CN115131795A - Image scanning anti-shake recognition method and system and storage medium - Google Patents
Image scanning anti-shake recognition method and system and storage medium Download PDFInfo
- Publication number
- CN115131795A CN115131795A CN202210739709.8A CN202210739709A CN115131795A CN 115131795 A CN115131795 A CN 115131795A CN 202210739709 A CN202210739709 A CN 202210739709A CN 115131795 A CN115131795 A CN 115131795A
- Authority
- CN
- China
- Prior art keywords
- image
- scanning
- shake
- information
- neural network
- 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 32
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 14
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 10
- 230000033001 locomotion Effects 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 239000011664 nicotinic acid Substances 0.000 claims description 3
- 238000003909 pattern recognition Methods 0.000 claims 1
- 238000013135 deep learning Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 102100032202 Cornulin Human genes 0.000 description 1
- 101000920981 Homo sapiens Cornulin Proteins 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010894 electron beam technology Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001629 suppression Effects 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
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/20—Combination of acquisition, preprocessing or recognition functions
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19173—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an image scanning anti-shake recognition method, an image scanning anti-shake recognition system and a storage medium, and relates to the field of scanning recognition. The invention comprises the following steps: acquiring a scanned image, and obtaining RGB information of the scanned image according to the scanned image; preprocessing RGB information of a scanned image, the preprocessing comprising: turning over, enlarging and reducing; processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be recognized; and identifying the information to be identified by combining the cyclic neural network with the OCR identification part of the deep neural network. The invention improves the accuracy of scanning identification and realizes the identification of text characters.
Description
Technical Field
The invention relates to the field of scanning identification, in particular to an image scanning anti-shake identification method, an image scanning anti-shake identification system and a storage medium.
Background
With the development of internet technology, image scanning devices (such as electronic scanning pens) have become an indispensable part of learning and living. The image scanning means that pictures or figures are displayed on a screen through the left and right movement of electron beams, radio waves and the like, and the pictures or the figures are copied, memorized and stored with certain characters and patterns.
In the prior art, the image scanning apparatus is rarely involved in the function of character recognition, which causes great inconvenience in entering characters. Furthermore, nowadays, more and more handheld scanning devices are provided, which can not avoid the problems of jitter and the like in the scanning process. Therefore, how to solve the above problems needs to be solved by those skilled in the art.
Disclosure of Invention
In view of this, the present invention provides an image scanning anti-shake recognition method, system and storage medium, so as to improve the accuracy of scanning recognition, realize recognition of text characters, and solve the problems in the background art.
In order to achieve the purpose, the invention adopts the following technical scheme:
disclosed is an image scanning anti-shake recognition method, which comprises the following steps:
acquiring a scanned image, and acquiring RGB information of the scanned image according to the scanned image;
preprocessing RGB information of a scanned image, the preprocessing comprising: turning over, enlarging and reducing;
processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be recognized;
and identifying the information to be identified by combining the cyclic neural network with the OCR identification part of the deep neural network.
Optionally, performing semantic segmentation on the information to be recognized, and performing preprocessing by using an OCR recognition part of the neural network;
inputting the preprocessed information to be identified into a cyclic convolution neural network, firstly extracting a feature map of the image to be identified through a convolution layer, and cutting the feature map into 4 blocks according to rows;
and each block is processed through a cycle layer and a softmax function, the prediction probability of the character corresponding to the element is listed, the prediction result is approximately in soft alignment with the class mark, and finally, a complete recognition result is obtained.
Optionally, the preprocessing of the OCA recognition part using the neural network is: and carrying out cell positioning on the character part, and carrying out image direction correction, format conversion, gray level conversion and binaryzation on a positioning result.
Optionally, the anti-shake algorithm includes the following steps:
inputting information to be identified;
feature point detection using SURF;
tracking the positions of the feature points by using an optical flow method based on feature matching, calculating the motion vector of the feature points, and estimating the motion path of the scanning equipment;
and smoothing the motion path of the scanning equipment by using a Kalman filtering algorithm so as to output a relatively stable scanning frame sequence.
