a first preset wavelet function to obtain a second to-be-identified image including a target to-be-identified image;
the identifying module 330 is configured to obtain an image identification result of the second image to be identified according to a preset image identification model, so as to obtain a target image.
In some embodiments, the denoising module 320 is specifically configured to obtain an approximate component and a detail component of the first image to be recognized according to a first preset wavelet function, where the detail component includes a high-frequency component; acquiring an approximation coefficient of the approximation component and a detail coefficient corresponding to the high-frequency component; performing Gaussian filtering on the high-frequency components to obtain filtered high-frequency components; and performing wavelet inverse transformation calculation according to the approximate component, the approximate coefficient, the filtered high-frequency component and the detail coefficient to obtain a second image to be identified.
In some embodiments, the identifying module 330 is further configured to perform wavelet transform processing on the second image to be identified according to a second preset wavelet function, so as to obtain feature information of the second image to be identified; and performing inverse wavelet transform processing on the characteristic information according to a second preset wavelet function to obtain an image identification result and obtain a target image.
In some embodiments, the obtaining module 310 is further configured to obtain a third image to be recognized; and carrying out preset data augmentation transformation on the third image to be recognized to obtain the first image to be recognized.
In some embodiments, the obtaining module 310 is further configured to perform color enhancement on the third image to be recognized after the augmentation transformation, so as to obtain the first image to be recognized.
In some embodiments, the image recognition apparatus 300 may further include a training module, configured to obtain a preset image recognition training sample, where the preset image recognition training sample includes a preset image obtained after being subjected to augmentation transformation according to preset data, and the preset image includes a preset image to be recognized; and training the preset image recognition network according to the preset image recognition training sample to obtain a preset image recognition model.
In some embodiments, the first preset wavelet function is a symmetric wavelet basis function.
It is understood that the image recognition apparatus 300 according to the embodiment of the present invention may correspond to an execution subject of the image recognition method according to the embodiment of the present invention, and specific details of operations and/or functions of each module/unit of the image recognition apparatus 300 may refer to the descriptions of corresponding parts in the image recognition method according to the embodiment of the present invention, which are not described herein again for brevity.
According to the image recognition device provided by the embodiment of the invention, after the first to-be-recognized image including the target to-be-recognized image is obtained, the first to-be-recognized image is subjected to noise reduction according to the first preset wavelet function, the first to-be-recognized image can be converted into the frequency domain through the first preset wavelet function, and the noise reduction is performed from the angle of the frequency domain, so that the noise reduction effect is achieved, the noise reduction effect is obviously improved, and then the second to-be-recognized image is obtained through wavelet inverse transformation reconstruction, so that the main characteristic information in the to-be-recognized image can be effectively reserved, and particularly, the detail characteristic in the image can be extracted and recognized. Moreover, the image area of the image to be recognized after the noise reduction processing is effectively reduced compared with the image area before the noise reduction processing, and the main characteristic information in the image to be recognized is combined, so that the recognition speed and the image recognition precision can be effectively improved when the image to be recognized after the noise reduction processing is performed according to the preset image recognition model, and an accurate image recognition result is obtained.
Fig. 4 is a schematic diagram of a hardware structure of an image recognition device according to an embodiment of the present invention.
As shown in fig. 4, the image recognition apparatus 400 in the present embodiment includes an input apparatus 401, an input interface 402, a central processor 403, a memory 404, an output interface 405, and an output apparatus 406. The input interface 402, the central processing unit 403, the memory 404, and the output interface 405 are connected to each other through a bus 410, and the input device 401 and the output device 406 are connected to the bus 410 through the input interface 402 and the output interface 405, respectively, and further connected to other components of the image recognition device 400.
Specifically, the input device 401 receives input information from the outside and transmits the input information to the central processor 403 through the input interface 402; the central processor 403 processes the input information based on computer-executable instructions stored in the memory 404 to generate output information, stores the output information temporarily or permanently in the memory 404, and then transmits the output information to the output device 406 through the output interface 405; the output device 406 outputs the output information to the outside of the image recognition device 400 for use by the user.
That is, the image recognition apparatus shown in fig. 4 may also be implemented to include: a memory storing computer-executable instructions; and a processor which, when executing computer executable instructions, may implement the image recognition method described in connection with the example shown in fig. 1.
In one embodiment, the image recognition apparatus 400 shown in fig. 4 includes: a memory 404 for storing programs; and a processor 403, configured to execute a program stored in the memory to perform the image recognition method according to the embodiment of the present invention.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium has computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement the image recognition method provided by embodiments of the present invention.
It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic Circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor Memory devices, Read-Only memories (ROMs), flash memories, Erasable Read-Only memories (EROMs), floppy disks, Compact disk Read-Only memories (CD-ROMs), optical disks, hard disks, optical fiber media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.
As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.
Claims (10)
1. An image recognition method, characterized in that the method comprises:
acquiring a first image to be recognized, wherein the first image to be recognized comprises a target image to be recognized;
according to a first preset wavelet function, denoising the first image to be recognized to obtain a second image to be recognized comprising the target image to be recognized;
and acquiring an image recognition result of the second image to be recognized according to a preset image recognition model to obtain a target image.
2. The method according to claim 1, wherein the denoising the first image to be recognized according to a first preset wavelet function to obtain a second image to be recognized including the target image to be recognized comprises:
obtaining an approximate component and a detail component of the first image to be identified according to the first preset wavelet function, wherein the detail component comprises a high-frequency component; acquiring an approximation coefficient of the approximation component and a detail coefficient corresponding to the high-frequency component;
performing Gaussian filtering on the high-frequency components to obtain filtered high-frequency components;
and performing wavelet inverse transformation calculation according to the approximate component, the approximate coefficient, the filtered high-frequency component and the detail coefficient to obtain the second image to be identified.
3. The method according to claim 1, wherein the obtaining of the image recognition result of the second image to be recognized according to a preset image recognition model comprises:
performing wavelet transformation processing on the second image to be identified according to a second preset wavelet function to obtain characteristic information of the second image to be identified;
and performing inverse wavelet transform processing on the characteristic information according to the second preset wavelet function to obtain the image identification result, so as to obtain the target image.
4. The method of claim 1, wherein the acquiring the first image to be recognized comprises:
acquiring a third image to be identified;
and carrying out preset data augmentation transformation on the third image to be recognized to obtain the first image to be recognized.
5. The method according to claim 4, wherein after the performing of the preset data augmentation transformation on the third image to be recognized, the method further comprises:
and carrying out color enhancement on the third image to be recognized after the amplification transformation to obtain the first image to be recognized.
6. The method according to claim 4, wherein before the obtaining of the image recognition result of the second image to be recognized according to the preset image recognition model, the method further comprises:
acquiring a preset image recognition training sample, wherein the preset image recognition training sample comprises a preset image obtained after amplification and transformation according to preset data, and the preset image comprises a preset image to be recognized;
and training a preset image recognition network according to the preset image recognition training sample to obtain the preset image recognition model.
7. The method of claim 1, wherein the first predetermined wavelet function is a symmetric wavelet basis function.
8. An image recognition apparatus, characterized in that the apparatus comprises:
the system comprises an acquisition module, a recognition module and a recognition module, wherein the acquisition module is used for acquiring a first image to be recognized, and the first image to be recognized comprises a target image to be recognized;
the denoising module is used for denoising the first image to be recognized according to a first preset wavelet function to obtain a second image to be recognized comprising the target image to be recognized;
and the identification module is used for acquiring an image identification result of the second image to be identified according to a preset image identification model to obtain the target image.
9. An image recognition apparatus, characterized in that the apparatus comprises: a processor, and a memory storing computer program instructions;
the processor reads and executes the computer program instructions to implement the image recognition method of any one of claims 1-7.
