Data enhancement method for insulator defect detection
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 insulator technology, which is applied in image enhancement, image data processing, instruments, etc., can solve the problems that cannot be applied multiple times, the deviation of real data is large, and the effect of model training is affected, so as to achieve rich feature diversity and context information , the effect of expanding the sample size
Pending Publication Date: 2021-12-31
的卢技术有限公司
View PDF0 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:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0006] Except for image flipping and affine transformation, other data enhancement methods do not change much the target features and context information. Although image flipping and affine transformation can achieve this goal, they cannot be applied multiple times, because after exceeding a certain standard, other The amplified data deviates greatly from the real data in the actual situation, which will affect the model training effect
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
[0031] Such as figure 1 As shown, a data augmentation method for insulator defect detection includes the following steps:
[0032] (1) Image cropping and splicing;
[0033] (2) roi extraction-acquire target features, superimpose to the picture after affine transformation;
[0034] (3) Comprehensively apply conventional data enhancement methods such as image flipping, contrast change, and random cropping to the image processed in steps (1) and (2).
[0035] Described step (1) is specifically:
[0036] (1.1) If figure 2 As shown in , the insulators are installed horizontally, divided into several parts along the horizontal direction, and then exchanged with each other, splicing them into a picture of the original size to enrich its context information;
[0037] (1.2) If image 3 As shown in , the insulators are installed vertically, divided into several parts along the vertical direction, and then exchanged with each other, splicing them into a picture of the original size...
Embodiment 2
[0043] A computer storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned data enhancement method for insulator defect detection is realized.
Embodiment 3
[0045] A computer device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, the above-mentioned data enhancement for insulator defect detection is realized method.
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 data enhancement method for insulator defect detection. The method comprises the following steps: (1) cutting and splicing pictures; (2) roi extraction: obtaining target features, and superposing the target features to a picture after affine transformation; and (3) comprehensively applying conventional data enhancement methods such as picture flipping, contrast change and cutting to the pictures processed in the steps (1) and (2). According to the invention, the context information of the picture is enriched through picture cutting and splicing; through feature extraction, affine transformation and random position superposition, feature diversity and context information are enriched; through combination of hyper-parameter setting and multiple methods, the sample size can be greatly expanded; according to the method, the target can be kept close to the real situation while the picture is amplified.
Description
technical field [0001] The invention relates to insulators, in particular to a data enhancement method for insulator defect detection. Background technique [0002] Insulators are important components in transmission lines, and the quality of insulators directly affects the safety of transmission lines. [0003] In recent years, the State Grid has been vigorously promoting intelligent power inspections, that is, combining drone aerial pictures and deep learning image processing technology to automatically locate potential safety hazards in power transmission lines. Among them, insulator defect detection is a key concern in intelligent inspection. Common insulator defects include insulator self-explosion, insulator contamination, and insulator shed damage. [0004] The realization of intelligent patrol inspection relies on a large number of image processing technologies, especially deep learning, for example, target detection and location of self-explosion. The application ...
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