Target area detection method, device, equipment and storage medium

Abstract

The present application discloses a target area detection method, device, electronic equipment and storage medium. The method includes: extracting features from the image to be detected to obtain target features, processing the target features to obtain a three-dimensional feature map, and according to each pixel in the three-dimensional feature map The number of channels of the point is the probability that each pixel in the image to be detected is the target area; according to the probability that each pixel in the image to be detected is the target area, the candidate area in the image to be detected is determined; the candidate area is analyzed and processed by connected domain, Get the target area of ​​the image to be detected. By adopting the above detection method of the present application, the target area in the image to be detected can be automatically and quickly detected.

Description

technical field [0001] The present application relates to the field of computer vision technology in artificial intelligence technology, and more specifically, to a target area detection method, device, equipment and storage medium. Background technique [0002] With the advancement of science and technology, the safety detection of the appearance, outer packaging or casing of objects is becoming more and more extensive. For example, for defective items (such as dirty, broken or abnormally marked items), Usually, it is necessary to repair the defective position of the item in time, so as to avoid affecting the use of the corresponding item by the user. [0003] In the prior art, it is usually necessary to collect the image of the object first to check whether the object is flawed, and the work of collecting the image of the object is completed at a fixed position. At present, a camera and a control box are usually installed on a fixed position, and the camera is controlled ...

AI Technical Summary

Problems solved by technology

Since the manual visual inspection largely depends on the proficiency of the inspection engineer, the inspection process is too inefficient, the inspection time is long, and there are cases of missing defects.

Images