Neural network deep learning-based cloth defect detection method

Abstract

The invention relates to a neural network deep learning-based cloth defect detection method. The method includes the following steps that: (1) high-speed line scanning imaging is performed; (2) an improved BP neural network cloth defect detection algorithm is adopted to accurately detect cloth defects; and (3) a convolutional neural network deep learning-based cloth defect classification algorithm is adopted to automatically select characteristic information of defect diversity, and non-linear and systematic processing and classification are performed. According to the method of the invention, algorithms such as image correction, splicing and de-noising are realized in an imaging system through a GPU, so that high-speed and high-quality image acquisition can be realized; the improved BP neural network cloth defect detection algorithm is adopted to detect and eliminate interference factors such as dust, dirt, cotton balls and folds; and the convolutional neural network deep learning-based cloth defect classification algorithm is adopted to monitor various kinds of detects in real time, and the classification algorithm can automatically select the characteristic information of defect diversity and carry out non-linear and systematic processing and classification.

Description

technical field [0001] The invention relates to the technical field of defect detection, in particular to a cloth defect detection method based on neural network deep learning. Background technique [0002] In the process of textile production, surface defects of cloth are the key factors affecting the quality of cloth. For a long time, cloth inspection is generally done manually. Manual inspection relies on the experience and proficiency of cloth inspection personnel, and the evaluation standards are unstable and inconsistent, so false detections and missed inspections often occur, and skilled cloth inspection personnel can only find about 70% of the defects. In addition, cloth defect detection is a heavy and tasteless physical labor for workers, and it greatly damages the eyesight of cloth inspection workers. Therefore, the use of machine vision instead of manual automatic identification of cloth defects has become an inevitable trend in the development of the textile in...

AI Technical Summary

Problems solved by technology

Due to the different raw materials of gray cloth in different weaving factories, the densities of the gray cloth produced are different, and it is difficult to have quantitative imaging standards in the gray cloth quality inspection imaging process, including light source intensity, camera aperture size, object distance and other parameters will be the debugging process of the developed testing equipment Parameters that are very difficult to tune in

[0005] 2. There are many types of defects, which are difficult to distinguish

There are more than 20 kinds of gray fabric defects, and the similarity between different defects is high, even experienced cloth inspectors are often difficult to distinguish

[0006] 3. There are too many factors affecting defect imaging

The production conditions of most textile mills are relatively harsh, and dirt or cotton balls are stuck to the surface of the gray cloth to varying degrees. These substances will largely affect the reliability of the cloth defect detection algorithm

Similarly, some gray fabrics will form wrinkles due to time storage and the vibration of the machine tool will also affect the reliability of the algorithm

Images