Glass defect detection method based on frequency domain and space combination based on segmentation network

Abstract

The present application discloses a glass defect detection method based on a segmentation network combined with frequency domain and space, which includes the following steps: constructing an image data set containing glass defect images, and performing grayscale processing on the objects in the image data set to obtain a predetermined Train the data model; collect images of glass products, and then process the collected images through frequency domain combined with spatial analysis to obtain the frequency characteristics of the images; deconvolve the processed images to extract high-dimensional features in the images as feature points; A convolutional neural network is used to segment and train the feature points in the image to be detected to obtain a test data set; the pre-trained data model is used to train and classify the test data set to obtain a detection result. The application can effectively improve the efficiency of glass detection, shorten the detection time and improve the yield rate of products, and effectively reduce the input of labor costs.

Description

technical field [0001] The present application relates to the technical field of visual recognition processing, and in particular to a glass defect detection method based on a segmentation network combined with frequency domain and space. Background technique [0002] With the continuous development and progress of industrial technology, users and manufacturers have higher and higher requirements for product quality. In addition to meeting the performance requirements, they must also have a good appearance, that is, a good surface quality. However, in the process of manufacturing products, surface defects are often unavoidable, such as bubbles, plaques, cracks, pitted surfaces, inclusions and scratches in glass products. Manual detection is a traditional detection method for product surface defects. This method has low sampling rate, low accuracy, poor real-time performance, low efficiency, high labor intensity, and is greatly affected by manual experience and subjective fac...

AI Technical Summary

Problems solved by technology

Manual detection is a traditional detection method for product surface defects. This method has low sampling rate, low accuracy, poor real-time performance, low efficiency, high labor intensity, and is greatly affected by manual experience and subjective factors. The detection method based on machine vision can To a large extent, it overcomes the above disadvantages, and at the same time can effectively improve the detection efficiency and reduce labor costs. However, there is still a lack of targeted detection methods to realize machine identification of glass defects.

Images