CN112100431B - Evaluation method, device and equipment of OCR system and readable storage medium - Google Patents
Evaluation method, device and equipment of OCR system and readable storage medium Download PDFInfo
- Publication number
- CN112100431B CN112100431B CN202011275415.1A CN202011275415A CN112100431B CN 112100431 B CN112100431 B CN 112100431B CN 202011275415 A CN202011275415 A CN 202011275415A CN 112100431 B CN112100431 B CN 112100431B
- Authority
- CN
- China
- Prior art keywords
- ocr system
- image
- character
- recognized
- deep learning
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5846—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using extracted text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
Abstract
The invention discloses an evaluation method, a device, equipment and a readable storage medium of an OCR system, wherein the method comprises the following steps: acquiring a training image, inputting the training image into an initial OCR system to train the initial OCR system, and obtaining a corresponding OCR system after the initial OCR system is trained; inputting the image to be recognized into an OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system; based on the character recognition result and the actual marking data corresponding to the image to be recognized, determining a character recall rate corresponding to the OCR system and a character recognition accuracy rate corresponding to the OCR system, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index. The method solves the problem that the evaluation index can not objectively reflect the overall performance of the OCR system due to the fact that character detection and character recognition are separately and independently evaluated in the prior art.
Description
Technical Field
The invention relates to the technical field of optical character recognition, in particular to an evaluation method, device and equipment of an OCR system and a readable storage medium.
Background
The OCR (optical character recognition) technology can convert printed characters in an image into a text format which can be processed by a computer, and the input and verification in the OCR technology are widely applied to scenes such as data comparison and the like, thereby becoming a key link for informatization and digitization application of various industries in national economy. With the continuous development of big data and deep learning technology, OCR technology has made a breakthrough, and OCR technology is widely applied to the application of printed document scanner recognition.
Currently, the evaluation of the recognition accuracy of the OCR system generally includes: and two links of text detection and text recognition. In the prior art, text detection mainly uses scores of a detection box and a labeling box when IOU =0.5 is a threshold as an evaluation index, and text recognition uses character accuracy or field accuracy as the evaluation index. Actually, in an OCR system, text recognition depends on a text detection positioning result, and sometimes a higher detection index may bring about a decrease in a recognition index instead, thereby causing a problem that the existing evaluation technology for the OCR system is difficult to reflect the overall performance of the OCR system.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide an OCR system evaluation method, an OCR system evaluation device, OCR system evaluation equipment and a readable storage medium, and aims to solve the technical problem that the existing OCR system evaluation technology is difficult to reflect the overall performance of an OCR system.
In order to achieve the above object, the present invention provides an evaluation method of an OCR system, comprising the steps of:
acquiring a training image, inputting the training image into an initial OCR system to train the initial OCR system, and obtaining a corresponding OCR system after the initial OCR system is trained;
inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
based on the character recognition result and the actual annotation data corresponding to the image to be recognized, determining a character recall rate corresponding to the OCR system and a character recognition accuracy rate corresponding to the OCR system, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index.
Optionally, the OCR system includes a character detection model and a character recognition model, and the step of inputting the training image into an initial OCR system to train the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained includes:
inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training is completed;
and training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is finished.
Optionally, the step of inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after the training is completed includes:
marking the training image to determine a preset marking position of a text box in the training image, inputting the training image containing the preset marking position to a first deep learning model, and determining a learning marking position corresponding to the training image;
determining first gradient information corresponding to the first deep learning model based on the preset labeling position and the learning labeling position;
optimizing the first deep learning model based on the first gradient information to determine the character detection model, wherein the character detection model is the optimized first deep learning model.
Optionally, the step of training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is completed includes:
acquiring a text strip image, and inputting the text strip image to the text detection model to obtain the text strip image with a marked text position;
inputting the text bar image labeled with the text position into a second deep learning model to obtain the learning text content corresponding to the text bar image;
and determining second gradient information corresponding to the second deep learning model based on the learned character content, and optimizing the second deep learning model based on the second gradient information to determine the character recognition model, wherein the character recognition detection model is the optimized second deep learning model.
Optionally, after the step of calculating an evaluation index of the OCR system based on the text recall rate and the text recognition accuracy rate to evaluate the performance of the OCR system based on the evaluation index, the method further includes:
when an image storage request is received, acquiring an image to be stored corresponding to the image storage request;
when the performance of the OCR system is up to standard, inputting the image to be stored into the OCR system, and determining whether identity card information exists in the image to be stored based on the OCR system;
and if the identity card information does not exist in the image to be stored, executing image storage operation corresponding to the image storage request.
Optionally, after the step of determining whether the identity card information exists in the image to be stored based on the OCR system, the method further includes:
if the identity card information exists in the image to be stored, inquiring a first target account associated with the identity card information, and matching the first target account with a second target account corresponding to local equipment;
if the first target account is matched with the second target account, executing image storage operation corresponding to the image storage request;
if the first target account is not matched with the second target account, sending verification information to the first target account, and executing image storage operation corresponding to the image storage request when feedback information corresponding to the verification information is received.
Optionally, the step of evaluating the performance of the OCR system based on the evaluation index includes:
and if the evaluation index of the OCR system reaches a preset threshold value, the performance of the OCR system reaches the standard.
Further, to achieve the above object, the present invention provides an evaluation apparatus of an OCR system, comprising:
the training module is used for acquiring a training image, inputting the training image into an initial OCR system, and training the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained;
the recognition module is used for inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
and the evaluation module is used for determining the character recall rate corresponding to the OCR system and the character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index.
Further, to achieve the above object, the present invention also provides an evaluation apparatus of an OCR system, comprising: a memory, a processor and an evaluation program of an OCR system stored on the memory and executable on the processor, the evaluation program of the OCR system when executed by the processor implementing the steps of the evaluation method of the OCR system as described above.
In addition, to achieve the above object, the present invention further provides a readable storage medium, on which an evaluation program of an OCR system is stored, the evaluation program of the OCR system, when executed by a processor, implementing the steps of the evaluation method of the OCR system as described above.
Training an initial OCR system by acquiring a training image and inputting the training image into the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained; then, inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system; and finally, determining the character recall rate corresponding to the OCR system and the character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating the evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index. In the embodiment, the character recall rate and the character recognition accuracy rate corresponding to the OCR system are calculated, and the evaluation index of the OCR system is calculated based on the character recall rate and the character recognition accuracy rate, so that the integral evaluation of the OCR system is performed, and the problem that the evaluation index cannot objectively reflect the integral performance of the OCR system due to the fact that character detection and character recognition are separately and independently evaluated in the prior art is solved. Further, the evaluation method of the OCR system uses a single index to evaluate the quality or performance of the OCR system, can help a user select a better OCR service, and promotes the development of informatization and digitization of various industries. In addition, due to the fact that the images have complex conditions that one-to-many matching, many-to-one matching and many-to-many matching exist between the answer frame and the detection frame due to the fact that the text frames are broken and adhered, misjudgment can be conducted on the conditions through an IOU-based evaluation mode in the prior art, and the evaluation method of the OCR system can effectively avoid misjudgment, so that evaluation of the model is more objective and fair.
Drawings
FIG. 1 is a schematic diagram of an evaluation device of an OCR system in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating a first embodiment of an OCR system evaluation method of the present invention;
FIG. 3 is a flowchart illustrating an evaluation method of an OCR system according to a second embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, fig. 1 is a schematic terminal structure diagram of a hardware operating environment according to an embodiment of the present invention.
The evaluation device of the OCR system in the embodiment of the present invention may be a PC, or may be a mobile terminal device having a display function, such as a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III, motion video Experts compression standard Audio Layer 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, motion video Experts compression standard Audio Layer 4) player, a portable computer, and the like.
As shown in fig. 1, the evaluation device of the OCR system may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Optionally, the evaluation device of the OCR system may further include a camera, a RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Such as light sensors, motion sensors, and other sensors. In particular, the light sensor may include an ambient light sensor that may adjust the brightness of the display screen according to the brightness of ambient light, and a proximity sensor that may turn off the display screen and/or backlight when an evaluation device of the OCR system is moved to the ear. As one of the motion sensors, the gravity acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes), detect the magnitude and direction of gravity when the mobile terminal is stationary, and can be used for applications (such as horizontal and vertical screen switching, related games, magnetometer attitude calibration) for recognizing the attitude of the mobile terminal, and related functions (such as pedometer and tapping) for vibration recognition.
Those skilled in the art will appreciate that the configuration of the evaluation equipment of the OCR system shown in figure 1 does not constitute a limitation of the evaluation equipment of the OCR system and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and an evaluation program of the OCR system.
In the evaluation device of the OCR system shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server and performing data communication with the background server; the user interface 1003 is mainly used for connecting a client (user side) and performing data communication with the client; and the processor 1001 may be used to invoke an evaluation program of the OCR system stored in the memory 1005.
In the present embodiment, the evaluation apparatus of the OCR system includes: a memory 1005, a processor 1001 and an evaluation program of the OCR system stored in the memory 1005 and operable on the processor 1001, wherein when the processor 1001 calls the evaluation program of the OCR system stored in the memory 1005, the following operations are performed:
acquiring a training image, inputting the training image into an initial OCR system to train the initial OCR system, and obtaining a corresponding OCR system after the initial OCR system is trained;
inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
based on the character recognition result and the actual annotation data corresponding to the image to be recognized, determining a character recall rate corresponding to the OCR system and a character recognition accuracy rate corresponding to the OCR system, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training is completed;
and training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is finished.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
marking the training image to determine a preset marking position of a text box in the training image, inputting the training image containing the preset marking position to a first deep learning model, and determining a learning marking position corresponding to the training image;
determining first gradient information corresponding to the first deep learning model based on the preset labeling position and the learning labeling position;
optimizing the first deep learning model based on the first gradient information to determine the character detection model, wherein the character detection model is the optimized first deep learning model.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
acquiring a text strip image, and inputting the text strip image to the text detection model to obtain the text strip image with a marked text position;
inputting the text bar image labeled with the text position into a second deep learning model to obtain the learning text content corresponding to the text bar image;
and determining second gradient information corresponding to the second deep learning model based on the learned character content, and optimizing the second deep learning model based on the second gradient information to determine the character recognition model, wherein the character recognition detection model is the optimized second deep learning model.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
when an image storage request is received, acquiring an image to be stored corresponding to the image storage request;
when the performance of the OCR system is up to standard, inputting the image to be stored into the OCR system, and determining whether identity card information exists in the image to be stored based on the OCR system;
and if the identity card information does not exist in the image to be stored, executing image storage operation corresponding to the image storage request.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
if the identity card information exists in the image to be stored, inquiring a first target account associated with the identity card information, and matching the first target account with a second target account corresponding to local equipment;
if the first target account is matched with the second target account, executing image storage operation corresponding to the image storage request;
if the first target account is not matched with the second target account, sending verification information to the first target account, and executing image storage operation corresponding to the image storage request when feedback information corresponding to the verification information is received.
Further, the processor 1001 may call an evaluation program of the OCR system stored in the memory 1005, and also perform the following operations:
and if the evaluation index of the OCR system reaches a preset threshold value, the performance of the OCR system reaches the standard.
The invention also provides an evaluation method of the OCR system, referring to fig. 2, and fig. 2 is a schematic flow chart of a first embodiment of the evaluation method of the OCR system of the invention.
Step S10, acquiring a training image, inputting the training image into an initial OCR system to train the initial OCR system, and obtaining a corresponding OCR system after the initial OCR system is trained;
the evaluation method of the OCR system provided by the invention is applied to the OCR system, the OCR system is an optical character recognition system and can recognize characters in an image and extract the characters in the image so as to convert the characters in the image into a format which can be processed by a computer, wherein the OCR system comprises a character detection model and a character recognition model. The character detection model and the character recognition model are deep learning models, the deep learning models can be network models such as a convolutional neural network or a cyclic neural network, and the network types to which the deep learning models belong are not limited in this embodiment. The character detection model is used for identifying the positions of characters in the picture, and the character identification model is used for identifying the character contents in the positions of the identified characters, namely identifying the character contents contained in the positions of the characters.
In the embodiment, in the process of training the OCR system, an initial OCR system is obtained first, a training image is obtained, and then the training image is input into the initial OCR model, so that the initial OCR model is trained based on the training image; and after the initial OCR system is trained, obtaining the corresponding OCR system after the initial OCR system is trained. The initial OCR system is in an initial state before being trained, the initial OCR system comprises a first deep learning model and a second deep learning model, the first deep learning model is used for training a character detection model, the second deep learning model is used for training a character recognition model, namely, the first deep learning model is the initial character detection model, and the second deep learning model is the initial character recognition model.
Step S20, inputting an image to be recognized into the OCR system, so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
in this embodiment, after training the initial OCR system to obtain the OCR system, the OCR system includes a character detection model and a character recognition model, and then an evaluation process of the OCR system is performed. Firstly, an image to be recognized is obtained, the image to be recognized is input into the OCR system, and a character recognition result corresponding to the image to be recognized is determined based on a character detection model and a character recognition model in the OCR system after training is completed. Specifically, an image to be recognized is input into a character detection model, and an intermediate recognition result corresponding to the image to be recognized is determined based on a first model parameter of the character detection model, wherein the intermediate recognition result is a text box obtained by recognizing the image to be recognized; and after the intermediate recognition result is obtained, inputting the image to be recognized containing the intermediate recognition result into the character recognition model to obtain a character recognition result. In order to improve the accuracy of evaluating the OCR system, the image to be recognized is an image which is inconsistent with the training image and contains character content, and the image to be recognized is used for evaluating the OCR system.
It should be noted that the image to be recognized is input into the character detection model, so that the character detection model determines the character position of the image to be recognized, that is, the text box corresponding to the image to be recognized determined by the character detection model is the character position in the image to be recognized. And inputting the image to be recognized containing the intermediate recognition result into the trained character recognition model, so that the character recognition model recognizes the character content in the image information of the image to be recognized based on the intermediate recognition result and the image information in the image to be recognized, thereby obtaining a character recognition result, namely, the character recognition result is the character content obtained after the character recognition model recognizes the image to be recognized.
Step S30, determining a character recall rate corresponding to the OCR system and a character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index.
In this embodiment, the text recall rate is a ratio between the number of correctly recognized characters and the actual number of characters, and the text recognition accuracy rate is a ratio between the number of correctly recognized characters in the text recognition result and all the characters in the text recognition result. The recall is the character recall rate, the precision is the character recognition precision, and the calculation formulas of the character recall rate recall and the character recognition precision are respectively as follows:
wherein the content of the first and second substances,
representing all the numbers of characters in the labeled answers of the images to be recognized,
the number of characters representing that the annotation answer of the image to be recognized is correctly recognized,
representing all the numbers of characters in the character recognition result of the image to be recognized,
representing the correct number of characters in the character recognition result of the image to be recognized.
And after the character recall rate and the character recognition accuracy rate are obtained, evaluating the OCR system, specifically, calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate of the OCR system for recognizing the image to be recognized so as to evaluate the performance of the OCR system based on the evaluation index subsequently. The scheme takes the score as an evaluation index of the OCR system, and the calculation formula of the evaluation index f1 is as follows:
further, if the evaluation index of the OCR system reaches a preset threshold value, the performance of the OCR system reaches the standard.
In the evaluation method of the OCR system provided by this embodiment, a training image is obtained and input into an initial OCR system, so as to train the initial OCR system, and obtain a corresponding OCR system after the initial OCR system is trained; then, inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system; and finally, determining the character recall rate corresponding to the OCR system and the character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating the evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index. In the embodiment, the character recall rate and the character recognition accuracy rate corresponding to the OCR system are calculated, and the evaluation teacher chart of the OCR system is calculated based on the character recall rate and the character recognition accuracy rate, so that the integral evaluation of the OCR system is performed, and the problem that in the prior art, the evaluation index cannot objectively reflect the integral performance of the OCR system due to the fact that character detection and character recognition are separately and independently evaluated is solved. Further, the evaluation method of the OCR system uses a single index to evaluate the quality or performance of the OCR system, can help a user select a better OCR service, and promotes the development of informatization and digitization of various industries. In addition, due to the fact that the images have complex conditions that one-to-many matching, many-to-one matching and many-to-many matching exist between the answer frame and the detection frame due to the fact that the text frames are broken and adhered, misjudgment can be conducted on the conditions through an IOU-based evaluation mode in the prior art, and the evaluation method of the OCR system can effectively avoid misjudgment, so that evaluation of the model is more objective and fair.
Based on the first embodiment, a second embodiment of the evaluation method of the OCR system of the present invention is proposed, referring to fig. 3, in this embodiment, step S10 includes:
step S11, inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training;
and step S12, training a second deep learning model based on the character detection model to obtain the corresponding character recognition model after the second deep learning model is trained.
In this embodiment, a first deep learning model in an initial OCR system is trained, a character detection model is obtained after the first deep learning model is completed, and then the first deep learning model and a second deep learning model are trained together. The OCR system comprises a character detection model and a character recognition model. Specifically, a training image is input into a first deep learning model in an initial OCR system for training, and after the first deep learning model is trained, a character detection model is obtained; and after the character detection model is obtained, training is carried out by combining the character detection model and the second deep learning model, and after the second deep learning model is trained, the character recognition model is obtained. The condition for completing the training of the first deep learning model or the second deep learning model may be that the training step reaches the maximum iteration step or the gradient corresponding to the gradient descent method reaches the minimum gradient value.
Further, the step of inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training is completed includes:
marking the training image to determine a preset marking position of a text box in the training image, inputting the training image containing the preset marking position to a first deep learning model, and determining a learning marking position corresponding to the training image;
determining first gradient information corresponding to the first deep learning model based on the preset labeling position and the learning labeling position;
optimizing the first deep learning model based on the first gradient information to determine the character detection model, wherein the character detection model is the optimized first deep learning model.
In this embodiment, the process of training the text detection model specifically includes the following steps: marking a training image, and marking a text box in the training image so as to determine a preset marking position of the text box in the training image; and then, inputting the training image containing the preset labeling position into a first deep learning model for training and learning, and outputting the learning labeling position corresponding to the training image by the first deep learning model. In the process of training the character detection model, optimizing the first deep learning model based on a gradient descent method, namely after obtaining a learning labeling position output by the first deep learning model, optimizing a first model parameter of the deep learning model based on first gradient information; and optimizing the first deep learning model based on the first gradient information until the first gradient information meets a first preset condition, and finishing training the first deep learning model to obtain a character detection model. The first preset condition may be that the first gradient information reaches a first minimum gradient value, and the first minimum gradient value may be set as needed.
Further, the step of training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is completed includes:
acquiring a text strip image, and inputting the text strip image to the text detection model to obtain the text strip image with a marked text position;
inputting the text bar image labeled with the text position into a second deep learning model to obtain the learning text content corresponding to the text bar image;
and determining second gradient information corresponding to the second deep learning model based on the learned character content, and optimizing the second deep learning model based on the second gradient information to determine the character recognition model, wherein the character recognition detection model is the optimized second deep learning model.
