Identification and evaluation system based on OCR (Optical Character Recognition) technology

Abstract

The invention provides a recognition and evaluation system based on an OCR (Optical Character Recognition) technology, and the system comprises an interface calling module which is used for calling an OCR interface recognition module for each piece of image-text data selected by a user; the OCR interface recognition module is used for determining a corresponding pre-training classifier based on the corresponding image-text data template, and performing OCR recognition on the corresponding image-text data by using the pre-training classifier to obtain a recognition result; the evaluation module is used for carrying out image-text data evaluation on the corresponding image-text data based on the identification result and a preset evaluation rule, and carrying out user evaluation on the user based on each image-text data evaluation result to obtain a user evaluation result; and the display module is used for displaying the user evaluation result to the user. According to the technical scheme provided by the embodiment of the invention, the recognition accuracy of related image-text data in an evaluation process can be improved, and the accuracy of an evaluation result is further improved; and the working efficiency of the evaluation process can be effectively improved by automatically identifying related image-text data.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an identification and evaluation system based on OCR technology. Background Art

[0002] With the development of computer and information technology, there is an increasing demand for automated processing of text information in graphic materials (such as paper documents and images). This includes digitizing paper documents, automatic form recognition, automated bill processing, etc.

[0003] At home and abroad, quantitative evaluation methods have penetrated into all levels of management of enterprises and institutions, such as performance evaluation, financial analysis, risk evaluation, etc. Enterprises and institutions hope to understand the work status of the evaluation object and their own business conditions more accurately and objectively through this method, so as to make more informed decisions. How to quickly, accurately and intelligently identify graphic materials related to quantitative evaluation and how to use the identification results for quantitative evaluation have become the key considerations for various enterprises and institutions in conducting quantitative evaluation.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, the present application aims to provide a recognition and evaluation system based on OCR technology, which can improve the recognition accuracy of relevant graphic and text materials in the evaluation process, thereby improving the accuracy of the evaluation results. By automatically recognizing relevant graphic and text materials, the work efficiency of the evaluation process can be effectively improved.

[0006] The present application embodiment provides a recognition evaluation system based on OCR technology, the system comprising:

[0007] The interface calling module is used to call the OCR interface recognition module for each graphic material selected by the user, wherein the graphic material includes the personal graphic material uploaded by the user through the corresponding graphic material template, and / or the group graphic material uploaded by the administrator claimed by the user through the corresponding graphic material template;

[0008] An OCR interface recognition module, used to determine a corresponding pre-trained classifier based on a corresponding graphic material template, and use the pre-trained classifier to perform OCR recognition on the corresponding graphic material to obtain a recognition result, wherein the corresponding graphic material template is a template applied when the corresponding graphic material is uploaded;

[0009] An evaluation module, used to evaluate the corresponding graphic materials based on the recognition results and preset evaluation rules, and to evaluate the user based on each graphic material evaluation result to obtain a user evaluation result;

[0010] A display module is used to display the user evaluation result to the user.

[0011] According to the technical solution provided in the embodiment of the present application, the recognition result includes keywords in the corresponding graphic material, and the preset evaluation rule includes a one-to-one correspondence between keywords and corresponding scores. Based on this, the evaluation module is specifically used to determine the score corresponding to the corresponding graphic material based on the keywords and the one-to-one correspondence between the keywords and the corresponding scores. After the score of each graphic material selected by the user is determined, the score of each graphic material is added to obtain the user's total score, and the user's total score is the user evaluation result.

[0012] According to the technical solution provided by the embodiment of the present application, the OCR interface recognition module includes a template creation unit, wherein the template creation unit is used to:

[0013] Receive reference and text materials uploaded by users;

[0014] Using OCR technology to identify reference fields and areas to be identified in reference text materials, wherein the reference fields are text content with fixed positions in the reference text materials, and the areas to be identified are areas corresponding to keywords to be identified in the reference text materials;

[0015] A graphic material template is created based on the reference field and the area to be identified, and a graphic material template ID is generated, wherein the graphic material template has the same format as the reference graphic material.

[0016] According to the technical solution provided in the embodiment of the present application, the interface calling module also includes a target graphic material template calling unit, wherein the target graphic material template calling unit is used to receive the target graphic material template ID input by the user before calling the OCR interface recognition module for each graphic material selected by the user, and request the OCR interface recognition module to call the target graphic material template corresponding to the target graphic material template ID, and save the called target graphic material template to a preset storage space, wherein the target graphic material template is used for users to upload personal graphic materials.

[0017] According to the technical solution provided in the embodiment of the present application, the OCR interface recognition module further includes a classifier training unit, wherein the classifier training unit is used to:

[0018] In response to a user's request to create an initial classifier, an initial classifier is created, wherein the initial classifier is configured with a classifier-related graphic and text material template selected by the user;

[0019] Based on the image and text material training set uploaded by the user and / or the classification keywords corresponding to the classifier-related image and text material template, the initial classifier is trained to obtain a pre-trained classifier, wherein the image and text material training set is a set of image and text materials with the same layout uploaded for each classifier-related image and text material template, and the pre-trained classifier is used to identify the image and text materials and classify the image and text materials into the corresponding image and text material template.

[0020] According to the technical solution provided in the embodiment of the present application, the pre-trained classifier includes three types: image feature classifier, text feature classifier and graphic-text collaborative classifier. Among them, the image feature classifier is trained based on the graphic-text data training set, the text feature classifier is trained based on the classification keywords corresponding to the graphic-text data template related to the classifier, and the graphic-text collaborative classifier is trained based on the classification keywords corresponding to the graphic-text data training set and the classifier related graphic-text data template.

[0021] According to the technical solution provided in the embodiment of the present application, the system also includes a preset evaluation rule determination module, wherein the preset evaluation rule determination module is used to determine the scoring items and the evaluation tags corresponding to each scoring item, and set a corresponding score and a keyword corresponding to the corresponding score for each evaluation tag, wherein the preset evaluation rules include one or more scoring items, and each scoring item corresponds to one or more evaluation tags.

[0022] According to the technical solution provided in the embodiment of the present application, the evaluation module also includes a group evaluation unit, wherein the group evaluation unit is used to perform weighted evaluation based on preset weights according to the evaluation results of each user in the corresponding group and the evaluation results given by management personnel for the corresponding group to obtain a group evaluation result.

[0023] According to the technical solution provided in the embodiment of the present application, the system further includes an evaluation ranking module and an information push module, wherein the evaluation ranking module is used to rank each user according to the user evaluation result of each user in the group and to rank each group according to the evaluation result of each group;

[0024] The information push module is used to push the ranking results obtained by the evaluation ranking module to the user.

[0025] According to the technical solution provided in the embodiment of the present application, the evaluation ranking module also includes a group dynamic scoring unit, wherein the group dynamic scoring unit is used to dynamically score each group according to the ranking order of each group according to preset dynamic scoring rules.

[0026] In summary, the present application proposes an identification and evaluation system based on OCR technology, including an interface calling module, which is used to call an OCR interface identification module for each graphic and text material selected by a user, wherein the graphic and text material includes personal graphic and text materials uploaded by the user through a corresponding graphic and text material template, and / or collective graphic and text materials uploaded by an administrator claimed by the user through a corresponding graphic and text material template; an OCR interface identification module, which is used to determine a corresponding pre-trained classifier based on the corresponding graphic and text material template, and use the pre-trained classifier to perform OCR identification on the corresponding graphic and text material to obtain a recognition result, wherein the corresponding graphic and text material template is a template applied when the corresponding graphic and text material is uploaded; an evaluation module, which is used to perform graphic and text material evaluation on the corresponding graphic and text material based on the recognition result and a preset evaluation rule, and perform a user evaluation on the user based on each graphic and text material evaluation result to obtain a user evaluation result; and a display module, which is used to display the user evaluation result to the user. By using OCR technology to identify the graphic and text materials selected by the user, the recognition accuracy of relevant graphic and text materials in the evaluation process can be improved, thereby improving the accuracy of the evaluation results; by automatically identifying relevant graphic and text materials instead of manual recognition, the work efficiency of the evaluation process can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the structure of an OCR-based recognition and evaluation system provided in an embodiment of the present application;

[0028] Figure 2 A front-end initial interface provided by an embodiment of the present application before a user uploads reference text materials;

[0029] Figure 3 A front-end interface provided by an embodiment of the present application after a user uploads reference text and image data;

[0030] Figure 4 A front-end interface for selecting a reference field in a reference text material provided in an embodiment of the present application;

[0031] Figure 5 A front-end interface for selecting a region to be identified in a reference graphic material provided in an embodiment of the present application;

[0032] Figure 6 An interface for creating an initial classifier provided in an embodiment of the present application;

[0033] Figure 7 An editing interface of an initial classifier provided in an embodiment of the present application;

[0034] Figure 8 A front-end interface for selecting a classifier-related graphic material template for an initial classifier provided in an embodiment of the present application;

[0035] Fig. 9 A front-end interface after the configuration of a classifier-related graphic and text data template is completed provided in an embodiment of the present application;

[0036] Fig.10 A front-end interface for uploading a training set of graphic materials provided in an embodiment of the present application;

[0037] Fig.11 A preset evaluation rule display interface provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] As mentioned in the background technology, in response to the problems in the prior art, this application proposes a recognition and evaluation system based on OCR technology. This embodiment is applicable to situations where it is necessary to evaluate the objects to be evaluated in a group based on uploaded graphic and text materials. Figure 1 A schematic diagram of the structure of an OCR-based recognition and evaluation system provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the recognition and evaluation system based on OCR technology includes:

[0041] The interface calling module 10 is used to call the OCR interface recognition module 11 for each graphic material selected by the user, wherein the graphic material includes personal graphic materials uploaded by the user through the corresponding graphic material template, and / or collective graphic materials uploaded by the administrator claimed by the user through the corresponding graphic material template.

[0042] The user is the object to be evaluated, for example, the user can be an individual member of a group, where the group can be a department or branch, etc. The graphic materials can be the relevant science and technology or management certificates (here only for example, not limiting, and can also be other categories) issued by the current user's country, province, city, industry association, group company, etc., for example, they can include national professional qualification certificates, paper award certificates, patent certificates, software copyright certificates, appraisal certificates, achievement certificates, honorary certificates, and award documents, etc.

[0043] In the specific operation process, the user can select the graphic material to be recognized in the current graphic material display interface, and the interface calling module 10 responds to the user's selection operation and calls the OCR (Optical Character Recognition, text recognition technology) interface recognition module 11 to recognize the graphic material. Optionally, each time the user selects a graphic material, the interface calling module 10 calls the OCR interface recognition module 11 once for the graphic material, and optionally, the user simultaneously selects multiple graphic materials with the same graphic material template, and the interface calling module 10 calls the OCR interface recognition module 11 once for the multiple graphic materials. It should be noted that in this embodiment, the graphic and text materials displayed in the current graphic and text material display interface can be the graphic and text materials of the individual uploaded by the user through the corresponding graphic and text material template, such as an advanced individual honorary certificate or a first prize certificate for scientific and technological progress, etc., or it can be the graphic and text materials of the group uploaded by the administrator claimed by the user through the corresponding graphic and text material template, such as an advanced department or a safety culture demonstration team, etc., wherein the corresponding graphic and text material template has the same layout as the graphic and text material uploaded through the template. The graphic and text material template in this embodiment can be a system-provided one, or it can be created by the user according to his own needs. Each graphic and text material template corresponds to a graphic and text material template ID (Identity document).

[0044] Preferably, the OCR technology-based recognition and evaluation system in this embodiment may also include a graphic and text material management module, which can manage the graphic and text materials uploaded by users or administrators, and specifically may include classifying and displaying the graphic and text materials, allowing users or administrators to edit, delete, and view the details of the graphic and text materials, and also allowing users or administrators to upload new graphic and text materials and other operations.