Optionally, the method further includes the steps of utilizing the bionic mode to identify and cover each class of learning samples according to the characteristic of continuous distribution of the similar samples in the feature space, and realizing supervision and classification by taking the optimal coverage of the samples as a criterion.
Also disclosed is an image scanning anti-shake recognition system, comprising:
an image acquisition module: the system comprises a scanning module, a processing module and a display module, wherein the scanning module is used for acquiring a scanning image and obtaining RGB information of the scanning image according to the scanning image;
an image preprocessing module: for pre-processing RGB information of a scanned image, the pre-processing comprising: turning over, enlarging and reducing;
anti-shake processing module: the system is used for processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be identified;
an image text recognition module: the system is used for identifying the information to be identified by utilizing the combination of the cyclic neural network and the OCR identification part of the deep neural network.
Finally a computer storage medium is disclosed, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of an image scanning anti-shake recognition method according to any one of the claims.
Compared with the prior art, the image scanning anti-shake recognition method, the image scanning anti-shake recognition system and the storage medium have the following advantages that:
1. according to the method, the calculated amount of picture scanning is effectively reduced through the anti-shake algorithm, so that the calculation performance requirement on the embedded equipment is effectively reduced, and high real-time performance and accuracy can be realized on the embedded equipment with low configuration;
2. the identification method of the invention has the advantages of obviously improved accuracy, and improved generalization and robustness.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow diagram of the present invention;
fig. 2 is a schematic structural diagram of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses an image scanning anti-shake identification method, which comprises the following steps as shown in figure 1:
s1: acquiring a scanned image, and acquiring RGB information of the scanned image according to the scanned image;
s2: preprocessing RGB information of a scanned image, wherein the preprocessing comprises the following steps: turning over, enlarging and reducing;
s3: processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be recognized;
s4: and identifying the information to be identified by combining the cyclic neural network with the OCR identification part of the deep neural network.
In this embodiment, the method further includes: performing semantic segmentation on information to be recognized, and performing preprocessing by utilizing an OCR recognition part of a neural network;
inputting the preprocessed information to be identified into a cyclic convolution neural network, firstly extracting a feature map of the image to be identified through a convolution layer, and cutting the feature map into 4 blocks according to rows;
and processing each block through a loop layer and a softmax function, listing the prediction probability of the character corresponding to the element, and carrying out approximate soft alignment on the prediction result and the class mark to finally obtain a complete recognition result.
The method also comprises the steps of utilizing the bionic mode to identify and cover each class of learning samples by utilizing a plurality of same geometric forms according to the characteristic of continuous distribution of the same class of samples in a characteristic space, and realizing supervision and classification by taking the optimal coverage of the samples as a criterion.
Specifically, the preprocessing of the OCA identification part by using the neural network is as follows: and carrying out cell positioning on the character part, and carrying out image direction correction, format conversion, gray level conversion and binaryzation on a positioning result. In the deep learning-based OCR task, the recognition process is divided into two steps: single word cutting and sorting tasks. Generally, people will cut a series of characters into a single font by using a projection method, and then send the font into a CNN network for character classification. In the embodiment, an end-to-end character recognition method CRNN based on deep learning is adopted, and the character recognition problem is converted into a sequence learning problem without explicitly utilizing a traditional algorithm to perform character segmentation. Although the input images have different scales and different text lengths, the whole text image can be recognized by adopting a translation layer method in an output stage through the loop processing of the CNN and the RNN. That is, the cutting of the text is also incorporated into the deep learning. Therefore, errors in the character cutting link can be avoided, and the recognition result is further influenced.
The anti-shake algorithm comprises the following steps:
inputting information to be identified;
feature point detection using SURF; the SURF algorithm detects characteristic points in a matrix form, the Hessian matrix determinant value represents the variation of surrounding pixel points, and then non-maximum suppression is carried out.