In this embodiment, the process of training the character recognition model is as follows: collecting a large number of single text strip images, determining preset character contents of the text strip images, and inputting the text strip images to the optimized character detection model so that the character detection model can output the text strip images marked with the detected character positions; then, inputting the text strip image with the marked character position into a second deep learning model for training, wherein the second deep learning model outputs the character content corresponding to the text strip image, namely outputs and identifies the character content in the text strip image; and then, optimizing the second deep learning model by using a gradient descent method, optimizing the first deep learning model based on the second gradient information, and completing the optimization of the second deep learning model until the second gradient information corresponding to the second deep learning model meets a second preset condition, thereby finally obtaining the character recognition model. The second preset condition may be that the second gradient information reaches a second minimum gradient value, and the second minimum gradient value may be set as needed.
Further, after the step of calculating an evaluation index of the OCR system based on the text recall rate and the text recognition accuracy rate to evaluate the performance of the OCR system based on the evaluation index, the method further comprises:
when an image storage request is received, acquiring an image to be stored corresponding to the image storage request;
when the performance of the OCR system is up to standard, inputting the image to be stored into the OCR system, and determining whether identity card information exists in the image to be stored based on the OCR system;
and if the identity card information does not exist in the image to be stored, executing image storage operation corresponding to the image storage request.
The identity card information comprises the identity card number, address, gender or native place of the client and the like. It should be noted that, when a client uses some terminal devices or platform systems, the client needs to upload the identity card information of the client in some cases, and therefore, these terminal devices or platform systems can acquire and store the identity card information of the client when the client uploads the identity card information, so that the client has an opportunity to steal the identity card information of the client, which causes the personal information and privacy of the client to be leaked, and thus, it is urgently needed to protect the privacy data of the client.
In this embodiment, when the terminal receives an image storage request to store an image to be stored of a client, the image to be stored corresponding to the image storage request is acquired, and an image recognition operation may be performed on the image to be stored based on the trained OCR system to recognize the image to be stored to identify whether the image to be stored includes the privacy data of the client. Therefore, no matter the terminal initiates an image storage request of any image, the image to be stored corresponding to the image storage request is obtained and identified so as to identify whether the image to be stored contains identity card information, and therefore the current image storage operation of the terminal can be monitored in real time so as to monitor whether the current storage operation is suspected of revealing privacy data of a client. And if the OCR system identifies that the image to be stored does not contain the identity card information, executing the image storage operation corresponding to the image storage request.
Further, after the step of determining whether the identity card information exists in the image to be stored based on the OCR system, the method further includes:
if the identity card information exists in the image to be stored, inquiring a first target account associated with the identity card information, and matching the first target account with a second target account corresponding to local equipment;
if the first target account is matched with the second target account, executing image storage operation corresponding to the image storage request;
if the first target account is not matched with the second target account, sending verification information to the first target account, and executing image storage operation corresponding to the image storage request when feedback information corresponding to the verification information is received.
In this embodiment, the target account is not limited, and further, the account information of the target account may include an international mobile subscriber identity number or a personal account number.
In this embodiment, if the OCR system recognizes that the image to be stored includes the identification card information, it indicates that the current image storage operation is suspected of revealing the privacy data of the client, so the image storage operation is prevented and controlled. Specifically, when the identity card information exists in the image to be stored, acquiring an identity card number in the identity card information based on an OCR system, and determining a first target account associated with the identity card information through the identity card number, such as determining a mobile phone number associated with the identity card number through the identity card number; then, a second target account bound by the local device (client terminal) is obtained, for example, when the local device is a mobile phone, the phone number of the SIM card on the local device can be obtained.
And then matching the first target account with the second target account. And when the first target account is matched with the second target account, the local terminal is allowed to execute the image storage operation corresponding to the image storage request operation if the local terminal is the security device held by the client. Conversely, if the first target account and the second target account are not matched, it indicates that the local terminal is not the device of the client and belongs to the unsafe device, and at this time, the local terminal cannot be allowed to execute the image storage operation corresponding to the image storage request operation. Instead, the verification information is sent to the first target account, and if feedback information fed back by the first target account is received, the image storage operation corresponding to the image storage request operation is executed.
According to the evaluation method of the OCR system provided by the embodiment, the training image is input into the first deep learning model, so that a character detection model corresponding to the first deep learning model after training is completed is obtained; and training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is finished. In this embodiment, first deep learning model in the initial OCR system is trained, and after the training is completed, the character detection model is obtained, and then the character recognition model is obtained by training in combination with the character detection model, so that the degree of cooperation between the character detection model and the character recognition model can be improved, and the accuracy of the OCR system can be further improved.
In addition, an embodiment of the present invention further provides an evaluation apparatus for an OCR system, where the evaluation apparatus for the OCR system includes:
the training module is used for acquiring a training image, inputting the training image into an initial OCR system, and training the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained;
the recognition module is used for inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
and the evaluation module is used for determining the character recall rate corresponding to the OCR system and the character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index.
