Chemical fiber drifting impurity identification method for mobile robot

Abstract

The invention belongs to the technical field of intelligent identification, and particularly relates to a chemical fiber drifting impurity identification method for a mobile robot. The method comprises steps of effectively removing white or high-brightness background interference through an image acquisition mode of monochromatic laser and an optical filter. combining a deep learning target detection method and an iterative position correction method, so that accurate identification and positioning of the highlight chemical fiber tow are realized; taking areas among the highlight chemical fiber tows as possible impurity drifting areas, and identifying impurity drifting chemical fibers based on an image processing method; using the distance relation between the drifting impurity chemical fibers and the highlight chemical fiber tows as a drifting impurity recognition judgment basis, so that the severity degree of drifting impurity occurrence can be quantified. The problem that other light exists in a close range of lump-shaped light can be effectively solved.

Description

technical field [0001] The invention belongs to the field of intelligent identification, in particular to a method for identifying floating impurities of chemical fibers for a mobile robot. Background technique [0002] The patent application publication number is CN 110992343 A "A Robot-Based Visual Inspection Method, Storage Medium and Electronic Equipment for Floating Fibers". Carry out supplementary light on the tow, and carry out visual inspection on the photo-imaged image with group-like highlights. If there are other bright lights outside the group-like bright light, it is considered that drifting has occurred. The disadvantages of the prior art are: [0003] The use of illuminating lamps for supplementary light can obtain images with highlighted tow areas, but cannot filter white or highlighted background objects. Such objects are easily identified as other bright lights, resulting in misidentification; [0004] The existing patents believe that there are other bri...

AI Technical Summary

Problems solved by technology

[0005] During the mobile shooting process of the robot, there is no guarantee that each station can obtain high-quality clusters of bright light, and other bright lights belonging to clusters of bright light are easily formed around the surrounding area. The existence of other bright lights outside the cluster of bright lights cannot be directly used as a judgment condition for floating

Images