[0045] The OCR interface recognition module 11 is used to determine the corresponding pre-trained classifier based on the corresponding graphic material template, and use the pre-trained classifier to perform OCR recognition on the corresponding graphic material to obtain a recognition result, wherein the corresponding graphic material template is the template applied when the corresponding graphic material is uploaded.

[0046] The OCR interface recognition module 11 in this embodiment can determine the corresponding graphic material template ID according to the graphic material selected by the user, and refer to the one-to-one correspondence between the pre-stored graphic material template ID and the pre-trained classifier ID to determine the pre-trained classifier used to identify the corresponding graphic material, and use the pre-trained classifier to perform OCR recognition on the corresponding graphic material. It can be understood that when calling the OCR interface recognition module 11, if the user selects one graphic material, the recognition result is the recognition result corresponding to one graphic material, and if the user selects multiple graphic materials with the same graphic material template at the same time, the recognition result is the recognition result corresponding to multiple graphic materials.

[0047] In this embodiment, the implementation process of the OCR recognition technology adopted by the OCR interface recognition module 11 can be roughly divided into the following steps: (1) Image preprocessing, which can specifically include: 1) Image input: input paper documents or pictures into the computer through a scanner, digital camera, mobile phone and other devices. 2) Grayscale: convert the color image into a grayscale image to simplify the image complexity. 3) Binarization: further convert the grayscale image into a binary image, that is, the text part in the image becomes black and the background becomes white, which helps to simplify the image information and facilitate subsequent text extraction and recognition. 4) Denoising: remove the messy information in the image, such as noise, stains, etc., to improve the clarity of the text. 5) Tilt correction: adjust the image direction to ensure that the text is arranged horizontally. (2) Text area detection, which can specifically include: 1) Edge detection: determine the area in the image that may contain text through algorithms such as image analysis and edge detection. 2) Contour analysis: further analyze the contour of the text area to more accurately locate the text area. (3) Character segmentation, which can specifically include: 1) segment the detected text area into individual characters. This process relies on features such as the spacing and connectivity between characters. 2) Feature extraction for each character. Common methods include projection, template matching, neural networks, etc. The extracted features include shape, angle, texture, etc., for subsequent classification and recognition. (4) Character recognition, which may specifically include: 1) Using a trained model or algorithm as a classifier to classify the extracted character features. These classifiers can be pre-trained models or custom trained models. 2) Converting the character results recognized by the classifier into a computer-recognizable character encoding. With the development of deep learning technology, recognition models based on convolutional neural networks have achieved remarkable results in the field of OCR. (5) Post-processing, which may specifically include: 1) Error correction: Correct the recognized text and remove erroneous characters. 2) Typesetting: Typeset the recognized text to meet the format requirements of the document. 3) Output: Output the recognition results in text form for subsequent processing by the user. The output format can be TXT, DOC, PDF and other formats.

[0048] The evaluation module 12 is used to evaluate the corresponding graphic materials based on the recognition results and preset evaluation rules, and to evaluate the user based on each graphic material evaluation result to obtain a user evaluation result.

[0049] Preferably, the recognition result includes keywords in the corresponding graphic material, and the preset evaluation rule includes a one-to-one correspondence between keywords and corresponding scores. Based on this, the evaluation module 12 is specifically used to determine the score corresponding to the corresponding graphic material based on the keywords and the one-to-one correspondence between the keywords and the corresponding scores. After the score of each graphic material selected by the user is determined, the score of each graphic material is added to obtain the user's total score, and the user's total score is the user evaluation result.

[0050] Exemplarily, the keywords obtained from the recognition result include province, science and technology, management, innovation, and third prize. The evaluation module 12 uses the above keywords to match the scores in the preset evaluation rules and determines that the corresponding score is 7. Then, the evaluation score of the graphic material corresponding to the above keywords can be determined to be 7. According to the above steps, the evaluation scores corresponding to other graphic materials selected by the user can also be determined. Another exemplary example is that the user has selected 5 graphic materials in total, and the evaluation scores of each graphic material determined by the evaluation module 12 according to the above steps are 6, 7, 9, 8, and 7 respectively. Then, the user evaluation result of the user can be determined to be 6+7+9+8+7=37.

