Visual identification device and method for robot positioning

Abstract

The invention discloses a visual identification device for robot positioning. The visual identification device comprises an image collection module, an image processing module in which a standard image is pre-stored, and a servo control system. The visual identification method comprises the following steps of 1) image collection; 2) image enhancement; 3) image filtering in which an image obtained in the step 2) is filtered by a median filtering algorithm; 4) edge detection; 5) image segmentation; and 6) compared identification. According to the visual identification device and method, the image processing method is improved to improve the anti-interference performance and accuracy of the visual identification system; algorithms related to image processing are relatively mature in the prior art, and can be referenced directly, and the system accuracy is ensured; and via a series of pretreatment, interference in the collection process is reduced, images can be identified via compared identification, and the accuracy is improved.

Description

technical field [0001] The invention relates to a visual recognition device, in particular to a visual recognition device for robot positioning and a recognition method thereof, belonging to the technical field of numerical control machine tools. Background technique [0002] Intelligent robots are the product of the development of industrial robots from non-intelligent to intelligent, from low-intelligence to high-intelligence. Its intelligent feature is that it has the function of coordinating with the external world, objects, environment and people. In this regard, vision, followed by robot vision, is the embodiment of simulating human vision on robots, so robot vision is considered to be the most important sensory ability of robots. Machine vision is generally defined as automatically acquiring and analyzing images to obtain data describing a scene or controlling certain actions. Its essence is to enable computers to have the ability to recognize surrounding environment...

AI Technical Summary

Problems solved by technology

In traditional industrial robot applications, the robot only operates according to a fixed program, and the initial pose and end pose of the object are specified in advance; when the working environment of the robot changes or the target object changes or even the environmental conditions are unknown, with The robot control system with traditional sensors has many limitations, and even cannot accurately complete the predetermined tasks. The introduction of visual devices to the robot, on the one hand, allows the robot to imitate the human visual ability to perceive the external environment, and second, it can achieve non-contact Pose measurement, thirdly, it improves the intelligence of the robot and the ability of the robot to adapt and self-learn to a certain extent; among them, in the process of visual recognition, its image processing method is the most critical, and the image processing method in the prior art is relatively complicated , and cannot well form a recognition system

Images