Scratch test - Reliability Test

Date：2021-11-04 17:45:51Views：1222

With the rapid progress of industry and society, China's manufacturing industry has developed rapidly in science and technology and occupied a leading position. In metal software, the detection technology of surface scratch has attracted much attention. Surface scratch detection technology was developed in the 1950s. It mainly adopts manual visual detection, and the inspector can directly observe the defects with the naked eye. The traditional surface scratch detection technology has many disadvantages; For example, the manual detection accuracy is low and workers are prone to fatigue. The manual detection needs to work in the environment of high temperature, noise, dust and vibration, which has caused great harm to the body and psychology.

With the development trend of industry 4.0 and the application of artificial intelligence visual inspection technology in various industries, more and more production-oriented enterprises use machine vision inspection equipment to control the quality of their products.Scratch detection is generally divided into two steps: first, determine whether there are scratches on the surface of the product; Then, after determining the scratch information, the scratches are extracted and detected, and the surface scratches are usually divided into the following three categories.

The first type of scratch is easy to identify from the appearance alone, and the change of its gray level is also obvious compared with the surrounding environment. You can choose a smaller threshold to mark the scratch directly

For the second type of scratch, the change of some gray values is not obvious, and the gray level of the overall image is also relatively average. The full level of the scratch is relatively small, only a few pixels, and the gray level is only slightly lower than the surrounding image, which is difficult to distinguish. The original image can be filtered by the mean value to obtain a smoother image, which is subtracted from the original image value. When the absolute value of the difference is greater than the threshold, it will be set as the target, and all targets will be marked, its area will be calculated, the targets with too small area will be removed, and the rest will be marked as scratches.

For the third type of scratch, the gray level difference of each part is large, and the shape is usually long strip. If a fixed threshold segmentation is adopted on an image, the marked defect part will be smaller than the actual part. Because the scratches of such images are narrow and long, the extension of defects will be missed by relying solely on gray detection. For this kind of image, the combination of double threshold and defect shape feature is selected according to its characteristics.

Due to the diversity of images in industrial detection, each image should be processed by analyzing and comprehensively considering various means to achieve the effect. Generally speaking, the gray value of the scratch part is darker than the surrounding normal part, that is, the gray value of the scratch part is too small; Moreover, most of them are on smooth surfaces, so the gray change of the whole image is very uniform and lacks texture features. Therefore, the scratch detection generally uses the method of statistical gray feature or threshold segmentation to mark the scratch part.

Scratch test is one of the standard methods used to test the bonding strength between surface film and matrix material. A material with a hardness larger than that of the surface film is made into a sharp tip to scratch on the surface of the sample or workpiece, and load is added to the scratching head until the membrane and the substrate are disengaged. The load is the critical load at this time, and the bonding strength between the surface film and the substrate can be calculated.

Application scope of scratch test

Scratch test is mainly used to test and evaluate the shear, scratch, planing, scraping and engraving resistance of anodized materials, rigid organic materials, powder coatings, soft metals, plastics, glass and coating adhesives.

Method standard

GB / t17657-1999, iso4586-2, jisk6902, federal series standards, terrazzo90322-9e-1, uni9428furniture, as / nzsas2924 2. Astmc217, cen series standards, din53799, din68861-4