Further, the training module is further configured to:
inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training is completed;
and training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is finished.
Further, the training module is further configured to:
marking the training image to determine a preset marking position of a text box in the training image, inputting the training image containing the preset marking position to a first deep learning model, and determining a learning marking position corresponding to the training image;
determining first gradient information corresponding to the first deep learning model based on the preset labeling position and the learning labeling position;
optimizing the first deep learning model based on the first gradient information to determine the character detection model, wherein the character detection model is the optimized first deep learning model.
Further, the training module is further configured to:
acquiring a text strip image, and inputting the text strip image to the text detection model to obtain the text strip image with a marked text position;
inputting the text bar image labeled with the text position into a second deep learning model to obtain the learning text content corresponding to the text bar image;
and determining second gradient information corresponding to the second deep learning model based on the learned character content, and optimizing the second deep learning model based on the second gradient information to determine the character recognition model, wherein the character recognition detection model is the optimized second deep learning model.
Further, the evaluation module is further configured to:
when an image storage request is received, acquiring an image to be stored corresponding to the image storage request;
when the performance of the OCR system is up to standard, inputting the image to be stored into the OCR system, and determining whether identity card information exists in the image to be stored based on the OCR system;
and if the identity card information does not exist in the image to be stored, executing image storage operation corresponding to the image storage request.
Further, the evaluation module is further configured to:
if the identity card information exists in the image to be stored, inquiring a first target account associated with the identity card information, and matching the first target account with a second target account corresponding to local equipment;
if the first target account is matched with the second target account, executing image storage operation corresponding to the image storage request;
if the first target account is not matched with the second target account, sending verification information to the first target account, and executing image storage operation corresponding to the image storage request when feedback information corresponding to the verification information is received.
Further, the evaluation module is further configured to:
and if the evaluation index of the OCR system reaches a preset threshold value, the performance of the OCR system reaches the standard.
Furthermore, an embodiment of the present invention further provides a readable storage medium, where an evaluation program of an OCR system is stored, and when executed by a processor, the evaluation program of the OCR system implements the steps of the evaluation method of the OCR system as described in any one of the above.
The specific embodiment of the readable storage medium of the present invention is substantially the same as the embodiments of the evaluation method of the OCR system, and is not described in detail herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system 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 system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An evaluation method of an OCR system, comprising the steps of:
acquiring a training image, inputting the training image into an initial OCR system to train the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained, wherein the OCR system comprises a character detection model and a character recognition model;
inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
determining a character recall rate corresponding to the OCR system and a character recognition accuracy rate corresponding to the OCR system based on the character recognition result and actual annotation data corresponding to the image to be recognized, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index;
wherein the step of inputting the image to be recognized into the OCR system to determine the character recognition result corresponding to the image to be recognized based on the OCR system comprises:
inputting an image to be recognized into the character detection model, and determining an intermediate recognition result corresponding to the image to be recognized based on the character detection model, wherein the intermediate recognition result is a text box obtained by recognizing the image to be recognized by the character detection model; after the intermediate recognition result is obtained, inputting the image to be recognized containing the intermediate recognition result into the character recognition model to obtain a character recognition result corresponding to the image to be recognized;
wherein the calculation formula for calculating the evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate is as follows:
wherein f1 represents the evaluation index, recall represents the text recall rate, and precision represents the text recognition accuracy rate.
2. An assessment method according to claim 1, wherein said OCR system comprises a character detection model and a character recognition model, said step of inputting said training image into an initial OCR system to train said initial OCR system, resulting in a corresponding OCR system after training said initial OCR system comprises:
inputting the training image into a first deep learning model to obtain a character detection model corresponding to the first deep learning model after training is completed;
and training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after the training is finished.
3. An evaluation method according to claim 2, wherein the step of inputting the training image into a first deep learning model to obtain a corresponding character detection model after the first deep learning model is trained comprises:
marking the training image to determine a preset marking position of a text box in the training image, inputting the training image containing the preset marking position to a first deep learning model, and determining a learning marking position corresponding to the training image;
determining first gradient information corresponding to the first deep learning model based on the preset labeling position and the learning labeling position;
optimizing the first deep learning model based on the first gradient information to determine the character detection model, wherein the character detection model is the optimized first deep learning model.
4. An evaluation method according to claim 2, wherein the step of training a second deep learning model based on the character detection model to obtain the character recognition model corresponding to the second deep learning model after training comprises:
acquiring a text strip image, and inputting the text strip image to the text detection model to obtain the text strip image with a marked text position;
inputting the text bar image labeled with the text position into a second deep learning model to obtain the learning text content corresponding to the text bar image;
and determining second gradient information corresponding to the second deep learning model based on the learned character content, and optimizing the second deep learning model based on the second gradient information to determine the character recognition model, wherein the character recognition detection model is the optimized second deep learning model.
5. An evaluation method for an OCR system according to claim 1, wherein said step of calculating an evaluation index of said OCR system based on said text recall ratio and said text recognition accuracy to evaluate the performance of said OCR system based on said evaluation index further comprises:
when an image storage request is received, acquiring an image to be stored corresponding to the image storage request;
when the performance of the OCR system is up to standard, inputting the image to be stored into the OCR system, and determining whether identity card information exists in the image to be stored based on the OCR system;
and if the identity card information does not exist in the image to be stored, executing image storage operation corresponding to the image storage request.
6. An evaluation method for an OCR system according to claim 5, wherein after the step of determining whether identity card information exists in the image to be stored based on the OCR system, further comprising:
if the identity card information exists in the image to be stored, inquiring a first target account associated with the identity card information, and matching the first target account with a second target account corresponding to local equipment;
if the first target account is matched with the second target account, executing image storage operation corresponding to the image storage request;
if the first target account is not matched with the second target account, sending verification information to the first target account, and executing image storage operation corresponding to the image storage request when feedback information corresponding to the verification information is received.
7. An evaluation method for an OCR system according to any one of claims 1 to 6, wherein said step of evaluating performance of said OCR system based on said evaluation index includes:
and if the evaluation index of the OCR system reaches a preset threshold value, the performance of the OCR system reaches the standard.
8. An evaluation apparatus of an OCR system, comprising:
the training module is used for acquiring a training image and inputting the training image into an initial OCR system so as to train the initial OCR system to obtain a corresponding OCR system after the initial OCR system is trained, wherein the OCR system comprises a character detection model and a character recognition model;
the recognition module is used for inputting an image to be recognized into the OCR system so as to determine a character recognition result corresponding to the image to be recognized based on the OCR system;
the evaluation module is used for determining the character recall rate corresponding to the OCR system and the character recognition accuracy rate corresponding to the OCR system based on the character recognition result and the actual annotation data corresponding to the image to be recognized, and calculating an evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate so as to evaluate the performance of the OCR system based on the evaluation index;
wherein the step of inputting the image to be recognized into the OCR system to determine the character recognition result corresponding to the image to be recognized based on the OCR system comprises:
inputting an image to be recognized into the character detection model, and determining an intermediate recognition result corresponding to the image to be recognized based on the character detection model, wherein the intermediate recognition result is a text box obtained by recognizing the image to be recognized by the character detection model; after the intermediate recognition result is obtained, inputting the image to be recognized containing the intermediate recognition result into the character recognition model to obtain a character recognition result corresponding to the image to be recognized;
wherein the calculation formula for calculating the evaluation index of the OCR system based on the character recall rate and the character recognition accuracy rate is as follows:
wherein f1 represents the evaluation index, recall represents the text recall rate, and precision represents the text recognition accuracy rate.
9. An evaluation apparatus of an OCR system, comprising: memory, a processor and an evaluation program of an OCR system stored on the memory and executable on the processor, the evaluation program of the OCR system when executed by the processor implementing the steps of the evaluation method of the OCR system according to any of claims 1 to 7.
10. A readable storage medium, characterized in that the readable storage medium has stored thereon an evaluation program of an OCR system, which when executed by a processor implements the steps of the evaluation method of the OCR system according to any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011275415.1A CN112100431B (en) | 2020-11-16 | 2020-11-16 | Evaluation method, device and equipment of OCR system and readable storage medium |
| PCT/CN2021/127185 WO2022100452A1 (en) | 2020-11-16 | 2021-10-28 | Ocr system evaluation method, apparatus and device, and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011275415.1A CN112100431B (en) | 2020-11-16 | 2020-11-16 | Evaluation method, device and equipment of OCR system and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112100431A CN112100431A (en) | 2020-12-18 |
| CN112100431B true CN112100431B (en) | 2021-02-26 |
Family
ID=73785570
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011275415.1A Active CN112100431B (en) | 2020-11-16 | 2020-11-16 | Evaluation method, device and equipment of OCR system and readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112100431B (en) |
| WO (1) | WO2022100452A1 (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112100431B (en) * | 2020-11-16 | 2021-02-26 | 深圳壹账通智能科技有限公司 | Evaluation method, device and equipment of OCR system and readable storage medium |
