Court judgment document-oriented multi-scale learning character recognition method and system
A character recognition and multi-scale technology, applied in the field of optical character recognition, can solve problems such as unsatisfactory requirements, achieve the effect of removing seal traces, improving recognition performance, and effective recognition
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0032] Example 1
[0033] As mentioned in the background art, documents with seals such as court judgment documents, contracts, etc., are all affected by the occlusion of seal traces during character recognition. Therefore, in this embodiment, the court judgment document is taken as an example. figure 1 As shown, a text recognition method for multi-scale learning of court judgment documents is provided, including:
[0034] S1: Obtain the image of the court judgment document to be recognized, and extract the seal area;
[0035] S2: Taking the maximum difference of the target RGB value as the objective function, taking the cyclic consistency loss as the constraint condition, building a seal generation network model based on the confrontation network, and using the trained seal generation network model to convert the RGB value of the seal traces in the seal area, and Delete the seal marks in the converted seal area;
[0036] S3: Perform feature extraction on the image of the court judgme...
Example Embodiment
[0109] Example 2
[0110] This embodiment provides a multi-scale learning character recognition system for court judgment documents, including:
[0111] The image acquisition module is used to acquire the image of the court judgment document to be recognized and extract the seal area;
[0112] The trace deletion module is used to take the maximum difference in the target RGB value as the objective function, and the cyclic consistency loss as the constraint condition, build a seal generation network model based on the confrontation network, and use the trained seal generation network model to calculate the RGB value of the seal trace in the seal area Conversion, and delete the seal traces in the converted seal area;
[0113] The text recognition module is used to extract the features of the court judgment document image from which the seal traces are deleted, perform global target detection and local detail detection on the obtained feature maps, and merge the mask and progressive mask...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap