Method and system for abnormal work detection of logistics transfer robot based on image processing

Abstract

The invention relates to the field of image data processing, in particular to a method and system for detecting abnormal work of a logistics transfer robot based on image processing; acquiring video image data corresponding to a logistics transfer robot, the video image data includes several frames of video images; extracting the corresponding video images of each frame Each connected domain is calculated, and the representative value of each connected domain is calculated; based on the representative value, each connected domain on two adjacent frames of video images is matched to obtain multiple matching pairs; according to the video image data, two adjacent frames of video images are calculated According to the image entropy of the difference image, the video image data is divided into multiple categories; based on the matching pairs, the discrete coefficients corresponding to various types of video images are calculated; the maximum similarity matching distance corresponding to various types of video images is calculated, and then the The frame transition stability corresponding to various video images; according to the frame transition stability, it is judged that the logistics transfer robot is abnormal. The invention can accurately judge whether the logistics transfer robot is abnormal.

Description

technical field [0001] The invention relates to the field of image data processing, in particular to a method and system for detecting abnormal work of a logistics transfer robot based on image processing. Background technique [0002] With the rapid development of e-commerce, the emergence of logistics transfer robots has replaced the traditional manual handling of goods, effectively reducing the labor intensity of workers and improving operating efficiency; however, due to the complexity of the working area environment, there may be Some adhesive fine particles, screws and other substances, if these substances adhere to the wheels of the logistics transfer robot, after a period of time, these substances will cause the wheels to slip, block, block and other abnormal conditions; abnormality of the wheels (Slipping, blocking, blocking) will cause the goods to be overturned on the ground, which will not only affect the work efficiency but also cause certain property losses; fo...

AI Technical Summary

Problems solved by technology

[0002] With the rapid development of e-commerce, the emergence of logistics transfer robots has replaced the traditional manual handling of goods, effectively reducing the labor intensity of workers and improving operational efficiency; however, due to the complexity of the working area environment, there may be Adhesive fine particles, screws and other substances, if these substances adhere to the wheels of the logistics transfer robot, after a period of time, these substances will cause abnormal conditions such as slipping, blocking, and blocking of the wheels; abnormalities of the wheels (Slipping, blocking, blocking) will cause the goods being transported to overturn on the ground, which will not only affect the work efficiency but also cause certain property losses; for example: when the logistics transfer robot transports heavy items, if the wheels slip, the logistics The transfer robot will not stop in time due to inertia and cause great losses; therefore, it is very necessary to detect whether the logistics transfer robot is abnormal in time and quickly, so as to avoid unnecessary losses

[0003] The existing detection method is to judge whether the robot is abnormal by using the uniformity of speed through the detection of different frames of video images, but the actual working environment of the robot is extremely complicated, and it is impossible to accurately detect whether the logistics transfer robot is abnormal by only using the uniformity of speed

Images