Safety helmet wearing detection method
A detection method and safety helmet technology, which is applied in the field of image recognition, can solve the problems of high missed detection rate of small targets and difficulty in detecting overlapping targets, and achieve the effect of optimizing non-maximum value suppression and accurate results
Pending Publication Date: 2022-01-11
ELECTRIC POWER RES INST OF STATE GRID ZHEJIANG ELECTRIC POWER COMAPNY
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The present invention aims at the shortcomings of the existing detection algorithm that it is difficult to detect overlapping targets or the rate of missed detection of small targets is high, and provides a safety helmet wearing detection method to improve the existing detection method
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 1
[0071] see Figure 1 to Figure 6 , a safety helmet wearing detection method according to Embodiment 1 of the present invention, said safety helmet wearing detection method comprising:
[0072] Step S1, data set production, including: collecting image samples and labeling image samples;
[0073] Step S2, model building, including:
[0074] Step S2.1, model selection, including: selecting improved YOLOv3 as the judgment model of whether to wear a safety helmet in the construction scene, so as to take into account the detection speed and detection accuracy. The improved YOLOv3 model uses new conv2d instead of the standard convolution layer, and the adopted formula is :
[0075] Y=x*w1+b1=x1*w2+b2
[0076] Among them, Y represents the convolution output, x represents the variable, w represents the weight, and b represents the bias;
[0077] Step S2.2, data set preprocessing, including: normalizing the obtained standard information of the hard hat wearing data set, and converti...
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 belongs to the technical field of image recognition, and particularly relates to a safety helmet wearing detection method. In order to overcome the defects of high overlapped target detection difficulty or high small target omission ratio in the existing detection algorithm, the invention adopts the following technical scheme: the safety helmet wearing detection method comprises the following steps: making a data set; carrying out model establishment: carrying out model selection, wherein the improved YOLOv3 model uses new conv2d to replace a standard convolutional layer, and the adopted formula is Y = x * w1 + b1 = x1 * w2 + b2; preprocessing the data set; initializing an anchor frame through a clustering algorithm, and carrying out model training; carrying out model testing, including feature extraction and bounding box prediction, and non-maximum suppression of a predicted bounding box; and carrying out target detection. The safety helmet wearing detection method has the beneficial effects that: network parameters are reduced through a matrix decomposition mode, and the detection rate is improved; and non-maximum suppression is optimized, so that the detection result is more accurate.
Description
technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a safety helmet wearing detection method. Background technique [0002] As an effective protective tool, safety helmets are widely used in various production and operation sites. Wearing safety helmets is one of the effective ways to prevent or reduce injuries of construction site workers, but some workers lack safety awareness, and behaviors of not wearing safety helmets often occur. Due to reasons such as insufficient supervision and audit, accidents caused by not wearing safety helmets occur from time to time. In order to ensure the personal safety of workers, real-time detection of helmets is very necessary. [0003] In recent years, with the rapid development of artificial intelligence, object detection technology has become a hot research direction, and a large number of excellent object detection algorithms have been born. For example, YOLO (You On...
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 IPC IPC(8): G06V20/52G06V10/25G06V10/44G06V10/762G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/23213G06F18/214
Inventor 韩睿刘黎赵泓闫云凤王文浩姜雄伟蒋鹏李特温典
Owner ELECTRIC POWER RES INST OF STATE GRID ZHEJIANG ELECTRIC POWER COMAPNY