Metal curved surface defect analysis system and method based on deep learning
A technology of deep learning and defect analysis, which is applied in the field of metal surface defect analysis system based on deep learning, can solve the problems of low recognition accuracy, large imaging difference, complex processing technology, etc.
Pending Publication Date: 2021-06-04
广州信邦智能装备股份有限公司
View PDF0 Cites 6 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Disadvantages of the existing technology: In the detection and identification of aero-engine blade defects, due to the variety of blade types and complex processing technology, the imaging of surface defects is quite different, and the image processing algorithm based on machine vision has low recognition accuracy when performing defect detection. And other issues
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 Construction
[0039] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0040] Such as figure 1 The metal surface defect analysis system based on deep learning shown includes the main interface, image preprocessing module, image training library module, convolutional neural network algorithm module and output module;
[0041] The function of the main interface module is to: collect and visualize images in real time; set camera parameters, including exposure time, acquisition frame rate and other parameter settings; visualize detection results, including visualization of result images and statistical results;
[0042] The role of the image preprocessing module is: the preprocessing algorithm, including image filtering, image contrast enhancement, etc., is used to reduce image noise and improve the quality of defect images;
[0043] The role of the image training library module is: basic training for deep ...
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 discloses a metal curved surface defect analysis system and method based on deep learning. The system comprises a main interface, an image preprocessing module, an image training library module, a convolutional neural network algorithm module and an output module. The system detects and recognizes a to-be-recognized metal curved surface image of a blade through an intelligent detection technology of deep learning, continuously trains the detection process of the system on various defect compositions of the metal curved surface by adopting the deep learning technology, and recognizes, classifies, analyzes and segments various defects of the metal curved surface such as the surface of the blade. And finally, intelligent high-precision recognition and detection of various defined defects of the metal curved surface are achieved, and a meaningful method is provided for detection of various defect types of the metal curved surface.
Description
technical field [0001] The invention relates to the field of high-gloss metal surface defect detection, in particular to a deep learning-based metal surface defect analysis system and method. Background technique [0002] With the development of science and technology, precision optical systems have higher and higher requirements for surface defects of high-gloss metal parts. Domestic detection of surface defects of high-gloss metal parts is mainly manual visual inspection, or through some common visual algorithms on the market. , Long-term detection by the human eye is prone to visual fatigue, ordinary algorithms are difficult to cover all defects, there are missed detections, and the efficiency is low, resulting in different judgment standards and difficult to quantitatively judge. At present, there are many optical inspection software based on machine vision technology, but they are all aimed at technical solutions with flat surfaces and uncomplicated background textures....
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): G06T7/00G06T7/11G06T7/136G06T7/194G06T7/73G06T5/00G06N3/04G06N3/08G01N21/88
CPCG06T7/0004G06T7/11G06T7/136G06T7/194G06T7/73G06N3/08G01N21/8851G01N2021/8854G01N2021/8887G06T2207/20081G06T2207/20084G06T2207/10004G06T2207/20024G06T2207/30136G06N3/045G06T5/70
Inventor 习勇张家业徐洪浩
Owner 广州信邦智能装备股份有限公司