Development Of Defect Detection Equipment For Aluminum Alloy Wheel Hub Based On Machine Vision

A lightweight motorcycle wheel hub is generally made from an aluminum alloy matrix and a steel spline sleeve using composite casting.Owing to the difference in heat transfer properties between the two metals,it is easy to cause excessive joint clearance.However,this affects the safety of the wheel hub.Therefore,it is necessary to detect defective products accurately.The traditional detection method involves coating the seam with ink and judging the quality of the product by applying a load(swing or torsion)manually and observing whether the ink breaks.Jinfei Group,the client of the project,has an annual output of nearly 200 million aluminum alloy motorcycle wheels.However,traditional detection methods are inefficient and labor-intensive.Therefore,it is urgent to develop defect detection equipment based on machine vision and automatic loading.The main contents are as follows:(1)Aiming at the problem that the uneven brightness of aluminum alloy wheel surface image affects the accuracy of edge recognition,an edge detection algorithm based on guided filter Retinex and improved canny is proposed to accurately locate the seam circle of composite casting,and compared with the traditional image enhancement and edge detection algorithm;(2)Aiming at the problem that the ink surface defect features are easily affected by alcohol volatilization and similar crack area before loading,which leads to low crack detection accuracy,a defect detection method of aluminum alloy wheel hub composite casting joint based on dual channel feature fusion was proposed;(3)The defect detection equipment of aluminum alloy wheel hub based on machine vision is designed and developed from the aspects of mechanism,electric control and software.The effectiveness and practicability of the algorithm proposed in this paper are verified on the equipment.On the basis of the above research content,this paper puts forward two innovation points:(1)An image edge detection method using Guided filtering Retinex and adaptive canny is proposed.The new algorithm greatly suppresses the edge error recognition caused by uneven brightness,and effectively improves the image edge contrast,and the edge connectivity is better than the traditional algorithm.(2)A defect detection method using dual channel feature fusion for composite casting joint of aluminum alloy wheel hub was proposed.The test results of 1000 groups qualified wheel image and defect image show that the recognition accuracy of the proposed algorithm is 98.8%.