10. A computer storage medium having computer program instructions stored thereon which, when executed by a processor, implement the image recognition method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011554905.5A CN112711999A (en) | 2020-12-24 | 2020-12-24 | Image recognition method, device, equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011554905.5A CN112711999A (en) | 2020-12-24 | 2020-12-24 | Image recognition method, device, equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112711999A true CN112711999A (en) | 2021-04-27 |
Family
ID=75545498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011554905.5A Pending CN112711999A (en) | 2020-12-24 | 2020-12-24 | Image recognition method, device, equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112711999A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113378972A (en) * | 2021-06-28 | 2021-09-10 | 成都恒创新星科技有限公司 | License plate recognition method and system in complex scene |
| CN114169430A (en) * | 2021-12-06 | 2022-03-11 | 深圳依时货拉拉科技有限公司 | Method and device for identifying copied image, storage medium and computer equipment |
| CN114398926A (en) * | 2022-01-12 | 2022-04-26 | 江苏金晟元控制技术有限公司 | Resistance spot welding plastic ring imaging method based on wavelet analysis and application thereof |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106056555A (en) * | 2016-06-03 | 2016-10-26 | 宁波大红鹰学院 | Image denoising method |
| CN109801282A (en) * | 2019-01-24 | 2019-05-24 | 湖北大学 | Pavement behavior detection method, processing method, apparatus and system |
| CN110009626A (en) * | 2019-04-11 | 2019-07-12 | 北京百度网讯科技有限公司 | Method and apparatus for generating image |
| CN110400267A (en) * | 2019-07-03 | 2019-11-01 | 安徽继远软件有限公司 | A kind of preprocess method based on inspection image |
| CN111046755A (en) * | 2019-11-27 | 2020-04-21 | 上海眼控科技股份有限公司 | Character recognition method, character recognition device, computer equipment and computer-readable storage medium |
-
2020
- 2020-12-24 CN CN202011554905.5A patent/CN112711999A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106056555A (en) * | 2016-06-03 | 2016-10-26 | 宁波大红鹰学院 | Image denoising method |
| CN109801282A (en) * | 2019-01-24 | 2019-05-24 | 湖北大学 | Pavement behavior detection method, processing method, apparatus and system |
| CN110009626A (en) * | 2019-04-11 | 2019-07-12 | 北京百度网讯科技有限公司 | Method and apparatus for generating image |
| CN110400267A (en) * | 2019-07-03 | 2019-11-01 | 安徽继远软件有限公司 | A kind of preprocess method based on inspection image |
| CN111046755A (en) * | 2019-11-27 | 2020-04-21 | 上海眼控科技股份有限公司 | Character recognition method, character recognition device, computer equipment and computer-readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| QIUFU LI 等: "Wavelet Integrated CNNs for Noise-Robust Image Classification", 《2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113378972A (en) * | 2021-06-28 | 2021-09-10 | 成都恒创新星科技有限公司 | License plate recognition method and system in complex scene |
| CN113378972B (en) * | 2021-06-28 | 2024-03-22 | 成都恒创新星科技有限公司 | License plate recognition method and system under complex scene |
| CN114169430A (en) * | 2021-12-06 | 2022-03-11 | 深圳依时货拉拉科技有限公司 | Method and device for identifying copied image, storage medium and computer equipment |
| CN114398926A (en) * | 2022-01-12 | 2022-04-26 | 江苏金晟元控制技术有限公司 | Resistance spot welding plastic ring imaging method based on wavelet analysis and application thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112711999A (en) | Image recognition method, device, equipment and computer storage medium | |
| CN103020916B (en) | Image denoising method combining two-dimensional Hilbert transform and BEMD | |
| CN102324021A (en) | Infrared dim-small target detection method based on shear wave conversion | |
| Bajpai et al. | Analysis of image enhancement techniques used in remote sensing satellite imagery | |
| CN112327259A (en) | Method and device for eliminating interference signals in SAR image | |
| CN117726537A (en) | SAR image denoising network method and system for self-adaptive multi-scale feature fusion AMFFD-Net | |
| CN110782406A (en) | Image denoising method and device based on information distillation network | |
| CN106157240B (en) | Remote sensing image super-resolution method based on dictionary learning | |
| CN114155161B (en) | Image denoising method, device, electronic equipment and storage medium | |
| Sheybani | An Algorithm for Real-Time Blind Image Quality Comparison and Assessment | |
| Gupta et al. | A noise robust edge detector for color images using hilbert transform | |
| Karthikeyan et al. | Energy based denoising convolutional neural network for image enhancement | |
| CN104036472A (en) | Method and device for enhancing quality of 3D image | |
| CN117911267B (en) | Back-scattered X-ray image noise reduction method based on wavelet transformation | |
| CN102663748A (en) | Low depth of field image segmentation method based on frequency domain | |
| Bhonsle et al. | De-noising of CT images using combined bivariate shrinkage and enhanced total variation technique | |
| CN103778615A (en) | Multi-focus image fusion method based on region similarity | |
| CN116385312A (en) | Low-illumination image denoising method based on phase correlation | |
| Huang et al. | An improved retinex algorithm for underwater image enhancement based on HSV model | |
| Pang | Improved image denoising based on Haar wavelet transform | |
| CN106845448A (en) | A kind of method for detecting infrared puniness target based on nonnegativity restrictions 2D variation mode decompositions | |
| CN112001870A (en) | Visible light and infrared image fusion method based on convolutional dictionary learning | |
| Zhang et al. | Research on Laser Polarization Image Reconstruction Based on Wavelet Transform and Deep Learning | |
| TW201617972A (en) | A method for extracting non-periodical patterns masked by periodical patterns, and a device implementing said method | |
| CN114581443B (en) | Image processing method and device, computer equipment and readable storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210427 |