Ceramic bathroom appearance quality detection system and detection method

Abstract

The invention is suitable for the technical field of visual inspection, and provides a ceramic bathroom appearance quality detection system, which comprises a workbench, and also comprises: a transportation system connected with the workbench and used for horizontally conveying a sample to be detected; the detection system is connected with the workbench, and the detection system can generate an external dimension image of the to-be-detected sample, so that the appearance of the to-be-detected sample is comprehensively detected; the conveying system comprises a belt conveying pair, the belt conveying pair is horizontally arranged along the side end face of the workbench, the workbench is fixedly connected with a camera obscura, and the belt conveying pair penetrates through a channel opening of the camera obscura. The detection system comprises a visual camera, a fixing frame and a six-axis industrial robot; a detection table is arranged in the camera obscura, the number of the visual cameras is four, and two of the visual cameras are fixedly connected with the six-axis industrial robot through fixing frames; and the other two vision cameras are respectively fixed on the two sides of the detection table and are arranged below the belt conveying pair in parallel.

Description

technical field [0001] The invention belongs to the technical field of visual inspection, and in particular relates to a ceramic bathroom appearance quality inspection system and inspection method. Background technique [0002] In recent years, some ceramic sanitary ware enterprises have basically realized production automation, but in the actual production line, due to uncontrollable factors such as product transportation and manufacturing process, the products inevitably have defects such as surface cracks and non-standard appearance dimensions, and the surface of ceramic sanitary ware products is smooth It is easy to reflect light, and the contrast of the defect part is not obvious and difficult to identify under ordinary ambient light sources. [0003] At present, the product quality inspection process is still mainly based on traditional manual inspection, relying on human eye observation and auxiliary touch for inspection and evaluation. Not only is the inspection spee...

AI Technical Summary

Problems solved by technology

[0002] In recent years, some ceramic sanitary ware enterprises have basically realized production automation, but in the actual production line, due to uncontrollable factors such as product transportation and manufacturing process, the products inevitably have defects such as surface cracks and non-standard appearance dimensions, and the surface of ceramic sanitary ware products is smooth It is easy to reflect light, and the contrast of the defect part is not obvious and difficult to identify under ordinary ambient light sources

[0003] At present, the product quality inspection process is still dominated by traditional manual inspection, relying on human eye observation and auxiliary touch for inspection and evaluation. Not only is the inspection speed of a single product slow, the inspection standards are vague, and the subjectivity is strong, making it difficult to reasonably classify products. , and it is difficult to trace the source of problems and feed them back to the production process, and there is no guarantee for product quality control

In addition, the size and weight of ceramic sanitary products are relatively large, and the inspectors have high work intensity. Touching the workpiece to be tested is easy to cause damage, which restricts the production efficiency and product quality of the factory.

The combination of the existing support vector machine and traditional image processing technology is suitable for defect detection of small samples, but the quality of the detection results largely depends on the quality of the extracted features, which requires high professional comprehensive quality of R & D personnel; and Although the ceramic flaw detection method based on the neural network is relatively high in accuracy, it requires a large amount of data sets to train the corresponding network, which consumes too much time and is too expensive

Images