| CN113220557B (en) * | 2021-06-01 | 2024-01-26 | 上海明略人工智能(集团)有限公司 | Cold start recommendation model evaluation method, system, computer equipment and storage medium |
| CN115061590B (en) * | 2022-08-17 | 2023-03-17 | 芯见(广州)科技有限公司 | KVM (keyboard video mouse) seat system control method based on video identification and KVM seat system |
| CN115512348B (en) * | 2022-11-08 | 2023-03-28 | 浪潮金融信息技术有限公司 | Object identification method, system, equipment and medium based on double identification technology |
| CN116612483B (en) * | 2023-07-19 | 2023-09-29 | 广州宏途数字科技有限公司 | Recognition method and device for handwriting vector of intelligent pen |
| CN117217876B (en) * | 2023-11-08 | 2024-03-26 | 深圳市明心数智科技有限公司 | Order preprocessing method, device, equipment and medium based on OCR technology |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技（深圳）有限公司 | OCR recognition methods and its electronic equipment |
| CN109934227A (en) * | 2019-03-12 | 2019-06-25 | 上海兑观信息科技技术有限公司 | System for recognizing characters from image and method |
| CN111191198A (en) * | 2019-11-25 | 2020-05-22 | 京东数字科技控股有限公司 | Account information processing method and device, computer readable medium and electronic equipment |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9177210B2 (en) * | 2007-10-30 | 2015-11-03 | Hki Systems And Service Llc | Processing container images and identifiers using optical character recognition and geolocation |
| US10510131B2 (en) * | 2018-03-07 | 2019-12-17 | Ricoh Company, Ltd. | Return mail services |
| CN108764226B (en) * | 2018-04-13 | 2022-05-03 | 顺丰科技有限公司 | Image text recognition method, device, equipment and storage medium thereof |
| US11037010B2 (en) * | 2019-01-11 | 2021-06-15 | Microsoft Technology Licensing, Llc | Compositional model for text recognition |
| CN110097049A (en) * | 2019-04-03 | 2019-08-06 | 中国科学院计算技术研究所 | A kind of natural scene Method for text detection and system |
| CN110399871A (en) * | 2019-06-14 | 2019-11-01 | 华南理工大学 | A kind of appraisal procedure of scene text testing result |
| CN112100431B (en) * | 2020-11-16 | 2021-02-26 | 深圳壹账通智能科技有限公司 | Evaluation method, device and equipment of OCR system and readable storage medium |
-
2020
- 2020-11-16 CN CN202011275415.1A patent/CN112100431B/en active Active
-
2021
- 2021-10-28 WO PCT/CN2021/127185 patent/WO2022100452A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技（深圳）有限公司 | OCR recognition methods and its electronic equipment |
| CN109934227A (en) * | 2019-03-12 | 2019-06-25 | 上海兑观信息科技技术有限公司 | System for recognizing characters from image and method |
| CN111191198A (en) * | 2019-11-25 | 2020-05-22 | 京东数字科技控股有限公司 | Account information processing method and device, computer readable medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112100431A (en) | 2020-12-18 |
| WO2022100452A1 (en) | 2022-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112100431B (en) | Evaluation method, device and equipment of OCR system and readable storage medium | |
| CN109919014B (en) | OCR (optical character recognition) method and electronic equipment thereof | |
| US10339383B2 (en) | Method and system for providing augmented reality contents by using user editing image | |
| EP3716132A1 (en) | Code scanning method, code scanning device and mobile terminal | |
| US9098888B1 (en) | Collaborative text detection and recognition | |
| CN109993150B (en) | Method and device for identifying age | |
| CN109919244B (en) | Method and apparatus for generating a scene recognition model | |
| KR102002024B1 (en) | Method for processing labeling of object and object management server | |
| CN110335139B (en) | Similarity-based evaluation method, device and equipment and readable storage medium | |
| CN109194689B (en) | Abnormal behavior recognition method, device, server and storage medium | |
| CN110264093B (en) | Credit model establishing method, device, equipment and readable storage medium | |
| CN109726110B (en) | Document testing method, device, apparatus and computer readable storage medium | |
| CN106504050A (en) | A kind of information comparison device and method | |
| CN112418214A (en) | Vehicle identification code identification method and device, electronic equipment and storage medium | |
| CN109947988B (en) | Information processing method and device, terminal equipment and server | |
| CN115205883A (en) | Data auditing method, device, equipment and storage medium based on OCR (optical character recognition) and NLP (non-line language) | |
| CN111881740A (en) | Face recognition method, face recognition device, electronic equipment and medium | |
| CN109885490B (en) | Picture comparison method and device | |
| CN110008926B (en) | Method and device for identifying age | |
| CN112950443A (en) | Adaptive privacy protection method, system, device and medium based on image sticker | |
| CN110046571B (en) | Method and device for identifying age | |
| CN113190646A (en) | User name sample labeling method and device, electronic equipment and storage medium | |
| CN113269730B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN111353422B (en) | Information extraction method and device and electronic equipment | |
| CN112395450A (en) | Picture character detection method and device, computer equipment and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Evaluation method, device, equipment and readable storage medium of OCR system Effective date of registration: 20210918 Granted publication date: 20210226 Pledgee: Shenzhen hi tech investment small loan Co.,Ltd. Pledgor: ONECONNECT FINANCIAL TECHNOLOGY Co.,Ltd. (SHANGHAI) Registration number: Y2021980009610 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20221130 Granted publication date: 20210226 Pledgee: Shenzhen hi tech investment small loan Co.,Ltd. Pledgor: ONECONNECT FINANCIAL TECHNOLOGY Co.,Ltd. (SHANGHAI) Registration number: Y2021980009610 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right |