A Novel Text Recognition Method Based on Counting Focusing Model
A text recognition and model technology, applied in the field of optical character recognition, can solve the problem of complex module design for focus weight calculation, and achieve the effects of simplified design and low code implementation requirements
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] The overall framework of the counting focus model is the same as the previous focus model, and consists of two parts: the encoder (decoder) based on convolutional neural network (CNN) extracts high-level features from the input image to obtain a high-level feature map (featuremap); The decoder of the long short-term memory network (LSTM) and the attention mechanism (Attention Mechanism) decodes the characters from left to right in sequence from the high-level feature map. Specific as figure 1 shown.
[0035] The encoder uses a common CNN, and the process of extracting high-level features to obtain a high-level feature map has no improvement compared with the prior art. The main improvement of the recognition method provided by the present invention lies in the calculation process of the decoder, such as figure 2 As shown, the calculation process of the decoder is as follows:
[0036] S21. Segment the high-level feature map from left to right along the horizontal dime...
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, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com