Self-adaption sorting system and method based on computer vision and machine learning

Abstract

The invention provides a self-adaption sorting system and method based on computer vision and machine learning. The system is composed of an article classification decision making module, an article information estimating module, a base, a camera, a variable-force suction cup, a flexible mechanical gripper and touch sensors. The system is simple in structure, low in manufacturing cost, stable andreliable in working, good in universality and interchangeability and suitable for multiple working environments. By arranging the variable-force suction cup and the flexible mechanical gripper, the self-adaption sorting system is suitable for fruit self-adaption sorting grading and other multiple production occasions and has the high application value.

Description

technical field [0001] The invention belongs to the field of robot applications, in particular to an adaptive sorting system and method using computer vision and machine learning. Background technique [0002] Nowadays, with the continuous development of robot technology, robots have played a pivotal role in the industrial production field of human society. The large-scale application of robots has improved the production automation of factories and improved the manufacturing efficiency of production lines. [0003] However, in some special production fields, such as fruit sorting, it is often necessary for the gripper of the robot to have certain flexible gripping characteristics, so as to achieve stable gripping without destroying the quality of the fruit. [0004] In particular, in the field of cherry sorting, it is usually necessary to classify the quality of cherries according to their size, weight, color, etc., and to achieve stable picking and sorting. [0005] Howev...

AI Technical Summary

Problems solved by technology

Its purpose is to use computer vision recognition algorithms to sort fruits in real time through information such as fruit type, size, and maturity in the process of fruit sorting, so as to improve the accuracy of the fruit sorting process and mainly solve the current problems. At this stage, users cannot achieve accurate and stable sorting according to the type and quality of fruits and other information.

Images