Electrical insulating glove real-time detection method and computer readable medium

A technology for electrical insulation and real-time detection, which is applied to the computer-readable medium storing the detection program of electrical insulation gloves, and in the field of real-time detection of electrical insulation gloves, can solve the problems of small head movement space and large hand movement space, and achieve improved The effect of extracting accuracy, realizing pedestrian detection, and improving recognition accuracy

Pending Publication Date: 2022-02-08
STATE GRID HENAN ELECTRIC ZHOUKOU POWER SUPPLY +1
View PDF1 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The aforementioned technical solution is used to detect the helmets of construction workers. Compared with the detection of safety helmets, the detection of insulating gloves has the following difficulties: the head movement space is small, but the hand movement space is large

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: a kind of electrical insulation gloves real-time detection method, comprises the following steps:

[0027] Obtain the frame pattern output by the camera;

[0028] Use a deep learning algorithm to detect whether there is a pedestrian pattern in the frame image, and if a pedestrian pattern is detected in the frame image, use a marker frame to mark the pedestrian pattern; in this embodiment, the deep learning algorithm is SSD (Single Shot Multi-Box Detector) algorithm , the feature extraction layer of the SSD algorithm includes the Inception structure.

[0029]Use the posture estimation algorithm to extract the wrist joint coordinates and elbow joint coordinates corresponding to the pedestrian pattern in the marked frame, and extract the hand in the marked frame according to the wrist joint coordinates, elbow joint coordinates, and hand extraction graphics of the pedestrian pattern in the marked frame In this embodiment, the posture estimation algorithm uses...

Embodiment 2

[0031] Embodiment 2: a kind of real-time detection method for electrical insulating gloves, comprising the following steps:

[0032] Obtain the frame pattern output by the camera;

[0033] Use a deep learning algorithm to detect whether there is a pedestrian pattern in the frame image, and if a pedestrian pattern is detected in the frame image, use a marker frame to mark the pedestrian pattern; in this embodiment, the deep learning algorithm uses tensorflow to build a convolutional neural network structure, input Layer, convolutional layer, excitation layer, pooling layer, fully connected layer and output layer, where the input layer receives image data, which is passed in as a tensor tensor, and the number of convolutional layers is 24. Generally, the convolutional layer A better detection effect can be obtained if the number is greater than or equal to 5. Each convolutional layer uses padding and pooling algorithms to scan and generate a localreceptive fields with a convolu...

Embodiment 3

[0036] Embodiment 3: A computer-readable medium storing a testing program for electrical insulating gloves, which can be run to implement the real-time testing method for electrical insulating gloves in Embodiment 1 or Embodiment 2.

[0037] In this embodiment, the testing program for electrical insulating gloves includes: an input module, an identification module and an output module.

[0038] The input module is used for receiving the image to be recognized.

[0039] The recognition module includes a deep learning algorithm for detecting pedestrian patterns, a pose estimation algorithm for extracting joint coordinates, a target area pixel extraction algorithm for extracting hand graphics, an HSV space conversion algorithm, and a comparator. When in use, the deep learning algorithm detects whether there is a pedestrian pattern in the image to be recognized, and if a pedestrian pattern is detected in the frame image, the mark frame is used to mark the pedestrian pattern in the...

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

No PUM Login to view more

Abstract

The invention relates to an electrician insulating glove real-time detection method and a computer readable medium storing an electrician insulating glove detection program. The method comprises the following steps: detecting whether a pedestrian pattern exists in a frame image or not by employing a deep learning algorithm, and marking the pedestrian pattern by employing a marking box if the pedestrian pattern is detected in the frame image; extracting wrist joint coordinates and elbow joint coordinates corresponding to the pedestrian pattern in the mark frame by using a posture estimation algorithm, and extracting a hand region image in the mark frame according to the wrist joint coordinates, the elbow joint coordinates and a hand extraction graph of the pedestrian pattern in the mark frame; and comparing the gray value of each pixel in a hand region image with the gray threshold of an insulating glove in HSV space to obtain the ratio of pixels with gray values located within the gray threshold range of the insulating glove to the pixels of the hand region image, comparing the ratio with an insulating glove identification ratio threshold, and outputting an insulating glove identification result. According to the invention, the detection efficiency of the insulating gloves of the electrician in the video can be improved.

Description

technical field [0001] The invention relates to the technical field of glove recognition in images, in particular to a real-time detection method for electrical insulating gloves and a computer-readable medium storing a detection program for electrical insulating gloves. Background technique [0002] Object recognition refers to finding a specified object in an image or video and framing the specific location of the object. There are many object recognition algorithms, and the simplest one is to judge only by the color of the object. However, the color is too sensitive to the lighting conditions and the stability is too poor. Later, HAAR and Adaboost were used for face recognition, HOG features for pedestrian detection, SIFT or SURF features for feature matching for target recognition, and so on. However, the recognition rates of these algorithms are easily disturbed by the environment, and the environmental factors of power operations are just complex and changeable. How ...

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
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06V20/52G06V10/25G06V10/44G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/214
Inventor 马学民李梦冉史晨昱李岩赵鹤董建刚李亚莉
Owner STATE GRID HENAN ELECTRIC ZHOUKOU POWER SUPPLY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap