Industrial vision detection method based on deep learning
A deep learning and visual detection technology, applied in neural learning methods, image data processing, image enhancement, etc., can solve the problems of low detection success rate and long learning and training time, and achieve the effect of reducing error rate and accurate detection
Pending Publication Date: 2021-02-09
苏州砺行信息科技有限公司
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Aiming at the deficiencies of the prior art, the present invention provides an industrial visual detection method based on deep learning, which solves the problems of long learning and training time and low detection success rate of deep learning industrial visual detection
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0035] like figure 1 As shown, the embodiment of the present invention provides a deep learning-based industrial vision detection method, including the following steps:
[0036] (1) Conduct deep learning training
[0037] A. Divide industrial lighting into three light intensities: bright, dim, and common light intensity, and then illuminate and take pictures of defective industrial parts. The photos taken are divided into three groups: bright group defect pictures and dark group defects Pictures and common light intensity group defect pictures, and then store the pictures in the sample storage;
[0038] B. Shuffle all the pictures in the sample storage, and then build the picture into a deep learning model through a certain program;
[0039] C. Establish a deep learning model;
[0040] I. Combining the training process of denoising autoencoder (DAE) and shrinkage autoencoder (CAE+H) with regularization, train a K-layer DCAE+H stacked deep neural network through the followin...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides an industrial vision detection method based on deep learning, and relates to the field of industrial vision detection. The industrial vision detection method based on deep learning comprises the following steps: (1) carrying out training of deep learning; A, photographing industrial parts in groups according to three illumination intensities; B, disordering the pictures in the sample storage and establishing a deep learning model; (2) performing image acquisition; a, putting industrial parts in part positions set by a detection system and illuminating the parts; b, collecting pictures of the industrial parts; (3) performing deep learning to detect the industrial parts; detecting and identifying the part picture shot by the industrial camera and the picture in the deep learning model; and (4) sorting and counting by a terminal. Through a reasonable detection method, three industrial part feature pictures are used for training deep learning, fixed illumination intensity and part positions are set for shooting of part pictures, the detection rate of defective parts is increased, and the method is worthy of vigorous popularization.
Description
technical field [0001] The invention relates to the technical field of industrial visual inspection, in particular to an industrial visual inspection method based on deep learning. Background technique [0002] Deep learning is the internal law and representation level of learning. The information obtained during the learning process is of great help to the interpretation of data such as text, images and sounds. Its ultimate goal is to enable machines to have the same ability to analyze and learn as humans , can recognize data such as text, images and sounds. Deep learning is a complex machine learning algorithm. The effect achieved in speech and image recognition far exceeds the previous related technologies. To complete the functions of identification, measurement, positioning, etc., the general visual inspection system is composed of cameras, lenses, and light sources, which can replace manual inspections of barcode characters, cracks, packaging, surface layer integrity, ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06K9/62G06N3/04G06N3/08
CPCG06T7/001G06N3/08G06T2207/10004G06T2207/30108G06T2207/30242G06N3/045G06F18/24147
Inventor 刘文军石景文杨梦铎陈晨陈涛
Owner 苏州砺行信息科技有限公司