Unlock instant, AI-driven research and patent intelligence for your innovation.
A deep learning-based flaw detection system and method
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A defect detection and deep learning technology, applied in optical testing defects/defects, measuring devices, scientific instruments, etc., can solve the problems of weak generality and function expansion of the underlying algorithm, and many open parameters of image processing software, etc., and achieve maintenance costs. The effect of low, strong compatibility and less labor input
Active Publication Date: 2021-06-11
TZTEK TECH
View PDF3 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
[0004] 2. The underlying algorithms of image processing software are weak in versatility and function expansion. For new products and new needs of customers, personnel are required to redevelop
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 more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0040] Such as figure 1 As shown, a deep learning-based flaw detection system consistent with this embodiment includes:
[0041] The belt transmission device 5 is used to transmit the tested product (siliconwafer);
[0042] A line-scanning camera 1 and a lens 2 connected thereto are used to scan the tested product 6 on the belt conveyor 5, and send the collected product surface image to the PC host 7;
[0043] Preferably, a light source 3 is also included, which is arranged on the side of the measurement center line 4, and is used for supplementing light for the camera to ensure the accuracy of image collection.
[0044] The PC host 7 is provided with image processing software, which detects and processes defects through the image processing software, and simultaneously displays and uploads the processing results to the cloud 8;
[0046] Preferably, the image processing software includes the following modules:
[0047] The ...
Embodiment 2
[0062] Such as figure 2 As shown, on the basis of the embodiment, this embodiment provides a method for detecting defects based on deep learning, including the following steps:
[0063] S1: the line-scanning camera 1 scans the tested product 6 on the belt conveyor 5, and sends the collected surface image of the product to the image processing software;
[0064] S2: Use image processing software to detect and process flaws, and display and upload the processing results to the cloud 8 at the same time;
[0066] Preferably, the step S2 specifically includes:
[0067] S21: Preprocessing the picture, the preprocessing includes but not limited to grayscale transformation of the image and cropping of the image;
[0068] S22: Perform prediction through convolutional neural network degree images, and obtain prediction results;
[0069] S23: Process the prediction result and obtain the processed picture;
[0070] S24: Display the process...
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 defect detection system based on deep learning, comprising: a belt transmission device, a line-scanning camera and a lens connected thereto, a PC host and a cloud; the PC host is provided with image processingsoftware, and through the image processingsoftware Perform flaw detection and processing, and display and upload the processing results to the cloud for big data analysis. The invention also discloses a defect detection method based on deep learning. The invention introduces the AI algorithm of deep learning to identify and extract the features of defects in the picture with high efficiency and high recognition rate. Less labor input, low maintenance cost for image processingsoftwaredetection performance. Strong compatibility, in the face of product upgrades, image processing software algorithms do not need to be developed separately, just collect more product samples for learning and training, and can quickly meet the flaw detection application of new products.
Description
technical field [0001] The present invention relates to the technical field of flaw detection, in particular to a flaw detection system and method based on deep learning. Background technique [0002] In the industrial production process, defect detection is an important step in many product quality inspection links. The defect detection device uses the image processing software to identify and process the product surface image collected by the industrial camera to find out the defects, and at the same time effectively classify and follow-up the defects. Traditional image processing software has several problems: [0003] 1. Image processing software has many open parameters, and it takes a lot of effort to debug to achieve better detection performance. [0004] 2. The underlying algorithms of image processing software are weak in versatility and function expansion. For new products and new needs of customers, personnel are required to redevelop them. [0005] Therefore, ...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More