Tracking the positions of the feature points by using an optical flow method based on feature matching, calculating the motion vector of the feature points, and estimating the motion path of the scanning equipment; and tracking the motion trail of the feature points acquired by the SURF algorithm by using an optical flow algorithm, and continuously positioning and tracking the main features of the target. Optical flow assumes that the luminance values do not change when the same object moves between different frames.
And smoothing the motion path of the scanning equipment by using a Kalman filtering algorithm so as to output a relatively stable scanning frame sequence.
Also disclosed is an image scanning anti-shake recognition system, as shown in fig. 2, including:
an image acquisition module: the system comprises a scanning module, a processing module and a display module, wherein the scanning module is used for acquiring a scanning image and obtaining RGB information of the scanning image according to the scanning image;
an image preprocessing module: for preprocessing the RGB information of a scanned image, the preprocessing comprising: turning over, zooming in and zooming out;
anti-shake processing module: the system is used for processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be identified;
an image text recognition module: the method is used for identifying the information to be identified by combining the cyclic neural network with the OCR identification part of the deep neural network.
Finally a computer storage medium is disclosed, on which a computer program is stored which, when being executed by a processor, carries out the steps of an image scanning anti-shake recognition method according to any one of the claims.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. An image scanning anti-shake identification method is characterized by comprising the following steps:
acquiring a scanned image, and acquiring RGB information of the scanned image according to the scanned image;
preprocessing RGB information of a scanned image, the preprocessing comprising: turning over, zooming in and zooming out;
processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be recognized;
and identifying the information to be identified by combining the cyclic neural network with the OCR identification part of the deep neural network.
2. The image scanning anti-shake recognition method according to claim 1, further comprising performing semantic segmentation on the information to be recognized and performing preprocessing by using an OCR recognition part of a neural network;
inputting the preprocessed information to be identified into a cyclic convolution neural network, firstly extracting a feature map of the image to be identified through a convolution layer, and cutting the feature map into 4 blocks according to rows;
and each block is processed through a cycle layer and a softmax function, the prediction probability of the character corresponding to the element is listed, the prediction result is approximately in soft alignment with the class mark, and finally, a complete recognition result is obtained.
3. The image scanning anti-shake recognition method according to claim 1, wherein the OCA recognition part using the neural network is preprocessed as: and carrying out cell positioning on the character part, and carrying out image direction correction, format conversion, gray level conversion and binarization on a positioning result.
4. The image scanning anti-shake recognition method according to claim 1, wherein the anti-shake algorithm comprises the following steps:
inputting information to be identified;
feature point detection using SURF;
tracking the positions of the feature points by using an optical flow method based on feature matching, calculating a motion vector of the feature points, and estimating a motion path of scanning equipment;
and smoothing the motion path of the scanning equipment by using a Kalman filtering algorithm so as to output a relatively stable scanning frame sequence.
5. The image scanning anti-shake recognition method according to claim 1, further comprising using a bionic pattern recognition to cover each class of learning samples with a plurality of same geometric shapes according to the characteristic of continuous distribution of similar samples in a feature space, and realizing supervised classification by taking the optimal coverage of the samples as a criterion.
6. An image scanning anti-shake recognition system, comprising:
an image acquisition module: the system comprises a scanning module, a display module and a display module, wherein the scanning module is used for acquiring a scanning image and obtaining RGB information of the scanning image according to the scanning image;
an image preprocessing module: for pre-processing RGB information of a scanned image, the pre-processing comprising: turning over, zooming in and zooming out;
anti-shake processing module: the system is used for processing the preprocessed RGB information by using an anti-shake algorithm to obtain information to be identified;
an image text recognition module: the system is used for identifying the information to be identified by utilizing the combination of the cyclic neural network and the OCR identification part of the deep neural network.
