Text detection method, device, medium and electronic device

Abstract

The invention relates to a text detection method, device, medium and electronic equipment. The method is applied to the Mask RCNN model. In the model, the region proposal network RPN is replaced by a segmentation network. The method includes: obtaining the first image of the original image containing the text Feature map; obtain the first segmentation result of the first feature map through the segmentation network, the first segmentation result includes multiple polygonal frames surrounding the text area; perform post-processing on the multiple polygonal frames to obtain multiple corrected rectangular frames; extract multiple corrected The second feature map of the rectangular frame, the third feature map is obtained by multiplying multiple second feature maps with the corresponding local feature maps in the first feature map; the detection result and the second segmentation result of the third feature map are obtained through this model , the detection result includes a plurality of candidate rectangular boxes, and the second segmentation result includes a feature map of the surrounding text area in each candidate rectangular box. The embodiments of the present invention can more accurately detect distorted, tilted and other texts in the recognized image.

Description

technical field [0001] The embodiments of the present disclosure relate to the technical field of image processing, and in particular, to a text detection method, a text detection device, and a computer-readable storage medium and electronic equipment for implementing the text detection method. Background technique [0002] At present, in the face of image data, how to efficiently and automatically obtain useful text information from images has become a research hotspot. [0003] In related technologies, the image text detection scheme adopts a target detection two-stage network such as Faster RCNN (Faster Region Convolutional Neural Networks), which can improve the efficiency of automatic text information detection to a certain extent. [0004] However, the current image text detection scheme fails to consider the text tilt and distortion of the text in the image, so the accuracy and efficiency of image text detection and recognition in the text tilt, distortion and other s...

AI Technical Summary

Problems solved by technology

[0004] However, the current image text detection scheme fails to consider the text tilt and distortion of the text in the image, so the accuracy and efficiency of image text detection and recognition in the text tilt, distortion and other scenarios are low.

Images