Oil production operation site smoker recognition processing method and system

A job site, recognition and processing technology, applied in neural learning methods, character and pattern recognition, instruments, etc., to achieve the effect of easy optimization, solving scale change problems, and simple structure

Active Publication Date: 2021-04-30
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF17 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide an anchor-free and multi-scale feature adaptive fusion method and system for identifying and processing smokers at oil production sites, so as to solve at least one technical problem in the above-mentioned background technology

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
  • Oil production operation site smoker recognition processing method and system
  • Oil production operation site smoker recognition processing method and system
  • Oil production operation site smoker recognition processing method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] Such as figure 1 As shown, Embodiment 1 of the present invention provides a smoking detection and identification processing system for personnel at an oil recovery operation site, the system includes:

[0057] The collection module is used to call the camera to take pictures of the oil extraction operation site to obtain environmental photos;

[0058] The recognition module is used to analyze the environmental photos by using a recognition model to determine whether there are cigarette butts in the environmental photos; wherein, the recognition model is obtained by using multiple sets of data through machine learning training; the multiple sets of data include The first type of data and the second type of data; each set of data in the first type of data includes: a photo containing cigarette butts and a label identifying that the photo contains cigarette butts; each set of data in the second type of data includes: A photograph that does not contain cigarette butts and ...

Embodiment 2

[0088] Embodiment 2 of the present invention provides an anchor-free and multi-scale feature self-adaptive fusion method for detection and identification of personnel smoking at oil production sites, including:

[0089] Utilize the camera installed on the oil extraction operation site to collect front view images with cigarette butts and without cigarette butts in the scene environment as a binary classification data set, and label the front view images to obtain the label information of the front view images after preprocessing; then input to Extracting a plurality of characteristic information from the pre-constructed initial oil recovery operation site smoking detection network, and merging the plurality of characteristic information to obtain fusion characteristic information;

[0090] Acquiring prediction information of each pixel of the front view image according to the fusion feature information, and calculating an error between the prediction information and the label i...

Embodiment 3

[0109] Such as figure 2 As shown, Embodiment 3 of the present invention provides an anchor-free and multi-scale feature self-adaptive fusion method for detection and identification of personnel smoking at the oil production site. The specific operation steps of the method are as follows:

[0110] Step S11: Deploy the pre-trained smoke detection network for personnel at the oil production site to the server at the oil production site.

[0111] Step S12: Connect the video stream generated by the camera at the oil production site to the server through the central processing unit, input the obtained video of the construction workers at the oil production site to the smoking detection network of the oil production site personnel that has completed the training in advance, and output the Video detection information.

[0112] Step S13: According to the detection information of the video, a judgment result is made on whether there are cigarette butts at the oil production site.

[...

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 provides a method and a system for recognizing and processing smoking personnel in an oil production operation site, and belongs to the technical field of machine vision. Determining whether a cigarette end exists in the environment picture or not by using an recognition model; wherein the recognition model is obtained through machine learning training by using multiple groups of data; wherein the multiple groups of data comprise first-class data and second-class data; wherein each group of data in the first type of data comprises a photo containing cigarette butts and a label for recognizing the cigarette butts contained in the photo; wherein each group of data in the second type of data comprises a photo which does not contain cigarette butts and a label which identifies that the photo does not contain the cigarette butts; and under the condition that the cigarette end exists in the environment picture, it is judged that the person on the oil production operation site has the smoking behavior, and an early warning signal is sent out. According to the method, the multi-scale features of the image are extracted, the multi-layer feature mapping is obtained, the feature pyramid is obtained, the scale change problem is solved, and cigarette end recognition in the on-site picture can be quickly and accurately carried out.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a method and system for identifying and processing smokers at oil production sites based on adaptive fusion of anchor-free and multi-scale features. Background technique [0002] Due to the complex environment of the oil extraction operation site, there are major safety hazards in the non-standard operation of the oil extraction construction workers, especially the problem of personnel smoking, which may cause major accidents. Real-time and accurate detection of the smoking behavior of personnel on the oil production site and an immediate alarm are of great significance to ensure the production safety of the oil production site. Fast and accurate detection of cigarette butts in on-site video images is crucial to identifying whether on-site personnel are smoking, but cigarette butts are small target problems and difficult to detect. [0003] In the field of computer vision...

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): G06K9/00G06K9/46G06K9/54G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/084G06V40/20G06V20/41G06V10/464G06V10/20G06V2201/07G06F18/24
Inventor 梁鸿杨莹魏学成巩亚明邵明文张千
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
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