7. A computer storage medium, characterized in that the computer storage medium has stored thereon a computer program which, when being executed by a processor, implements the steps of an image scanning anti-shake recognition method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210739709.8A CN115131795A (en) | 2022-06-28 | 2022-06-28 | Image scanning anti-shake recognition method and system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210739709.8A CN115131795A (en) | 2022-06-28 | 2022-06-28 | Image scanning anti-shake recognition method and system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115131795A true CN115131795A (en) | 2022-09-30 |
Family
ID=83380529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210739709.8A Pending CN115131795A (en) | 2022-06-28 | 2022-06-28 | Image scanning anti-shake recognition method and system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115131795A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845472A (en) * | 2016-12-30 | 2017-06-13 | 深圳仝安技术有限公司 | A kind of novel intelligent wrist-watch scans explanation/interpretation method and novel intelligent wrist-watch |
| CN110796010A (en) * | 2019-09-29 | 2020-02-14 | 湖北工业大学 | Video image stabilization method combining optical flow method and Kalman filtering |
| CN111898603A (en) * | 2020-08-10 | 2020-11-06 | 上海瑞美锦鑫健康管理有限公司 | Physical examination order recognition method and system based on deep neural network |
| CN112132151A (en) * | 2020-09-19 | 2020-12-25 | 娄忠富 | Image character recognition system and method based on recurrent neural network recognition algorithm |
-
2022
- 2022-06-28 CN CN202210739709.8A patent/CN115131795A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845472A (en) * | 2016-12-30 | 2017-06-13 | 深圳仝安技术有限公司 | A kind of novel intelligent wrist-watch scans explanation/interpretation method and novel intelligent wrist-watch |
| CN110796010A (en) * | 2019-09-29 | 2020-02-14 | 湖北工业大学 | Video image stabilization method combining optical flow method and Kalman filtering |
| CN111898603A (en) * | 2020-08-10 | 2020-11-06 | 上海瑞美锦鑫健康管理有限公司 | Physical examination order recognition method and system based on deep neural network |
| CN112132151A (en) * | 2020-09-19 | 2020-12-25 | 娄忠富 | Image character recognition system and method based on recurrent neural network recognition algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108520247B (en) | Method, device, terminal and readable medium for identifying object node in image | |
| CN112348783B (en) | Image-based person identification method and device and computer-readable storage medium | |
| CN110176027B (en) | Video target tracking method, device, equipment and storage medium | |
| CN110046529B (en) | Two-dimensional code identification method, device and equipment | |
| CN110517283B (en) | Gesture tracking method, gesture tracking device and computer readable storage medium | |
| CN110956171A (en) | Automatic nameplate identification method and device, computer equipment and storage medium | |
| CN111612696B (en) | Image stitching method, device, medium and electronic equipment | |
| CN112101386B (en) | Text detection method, device, computer equipment and storage medium | |
| CN111985465A (en) | Text recognition method, device, equipment and storage medium | |
| CN113989604B (en) | Tire DOT information identification method based on end-to-end deep learning | |
| CN111401322A (en) | Station entering and exiting identification method and device, terminal and storage medium | |
| CN114037992A (en) | Instrument reading identification method and device, electronic equipment and storage medium | |
| CN115810197A (en) | Multi-mode electric power form recognition method and device | |
| CN115909378A (en) | Document text detection model training method and document text detection method | |
| CN111222355A (en) | Method and system for positioning bar code on PCB | |
| CN115188000A (en) | Text recognition method and device based on OCR (optical character recognition), storage medium and electronic equipment | |
| WO2024174726A1 (en) | Handwritten and printed text detection method and device based on deep learning | |
| CN115131795A (en) | Image scanning anti-shake recognition method and system and storage medium | |
| CN114998347B (en) | Semiconductor panel corner positioning method and device | |
| CN111274863A (en) | Text prediction method based on text peak probability density | |
| CN116432160A (en) | Slider verification code identification method and system based on RPA and LBP characteristics | |
| CN116229098A (en) | Image recognition method based on mask contour tracking and related products | |
| CN116246064A (en) | Multi-scale space feature enhancement method and device | |
| CN114511877A (en) | Behavior recognition method and device, storage medium and terminal | |
| CN115063813A (en) | Training method and training device of alignment model aiming at character distortion |
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 |