Target detecting method and device and computer readable storage medium

Abstract

The invention discloses a target detecting method and device and a computer readable storage medium. On the basis of fusion of infrared light and visible light images, contour information of a shape template and a to-be-detected object is compared and matched to determine the contour information meeting a matching similarity condition as a target information set. In the prior art, an object that is not a to-be-detected target but has a shape similar to that of the to-be-detected target occurs frequently in an actual scene; for example, when a to-be-detected image is a traffic road image and ato-be-detected target is pedestrian, a streetlight or a tree like the pedestrian may occur in the traffic road image, so that the target detection result has the low accuracy. However, with the targetdetecting method disclosed by the invention, the technical problem can be solved.

Description

technical field [0001] The present invention relates to image detection, in particular to an object detection method, equipment and computer-readable storage medium. Background technique [0002] Target detection technology, also called target extraction, is an image segmentation based on the geometric and statistical characteristics of the target. It combines target segmentation and recognition into one technology, which can identify the target to be detected in the image to be detected. Surveillance, public security, automatic driving, biomedicine and other fields have a wide range of applications. More generally, when the image to be detected is a traffic road image and the target to be detected is a pedestrian, based on the target detection technology, the traffic road image can be Pedestrian information is identified. [0003] In the actual implementation process, the inventor found that in the actual scene, objects similar to the shape of the target to be detected but...

AI Technical Summary

Problems solved by technology

[0004] The object detection method provided by the present invention solves the problem that objects that are similar to the target to be detected but not the target to be detected often appear in the actual scene. For example, when the image to be detected is a certain traffic road image, the target to be detected When it is a pedestrian, there may be street lights, trees, etc. similar to pedestrians on the traffic road image, resulting in inaccurate target detection results.

Images