Image processing method, device and equipment and computer storage medium

Abstract

The embodiment of the invention provides an image processing method, device AND equipment and a computer storage medium, relates to the field of image detection, and is used for improving the accuracy of positioning detection of a graphic code in an image in a complex scene. The method comprises the following steps: acquiring a to-be-detected image; inputting the to-be-detected image into a pre-trained key point detection model, determining the reference confidence degree of an object in the to-be-detected image and multiple groups of reference angular point coordinate information of the object, wherein the key point detection model is obtained through training according to a training sample set, the training sample set comprises a plurality of sample graphic codes and label information of each sample graphic code, and the label information comprises coordinate information of a plurality of angular points and coordinate information of a central point of the sample graphic code; under the condition that the reference confidence coefficient is greater than a preset confidence coefficient threshold value, determining that the object is a to-be-positioned graphic code; and screening a plurality of reference frames determined by the plurality of groups of reference angular point coordinate information, and determining frame position information of the graphic code to be positioned.

Description

technical field [0001] The present application belongs to the field of image detection, and in particular relates to an image processing method, device, equipment and computer storage medium. Background technique [0002] As the most common coding form in daily life, graphic code is widely used in scenarios such as mobile payment and information acquisition. However, in the graphic code image, due to printing problems, such as printing strips, underprinting, etc., as well as the small proportion of the graphic code image, blur, dirt, uneven illumination, geometric uneven deformation, etc., the imaging of the graphic code is abnormal. , so it is impossible to accurately locate the graphic code. [0003] In the existing graphic code positioning and recognition technology, the main positioning method is to scan the image to judge the boundary through the row-by-row and column-by-point scanning algorithm, and obtain the light and dark width flow according to the gradient transf...

AI Technical Summary

Problems solved by technology

The above method can achieve a high accuracy recognition rate for perfect graphic codes, that is, images with good printing, no abnormal imaging, and a single background, and the real-time performance can also meet the requirements. However, for graphic code images in complex scenes, it can reduce generalization. At the same time, the speed of positioning position detection graphics is slow and inaccurate, resulting in low accurate recognition rate of graphic codes and slow recognition speed, which cannot meet the requirements

Images