Text detection method, system, device and storage medium
A text detection and feature map technology, applied in character and pattern recognition, image analysis, image enhancement, etc., can solve the problem that the background noise in the text area cannot be well eliminated, and achieve the effect of suppressing texture noise
Active Publication Date: 2021-04-16
合肥综合性国家科学中心人工智能研究院
View PDF6 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The embodiment of the present application provides a text detection method, system, device, and storage medium, aiming to solve the problem that the text detection method in the prior art only expresses the characteristics of the text area in two-dimensional space, and cannot well eliminate the background of the text area. noise problem
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0056] In order to better understand the above-mentioned technical solutions, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0057] The invention provides a character detection device. Such as figure 1 as shown, figure 1 It is a schematic structural diagram of the hardware operating environment involved in the solution of the embodiment of the present invention.
[0058] It should be noted, figure 1 That is, it is a structural schematic diagram of the hardware operating environment...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a text detection method, system, device and storage medium. The method includes using a residual network to perform feature extraction to obtain a feature image; performing contextual attention processing on the feature image to generate a semantically enhanced feature map; using a semantically enhanced feature map to The pyramid network is processed to generate a shared feature map; the shared feature map is processed by a region proposal network to obtain a prediction suggestion box; based on the prediction suggestion box, the boundary segmentation branch is used to obtain candidate boundary points; pixel-level embedding parallel to the boundary segmentation branch is used The branch filters the candidate boundary points to obtain the final boundary points to locate the text area. It solves the problem that the text detection method in the prior art only expresses the characteristics of the text area in two-dimensional space, and cannot well eliminate the background noise of the text area. It can not only detect text of any shape, but also suppress the background similar to text texture. Noise for more precise text localization in arbitrarily shaped scenes.
Description
technical field [0001] The present invention relates to the technical field of text detection, in particular to a text detection method, system, equipment and storage medium. Background technique [0002] Scene text detection technology aims to locate the text area in the scene image, which plays an important role in computer vision tasks such as automatic driving, trademark recognition, and blind guidance. [0003] In recent years, text representation based on boundary points has become more popular, mainly because boundary points can fit the shape of curved text more flexibly, and the detection task of curved text is very useful. However, areas similar to text textures in the background are easily misjudged by the model as text areas, so it is inaccurate to directly use the segmentation or regression results of boundary points for subsequent text localization. The current text detection method only expresses the features of the text area in two-dimensional space. Due to t...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/32G06K9/62G06N3/04G06T7/11G06T7/12
CPCG06T7/12G06T7/11G06T2207/20016G06T2207/20084G06T2207/20081G06V20/62G06N3/045G06F18/253
Inventor 张勇东行梦婷谢洪涛
Owner 合肥综合性国家科学中心人工智能研究院