[0051] The display module 13 is used to display the user evaluation result to the user.

[0052] In this embodiment, it is written that the system also includes a certificate management module

[0053] The embodiment of the present application proposes an identification and evaluation system based on OCR technology, including an interface calling module, which is used to call an OCR interface identification module for each graphic material selected by a user, wherein the graphic material includes personal graphic materials uploaded by the user through a corresponding graphic material template, and / or graphic materials of a group uploaded by an administrator claimed by the user through a corresponding graphic material template; an OCR interface identification module, which is used to determine a corresponding pre-trained classifier based on the corresponding graphic material template, and use the pre-trained classifier to perform OCR recognition on the corresponding graphic material to obtain a recognition result, wherein the corresponding graphic material template is a template applied when the corresponding graphic material is uploaded; an evaluation module, which is used to perform graphic material evaluation on the corresponding graphic material based on the recognition result and a preset evaluation rule, and perform user evaluation on the user based on each graphic material evaluation result to obtain a user evaluation result; and a display module, which is used to display the user evaluation result to the user. By using OCR technology to identify the graphic material selected by the user, the recognition accuracy of the relevant graphic materials in the evaluation process can be improved, thereby improving the accuracy of the evaluation result; the work efficiency of the evaluation process can be effectively improved by automatically identifying the relevant graphic materials instead of manual recognition.

[0054] On the basis of the above embodiments, further, the OCR interface recognition module 11 includes a template creation unit, wherein the template creation unit is used to:

[0055] Receive reference and text materials uploaded by users;

[0056] Using OCR technology to identify reference fields and areas to be identified in reference text materials, wherein the reference fields are text content with fixed positions in the reference text materials, and the areas to be identified are areas corresponding to keywords to be identified in the reference text materials;

[0057] A graphic material template is created based on the reference field and the area to be identified, and a graphic material template ID is generated, wherein the graphic material template has the same format as the reference graphic material.

[0058] When a user needs to upload a certain format of graphic material in the system, if the graphic material template of the format does not exist in the system, the user can create a graphic material template of the format in the OCR interface recognition module 11, and after the creation, the graphic material template of the format can be called locally for use. The above-mentioned user can create a graphic material template according to actual needs, which can meet the user's use needs and improve the user's use experience.

[0059] Figure 2 The embodiment of the present application provides a front-end initial interface before a user uploads reference text materials. Figure 3 A front-end interface provided by an embodiment of the present application after a user uploads reference text materials, Figure 4 A front-end interface for selecting a reference field in a reference text material provided in an embodiment of the present application, Figure 5 A front-end interface for selecting a region to be identified in a reference text material provided in an embodiment of the present application. For example, see Figure 2-5 , the user clicks Figure 2 Click the "Click to select picture" button in the interface to upload the reference text material. In this embodiment, the reference text material is an award certificate. After uploading the reference text material, the user can Figure 3 Customize the template name in the interface. Users can select the reference field area and the difference to be identified according to the prompt box on the interface (such as Figure 4 and Figure 5 As shown), the template creation unit uses OCR technology to identify reference fields and keywords to be identified from the area selected by the user, and generates and publishes a graphic material template based on the recognition result.

[0060] On the basis of the above embodiments, further, the interface calling module 10 also includes a target graphic material template calling unit, wherein the target graphic material template calling unit is used to receive the target graphic material template ID input by the user before calling the OCR interface identification module 11 for each graphic material selected by the user, and request the OCR interface identification module 11 to call the target graphic material template corresponding to the target graphic material template ID, and save the called target graphic material template to a preset storage space, wherein the target graphic material template is used by the user to upload personal graphic materials.

[0061] After the user creates the graphic material template in the OCR interface recognition module 11, the user cannot directly use the graphic material template locally. If the user needs to use the graphic material template, the interface calling module 10 can be used to request the OCR interface recognition module 11 to call the graphic material template, and the graphic material template can be used only after the interface calling module 10 obtains the graphic material template. The specific request method is to input the graphic material template ID. It should be noted here that the graphic material template created by the user in the OCR interface recognition module 11, and its corresponding graphic material template ID will also be saved in the local preset storage space accordingly for the user to view and use.

[0062] On the basis of the above embodiments, further, the OCR interface recognition module 11 further includes a classifier training unit, wherein the classifier training unit is used to:

[0063] In response to a user's request to create an initial classifier, an initial classifier is created, wherein the initial classifier is configured with a classifier-related graphic and text material template selected by the user;

[0064] Based on the image and text material training set uploaded by the user and / or the classification keywords corresponding to the classifier-related image and text material template, the initial classifier is trained to obtain a pre-trained classifier, wherein the image and text material training set is a set of image and text materials with the same layout uploaded for each classifier-related image and text material template, and the pre-trained classifier is used to identify the image and text materials and classify the image and text materials into the corresponding image and text material template.

[0065] After the user creates a new graphic material template, it is necessary to train a pre-trained classifier corresponding to the new graphic material template so that the trained pre-trained classifier can identify graphic materials with the same layout as the new graphic material template and classify the graphic materials into the new graphic material template.

[0066] Figure 6 An interface for creating an initial classifier provided in an embodiment of the present application, Figure 7 An editing interface of an initial classifier provided in an embodiment of the present application, Figure 8 A front-end interface for selecting a classifier-related graphic data template for an initial classifier provided in an embodiment of the present application, Fig. 9 A front-end interface after the configuration of a classifier-related graphic and text data template is completed provided in an embodiment of the present application. Fig.10 A front-end interface for uploading a training set of graphic materials provided in an embodiment of the present application is exemplary, see Figure 6-10 , users can Figure 6 Enter the classifier name and function description in the interface, and click the "OK" button to generate the initial classifier. Figure 7 This is the editing interface of the initial classifier. Users can use this interface to configure the classifier-related graphic and text templates for the initial classifier. The specific operation is to click the "Add prefabricated template (prefabricated templates are system-provided templates)" or "Add custom template (custom templates are templates created by users)" button in this interface to jump to Figure 8 Interface, in which the user can select the appropriate graphic template according to actual needs. After the selection is completed, the user clicks the "OK" button to complete the operation of configuring the classifier-related graphic template for the initial classifier. Fig. 9 Click the "Start Training" button in the interface to jump to Fig.10 interface, you can upload the image and text training set (for the accuracy of training, the number of image and text training sets corresponding to each image and text template should generally be greater than 30). Fig. 9 Click the "Please Add" button in the classification keyword column on the interface to add classification keywords to the image and text template related to the classifier. After the image and text training set and / or classification keywords are added, the initial classifier can be trained to obtain a pre-trained classifier.

[0067] On the basis of the above embodiments, further, the pre-trained classifier includes three types: picture feature classifier, text feature classifier and graphic-text collaborative classifier. Among them, the picture feature classifier is trained based on the graphic-text material training set, the text feature classifier is trained based on the classification keywords corresponding to the graphic-text material template related to the classifier, and the graphic-text collaborative classifier is trained based on the classification keywords corresponding to the graphic-text material training set and the classifier related graphic-text material template.

[0068] This embodiment can train different types of pre-trained classifiers for different training sets, thereby improving the applicability of the system.

[0069] During the training process, if the user has uploaded a preset number of image and text data training sets for each classifier-related image and text data template, but has not filled in the classification keywords, the pre-trained classifier obtained through training is an image feature classifier; if the user has filled in the classification keywords for each classifier-related image and text data template, but some or all of the classifier-related image and text data templates have uploaded a preset number of image and text data training sets, the pre-trained classifier obtained through training is a text feature classifier; if the user has uploaded a preset number of image and text data training sets for each classifier-related image and text data template, and has filled in the classification keywords, the pre-trained classifier obtained through training is a image-text collaborative classifier.

[0070] On the basis of the above embodiments, further, the system also includes a preset evaluation rule determination module, wherein the preset evaluation rule determination module is used to determine the scoring items and the evaluation tags corresponding to each scoring item, and set a corresponding score and a keyword corresponding to the corresponding score for each evaluation tag, wherein the preset evaluation rules include one or more scoring items, and each scoring item corresponds to one or more evaluation tags.

[0071] Fig.11 A preset evaluation rule display interface provided in the embodiment of the present application, see Fig.11 The preset evaluation rules include scoring items, evaluation tags (tag names in the figure), scores and keywords. In this interface, users can maintain, add and delete various scoring items and evaluation tags. This embodiment can evaluate the evaluation object more fairly and impartially by refining the evaluation rules.

[0072] On the basis of the above embodiments, further, the evaluation module 12 also includes a group evaluation unit, wherein the group evaluation unit is used to perform weighted evaluation based on preset weights according to the evaluation results of each user in the corresponding group and the evaluation results given by management personnel for the corresponding group to obtain a group evaluation result.

[0073] This system can evaluate not only individual members within a group, but also groups, increasing the scope of application of the system. It should be noted that the system of this embodiment can also include an organizational structure management module, based on which relevant information of management personnel can be obtained.

[0074] On the basis of the above embodiments, the system further includes an evaluation ranking module and an information push module, wherein the evaluation ranking module is used to rank each user according to the user evaluation result of each user in the group and to rank each group according to the evaluation result of each group. Ranking individuals and groups enables individuals and groups to have a comprehensive understanding of their own situation, recognize their own shortcomings and advantages, so that they can work with clear goals in the future.

[0075] Preferably, the evaluation ranking module may further include a group dynamic scoring unit, wherein the group dynamic scoring unit is used to dynamically score each group according to a preset dynamic scoring rule and a ranking order of each group.

[0076] For example, the preset dynamic scoring rule can be that the first-ranked group gets a full score of 100 points, and then the score decreases by 10 points for each subsequent drop in rank, that is, the second-ranked group gets 90 points, the third-ranked group gets 80 points, and so on. The dynamic scoring method can accurately quantify the evaluation of the group. This mechanism ensures the fairness and objectivity of the evaluation, and effectively motivates each group to actively participate and strive to improve their own performance, thereby promoting the improvement of overall work efficiency and quality.

[0077] The information push module is used to push the ranking results obtained by the evaluation ranking module to the user. In this embodiment, the information push module can also push to-do items and assigned items to the user so that the user can be informed of unfinished tasks in a timely manner.

[0078] On the basis of the above embodiments, the system further includes an evaluation activity management module, wherein the evaluation activity management module is used to respond to the administrator's evaluation activity initiation request (including activity name, activity start time, and activity end time) to create the evaluation activity. The evaluation activity management module can also allow the administrator to maintain the evaluation activity, which can specifically include adding new activities, editing activity content, and activity status, etc. Exemplarily, the activity status can be ongoing or ended, etc. The evaluation activity management module can also allow the administrator to review or reject the information of the persons to be evaluated who participate in the evaluation activity. The evaluation activity management module can facilitate the administrator to manage the evaluation activities.

[0079] It should be noted that the terms used in this application are only for describing specific embodiments, rather than limiting the scope of this application. As shown in the specification and claims of this application, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0080] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0081] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and its core ideas of this application. The above is only the preferred implementation method of this application. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, ordinary technicians in this technical field can make several improvements, modifications or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the concept and technical solution of the invention to other occasions without improvement, should be regarded as the scope of protection of this application.

Claims

1. A recognition and evaluation system based on OCR technology, characterized in that: The system comprises: The interface calling module is used to call the OCR interface recognition module for each graphic material selected by the user, wherein the graphic material includes the personal graphic material uploaded by the user through the corresponding graphic material template, and / or the group graphic material uploaded by the administrator claimed by the user through the corresponding graphic material template; An OCR interface recognition module, used to determine a corresponding pre-trained classifier based on a corresponding graphic material template, and use the pre-trained classifier to perform OCR recognition on the corresponding graphic material to obtain a recognition result, wherein the corresponding graphic material template is a template applied when the corresponding graphic material is uploaded; An evaluation module, used to evaluate the corresponding graphic materials based on the recognition results and preset evaluation rules, and to evaluate the user based on each graphic material evaluation result to obtain a user evaluation result; A display module is used to display the user evaluation result to the user.

2. The evaluation system based on OCR technology according to claim 1, characterized in that: The recognition result includes keywords in the corresponding graphic material, and the preset evaluation rule includes a one-to-one correspondence between keywords and corresponding scores. Based on this, the evaluation module is specifically used to determine the score corresponding to the corresponding graphic material based on the keywords and the one-to-one correspondence between the keywords and the corresponding scores. After the score of each graphic material selected by the user is determined, the score of each graphic material is added to obtain the user's total score, and the user's total score is the user evaluation result.

3. The evaluation system according to claim 1 or 2, characterized in that: The OCR interface recognition module includes a template creation unit, wherein the template creation unit is used to: Receive reference and text materials uploaded by users; Using OCR technology to identify reference fields and areas to be identified in reference text materials, wherein the reference fields are text content with fixed positions in the reference text materials, and the areas to be identified are areas corresponding to keywords to be identified in the reference text materials; A graphic material template is created based on the reference field and the area to be identified, and a graphic material template ID is generated, wherein the graphic material template has the same format as the reference graphic material.

4. The system according to claim 3, characterized in that The interface calling module also includes a target graphic material template calling unit, wherein the target graphic material template calling unit is used to receive a target graphic material template ID input by a user before calling the OCR interface recognition module for each graphic material selected by the user, and request the OCR interface recognition module to call a target graphic material template corresponding to the target graphic material template ID, and save the called target graphic material template into a preset storage space, wherein the target graphic material template is used by the user to upload personal graphic materials.

5. The system according to claim 1 or 2, characterized in that: The OCR interface recognition module further includes a classifier training unit, wherein the classifier training unit is used to: In response to a user's request to create an initial classifier, an initial classifier is created, wherein the initial classifier is configured with a classifier-related graphic and text material template selected by the user; Based on the image and text material training set uploaded by the user and / or the classification keywords corresponding to the classifier-related image and text material template, the initial classifier is trained to obtain a pre-trained classifier, wherein the image and text material training set is a set of image and text materials with the same layout uploaded for each classifier-related image and text material template, and the pre-trained classifier is used to identify the image and text materials and classify the image and text materials into the corresponding image and text material template.

6. The system according to claim 5, characterized in that The pre-trained classifiers include three types: image feature classifiers, text feature classifiers, and graphic-text collaborative classifiers. The image feature classifiers are trained based on a graphic-text data training set, the text feature classifiers are trained based on classification keywords corresponding to graphic-text data templates related to the classifiers, and the graphic-text collaborative classifiers are trained based on classification keywords corresponding to graphic-text data training sets and graphic-text data templates related to the classifiers.

7. The system according to claim 2, characterized in that The system also includes a preset evaluation rule determination module, wherein the preset evaluation rule determination module is used to determine the scoring items and the evaluation tags corresponding to each scoring item, and set a corresponding score and a keyword corresponding to the corresponding score for each evaluation tag, wherein the preset evaluation rules include one or more scoring items, and each scoring item corresponds to one or more evaluation tags.

8. The system according to claim 1 or 2, characterized in that: The evaluation module further includes a group evaluation unit, wherein the group evaluation unit is used to perform weighted evaluation based on preset weights according to the evaluation results of each user in the corresponding group and the evaluation results given by management personnel for the corresponding group to obtain a group evaluation result.

9. The system according to claim 8, characterized in that The system further comprises an evaluation ranking module and an information push module, wherein the evaluation ranking module is used to rank each user according to the user evaluation result of each user in the group and to rank each group according to the evaluation result of each group; The information push module is used to push the ranking results obtained by the evaluation ranking module to the user.

10. The system according to claim 9, characterized in that The evaluation ranking module also includes a group dynamic scoring unit, wherein the group dynamic scoring unit is used to dynamically score each group according to a preset dynamic scoring rule and a ranking order of each group.

Images