Floor defect detection method based on deep learning

Abstract

The invention discloses a floor defect detection method based on deep learning. The method comprises the following steps: 1) collecting a high-resolution image of a floor by using an industrial camerato establish a floor detection image library; 2) manually labeling the defect area of each defect image in the floor detection image library in the step 1) to form a floor detection image label library; 3) establishing a semantic segmentation model and an image classification model based on deep learning; 4) expanding the floor detection image library acquired in the step 1) by adopting a data enhancement technology; 5) carrying out deep learning training on a floor defect semantic segmentation model and a floor defect classification model on the image library expanded in the step 4); and 6)collecting the floor surface image and performing defect detection on the floor surface based on the trained floor defect classification model. According to the floor defect detection method, the precision of manual screening is achieved under the condition that manual interference is not needed, the production cost can be reduced, and the intellectualization of the floor industry is promoted.

Description

Technical field: [0001] The invention relates to the technical field of floor defect detection, in particular to a deep learning-based floor defect detection method. Background technique: [0002] Defect detection of flooring is a key link in the quality control of its production process. The existing defect detection method is mainly completed by artificial naked eye screening. The existing method has the following disadvantages: slow detection speed, detection results are greatly affected by the experience of quality inspection workers, uncontrollable detection accuracy, high cost of quality inspection workers, etc. . With the emergence and development of computer technology, artificial intelligence and other science and technology, object surface defect detection technology based on machine vision technology has emerged as the times require, which has greatly improved the detection effect of object surface defects and increased the detection rate of object surface defect...

AI Technical Summary

Problems solved by technology

The existing defect detection method is mainly completed by artificial naked eye screening. The existing method has the following disadvantages: slow detection speed, detection results are greatly affected by the experience of quality inspection workers, uncontrollable detection accuracy, high cost of quality inspection workers, etc.

[0003] Detection technology based on traditional image processing and recognition technology still needs to manually extract features, and the design of these features requires a lot of prior knowledge and experience

In the flooring industry, the surface of the product is dominated by irregular textures and there are many types. The performance of traditional image algorithms is not satisfactory in this case.

Images