Electric power tool detection method based on Yolo V5 model
A detection method and technology of tools, applied in the direction of neural learning methods, biological neural network models, instruments, etc., can solve the problems of inaccurate identification and detection of electrical tools, and achieve the effect of improving accuracy
Pending Publication Date: 2022-01-18
STATE GRID ZHEJIANG ELECTRIC POWER CO LTD QUZHOU POWER SUPPLY CO +1
View PDF1 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The invention solves the problem of inaccurate identification and detection of electrical tools in the prior art, and proposes a method for detecting electrical tools based on the YoloV5 model. The invention realizes the identification and classification of electrical tools based on the Yolo V5 algorithm
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
[0031] This embodiment proposes a method for detecting electrical appliances based on the Yolo V5 model, refer to figure 1 , including the following steps:
[0032] S1, collecting and labeling sample image data of electrical appliances; S1 specifically includes the following steps: collecting sample image data of electrical appliances, labeling the sample image data with an image labeling tool labelimg, and obtaining labeled sample image data. Use the labeling tool LableImg to label the target object in the sample image data. The labels obtained include seven categories: RAM, NoRAM, RAMs, CPU, NoCPU, CPUFan, and RAID.
[0033] S2, performing feature enhancement on the labeled sample image data to obtain a COCO dataset; S2 specifically includes the following steps:
[0034] S201, perform feature enhancement on the marked sample image data to obtain enhanced image data, feature enhancement includes adding noise, changing brightness and changing chroma; adding noise includes: in...
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 electric power tool detection method based on a Yolo V5 model. The method comprises the following steps: collecting sample image data of an electric power tool and marking the sample image data; performing feature enhancement on the labeled sample image data to obtain a COCO data set; constructing a Yolo V5 model, wherein the loss function of the Yolo V5 model is a Focal Loss loss function; using a COCO data set to train the Yolo V5 model to obtain a target parameter and a target weight of electric power tool detection, and using the target parameter and the target weight as the parameter and the weight of the trained Yolo V5 model; and using the trained Yolo V5 model to detect the standardized image of the electric power tool, and outputting the category of the corresponding electric power tool. According to the invention, identification and classification of the electric power tools are realized based on the Yolo V5 algorithm. And the Yolo V5 algorithm performs data enhancement on the electric power tool model, and performs transfer learning on a data set through preprocessing, so that a database is expanded, and the accuracy of tool identification is further ensured.
Description
technical field [0001] The invention relates to the technical field of electric tool detection, in particular to a method for detecting electric power tools based on the Yolo V5 model. Background technique [0002] In the process of electric power work, all kinds of tools are commonly used, and the violations in the process of power work often involve some tools and tools. In order to better supervise the violations of operators, this paper focuses on identifying power tools in the process The purpose is to lay the foundation for violation detection and analysis. There are many electrical appliances, and with the continuous optimization and updating of electrical hardware equipment, there are more and more types of electrical appliances, which brings certain difficulties to identification and detection. Traditional machine learning related algorithms cannot better distinguish the complicated Electric tools. At present, there are few studies on the identification and classi...
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): G06T3/40G06T5/00G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06T3/4038G06N3/08G06T2207/20084G06T2207/20088G06T2207/30108G06N3/045G06F18/241G06T5/00
Inventor 钱斌张小龙王绍强司海涛周玮俞阳章华田仲旭何秀明杨凯
Owner STATE GRID ZHEJIANG ELECTRIC POWER CO LTD QUZHOU POWER SUPPLY CO