Surface crack detection method of winding inductor under complex background

Winding inductors are widely used in power supply circuits. The quality of the inductor is directly related to the performance and safety of power equipment. The inspection of surface defects is an important part of inductor production. The detection of surface cracks in inductors is the key and difficult point in the detection of surface defects of inductors. Black spots and watermarks on the surface of inductors will interfere with the detection effect. This paper studies crack detection methods in complex backgrounds. Firstly, it studies the interference information suppression method based on frequency domain transformation to reduce the interference of the black dots and watermarks; then, studies the slight crack recognition method based on the second derivative of the image to highlight the make slight cracks in the complex background; finally, studies the target screening method based on crack morphology analysis, to improve the accuracy of crack detection according to the morphological difference between cracks and interference items. Experiments show that the method proposed in this paper can effectively improve the accuracy of crack detection under complex background. In the meantime, the method proposed in this paper can be extended to defect detection in related complex backgrounds.

I. Introduction

Quality inspection is a key procedure in industrial production and it is the last guarantee for product quality. Quality testing can be divided into two types: destructive testing and non-destructive testing. Among them, destructive testing is suitable for random inspection, while non-destructive testing is suitable for full inspection. With the increasing requirements for product quality in the industrial field, full product inspection will be the trend. Vision-based non-destructive inspection of defects is a commonly used method in the industrial field. Traditional manual visual inspection has problems such as strong subjectivity and low consistency, which prompted researchers to develop automatic inspection systems based on machine vision to improve the accuracy and consistency of defect detection. However, surface defect detection is not the same as general size detection, especially surface defects under complex backgrounds are easily disturbed, resulting in low consistency of detection results. This paper studies the surface defect detection of winding inductors. Winding inductors are suitable for power supply circuits. This product is widely used in chargers, cameras, audio-visual equipment, etc. The quality of winding inductors directly affects the safety and life of electrical products.