Visual detection method for micro-defects on surface of magnetic tile

Abstract

The invention provides a visual detection method for micro-defects on the surface of a magnetic tile. The method comprises the steps of reading a magnetic tile image, detecting magnetic tile image defects, obtaining a defect region K of the magnetic tile image, and judging whether or not the area of the defect region K is greater than a first set value; detecting the magnetic tile image defects, obtaining a second-class defect effect map of the magnetic tile image, and judging whether or not the length of the second-class defect effect map is greater than a second set value; detecting the magnetic tile image defects, obtaining a circular degree of a communication domain pixel of an edge detection image Q' of the magnetic tile image, and judging whether or not the circular degree of the communication domain pixel of the edge detection image Q' is greater than a third set value, wherein whether or not the defects of the magnetic tile belong to three types can be judged through the threejudgement processes. The visual detection method for the micro-defect on the surface of the magnetic tile is high in adaptability to illumination changes and magnetic tile type changes; various different types of defects of the magnetic tile can all be detected; compared with magnetic tile defect images obtained by means of traditional methods, the magnetic tile defect image obtained by means of the method is clearer and more accurate.

Description

technical field [0001] The invention relates to a method for detecting micro-defects on the surface of magnetic tiles, in particular to a method for visually detecting micro-defects on the surface of magnetic tiles. Background technique [0002] In the manufacturing process of magnetic tiles, due to the influence of factors such as raw material composition, equipment usage, processing technology and worker operation, some processing defects will inevitably appear on the surface, such as cracks, chipping, indentation, trachoma and Grinding and so on. The existence of defects will have a great impact on the magnetic properties and service life of the magnetic tiles. If the defective magnetic tiles are used in wind power generation, new energy vehicles, aerospace and other fields, it will cause huge safety hazards, and even directly cause disastrous consequences. Therefore, the quality inspection of the magnetic tiles must be carried out before leaving the factory, and the ma...

AI Technical Summary

Problems solved by technology

The manual detection method mainly relies on the quality inspection personnel who have undergone certain training to judge the quality of magnetic tiles through observation, measuring tool measurement and touch according to their own experience. There are many limitations and shortcomings: (1) Due to the differences in the workers' own abilities , there is a certain chance, and the evaluation standards are different; (2) The detection speed is very slow, and it is difficult to meet the current production speed and requirements

[0005] (1) The surface image of the magnetic tile has the characteristics of non-flat shape, dark color, low contrast, etc., and the models are diverse, resulting in poor adaptability of the algorithm;

[0006] (2) The molding, sintering and grinding processes cause the surface of the magnetic tile to have complex textures, which affect the extraction of defects and their features, resulting in a high rate of misjudgment;

[0007] (3) Some scratches and cracks on the surface of small magnetic tiles are very small, and the width after imaging may be less than one pixel, which is easily disturbed and difficult to detect;

[0008] (4) The algorithm lacks versatility, and most methods are aimed at specific defects, but cannot detect all types or several types of defects at the same time

Moreover, most of them take a long time to process and are not suitable for online use;

[0009] (5) The light source is complex, and different surfaces and defects require different lighting forms

[0010] The visual detection of micro-defects on the surface of small magnetic tiles often requires that various types of defects and tiny defects on the surface can be detected more accurately under high-speed production conditions. Accuracy and robustness requirements, but in computer vision inspection, image analysis and processing algorithms are often difficult to balance efficiency, robustness, and accuracy

Images