China is a manufacturing powerhouse, producing a massive amount of industrial products daily. Users and manufacturers alike are increasingly demanding higher product quality standards, requiring not only performance but also a pleasant appearance, i.e., good surface quality. However, surface defects are often unavoidable during the manufacturing process. Different products have different definitions and types of surface defects. Generally, surface defects are areas of uneven physical or chemical properties on a product's surface, such as scratches, spots, and holes on metal surfaces; color differences and indentations on paper surfaces; and inclusions, breaks, and stains on non-metallic surfaces like glass.
Surface defects not only affect the aesthetics and comfort of products, but also generally have an adverse impact on their performance. Therefore, manufacturers attach great importance to the detection of surface defects in their products in order to detect them in a timely manner, thereby effectively controlling product quality. They can also analyze certain problems in the production process based on the detection results, thereby eliminating or reducing the generation of defective products, preventing potential trade disputes, and safeguarding the company's reputation.
Manual inspection is the traditional method for detecting surface defects in products. This method has low sampling rate, low accuracy, poor real-time performance, low efficiency, high labor intensity, and is greatly affected by human experience and subjective factors. However, machine vision -based inspection methods can largely overcome the above drawbacks.
The Robotics Industries Association (RIA) defines machine vision as: "Machine vision is a device that automatically receives and processes images of a real object through optical devices and non-contact sensors to obtain the required information or to control the movement of a robot."
Machine vision is a non-contact, non-destructive automated inspection technology. It is an effective means to achieve equipment automation, intelligence, and precision control, and has outstanding advantages such as safety and reliability, wide spectral response range, ability to work for extended periods in harsh environments, and high production efficiency. A machine vision inspection system acquires surface images of a product through an appropriate light source and image sensor (CCD camera), extracts feature information from the image using corresponding image processing algorithms, and then performs operations such as locating, identifying, and classifying surface defects, as well as statistical analysis, storage, and retrieval based on the feature information.
The basic components of a visual surface defect detection system mainly include an image acquisition module, an image processing module, an image analysis module, a data management module, and a human-machine interface module.
In machine vision surface defect detection systems, image processing and analysis algorithms are crucial. The typical workflow includes image preprocessing, target region segmentation, feature extraction and selection, and defect identification and classification. Numerous algorithms have emerged for each processing step, each with its own advantages, disadvantages, and applicable scope. Improving the accuracy, execution efficiency, real-time performance, and robustness of these algorithms has been a continuous research focus.
Machine vision surface inspection is quite complex, involving many disciplines and theories. Machine vision is a simulation of human vision, but the mechanisms of human vision are still not fully understood. Although every normal person is a "visual expert," it is difficult to express their visual process using a computer. Therefore, the construction of machine vision inspection systems needs to be further improved by studying the biological vision mechanism, so that inspection can be further developed towards automation and intelligence.
Machine vision surface inspection is quite complex, involving many disciplines and theories. Machine vision is a simulation of human vision, but the mechanisms of human vision are still not fully understood. Although every normal person is a "visual expert," it is difficult to express their visual process using a computer. Therefore, the construction of machine vision inspection systems needs to be further improved by studying the biological vision mechanism, so that inspection can be further developed towards automation and intelligence.
