Three-proofing paint coating quality detection system and method

Abstract

The invention discloses a three-proofing paint coating quality detection system and method and relates to the technical field of safety coatings. The system comprises a storage module, a database establishment module, a prediction model module, a to-be-detected image acquisition module, and a to-be-detected image analysis module; wherein the storage module is used for storing data; the database establishment module is used for establishing a to-be-tested batch component surface material and coating database; the prediction model module is used for establishing a conformal coating thickness prediction model according to the data in the database, and substituting the processed data of the to-be-detected coating surface image into the conformal coating thickness prediction model to predict the coating quality; the to-be-detected image acquisition module is used for acquiring a to-be-detected coating surface image; and the to-be-detected image analysis module is used for processing the to-be-detected coating surface image to obtain processed data of the to-be-detected coating surface image. According to the method, the image processing calculation amount is small, and three-proofing paint coating quality problems of parts which are not coated with the three-proofing paint, parts which cannot be effectively protected by the three-proofing paint and parts which influence the performance of components can be detected at the same time.

Description

technical field [0001] The invention relates to the technical field of safety coatings, in particular to a coating quality inspection system and method for conformal paint. Background technique [0002] Conformal paint is a specially formulated coating used to protect circuit boards and related equipment from environmental erosion. Conformal paint has good high and low temperature resistance; after curing, it forms a transparent protective film, which has excellent insulation, moisture-proof, leakage-proof, shock-proof, dust-proof, corrosion-proof, anti-aging, corona resistance and other properties. [0003] In the usage scenario, the surface of the three anti-paint coating is often the first to be affected by the environment, and problems such as corrosion, softening, deformation, and mildew will occur, which will lead to circuit board circuit failures, which will affect the life of the equipment and normal production activities in the slightest. its safety incident. This...

AI Technical Summary

Problems solved by technology

[0003] In the usage scenario, the surface of the three anti-paint coating is often the first to be affected by the environment, and problems such as corrosion, softening, deformation, and mildew will occur, which will lead to circuit board circuit failures, which will affect the life of the equipment and normal production activities in the slightest. its security incident

This makes it very important to detect the coating quality of the three-proof paint. However, the three-proof paint is a kind of transparent coating with a special formula, and direct human eye detection is not feasible. In production, artificial observation under ultraviolet light is often used. The method will first cause harm to the health of the testing personnel, and secondly, the human error is large, the accuracy is low, and the efficiency is low

Therefore, in large-scale production, manual inspection cannot be applied on a large scale. Therefore, people have developed a computer system for inspecting the quality of three anti-paint coatings.

[0004] However, the current detection system often adopts the method of image color comparison for analysis. Among them, graying the image and then comparing is a common method, but the tenderness of this method lies in: the coating thickness is thinner and the area Uncoated areas are indistinguishable due to similar grayscale values ​​in the grayscale image

At present, there is a technology to solve this problem by dividing the RGB color image into channels, using the color index method to calculate the grayscale image, and then thresholding the grayscale image to solve this problem, but this will increase the amount of calculation and the threshold standard is not clear , it is impossible to avoid the influence of the color difference of the material components themselves, and there are still some problems of low detection accuracy

In addition, due to the flow of the conformal paint on the surface of the device after coating and before drying, not only will the surface of some components not be coated, but also the thickness of the coating on the surface of some components will be too large, which will affect the performance of the device or use, traditional methods are unable to detect and discover these problems

Images