基于多特征的雾状气体泄漏视觉检测方法
作者单位:

南京航空航天大学

基金项目:

中央高校基本科研业务费专项资金资助(NJ2024012)


Multi-feature based visual detection of misty gas leakage
Affiliation:

Nanjing University of Aeronautics and Astronautics

Fund Project:

the Fundamental Research Funds for the Central Universities(NJ2024012)

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [31]
  • | | | |
  • 文章评论
    摘要:

    为解决可见光雾状气体泄漏样本数据稀缺、针对性特征缺乏情况下的泄漏检测问题,研究基于多特征的雾状气体泄漏视觉检测方法。利用雾状气体泄漏前后图像特征差异,设计包含透射率、相合度、均匀度的像素级雾状气体泄漏特征。结合支持向量机的决策机制和指数平滑思想,引入置信度优化、平滑因子约束集成算法权值,构建自适应雾状气体泄漏分类模型,实现可见光图像中雾状气体泄漏状态的准确判定。在典型泄漏场景对该方法展开验证的结果表明:该方法能较好地判定气体泄漏状态,成功率达到95%以上,单帧平均耗时约为35 ms,满足了检测实时性要求。

    Abstract:

    In order to solve the problem of leakage detection in the case of scarce visible foggy gas leakage sample data and lack of targeted features, a multi-feature based visual detection method for foggy gas leakage is investigated. Using the differences in image features before and after fog gas leakage, pixel-level fog gas leakage features including transmittance, coherence and uniformity are designed. Combining the decision-making mechanism of support vector machine and the idea of exponential smoothing, we introduce confidence optimization and smoothing factor constraints to integrate the weights of the algorithm, construct an adaptive foggy gas leakage classification model, and realize the accurate determination of the foggy gas leakage state in the visible light image. The results of the validation of the method in a typical leakage scenario show that the method can better determine the gas leakage state, with a success rate of more than 95%, and the average time consumed for a single frame is about 35 ms, which meets the real-time requirements of detection.

    参考文献
    [1] 吕伟,蒋翠玲,姜雅娟,等.考虑多因素影响的毒气扩散风险潜势图构建[J].中国安全科学学报,2020,30(11):175-181.
    [2] Shuangyan X,Jinghong W,Jia L, et al.System dynamics research of non-adaptive evacuation psychology in toxic gas leakage emergencies of chemical park[J].Journal of Loss Prevention in the Process Industries,2021,(prepublish):104556-.
    [3] 杨德振,柴欣,刘彤,等.一种空中加油机的油气泄漏检测方法、装置[P].北京市:CN202210727793.1,2022-11-15.
    [4] 翁静,袁盼,王铭赫等.基于支持向量机的泄漏气体云团热成像检测方法[J].光学学报,2022,42(09):104-111.
    [5] 李泉成,曹江涛,姬晓飞.光流增强的红外成像气体泄漏检测方法[J].电子测量与仪器学报,2023,37(03):50-56.
    [6] 刘泽军.便携式超声波气体泄漏检测仪的设计与实现[D].中北大学,2022.
    [7] 黄晓明,王忠华.基于传感网络和PC_DSR GA的泄漏源定位系统[J].实验室研究与探索,2020,39(07):128-132+157.
    [8] 彭敏,马宁,王少军,等. 基于可见光图像的工业过程气体泄漏检测研究 [J]. 电子测量与仪器学报, 2019, 33 (06): 1-8.
    [9] 邓永川. 基于图像处理的油田漏油检测方法设计与研究[D]. 电子科技大学,2020.
    [10] 施琦.基于深度学习的可见光遥感图像在轨目标识别技术研究[D].中国科学院大学(中国科学院微小卫星创新研究院),2023.
    [11] 李宝奇,黄海宁,刘纪元,等.基于改进SSD的合成孔径声纳图像感兴趣小目标检测方法[J].电子学报,2024,52(03):762-771.
    [12] Yin X,Chen L . Image Object Detection Method Based on Improved Faster R-CNN [J]. Journal of Circuits, Systems and Computers, 2023, 33 (07):1-15.
    [13] 王云龙. 基于机器学习的水下目标检测方法研究[D].哈尔滨工程大学,2022.
    [14] 林奕桐,梁健,刘书田,等.基于无人机可见光通道及支持向量机模型的柑橘黄龙病个体识别方法[J].西南农业学报,2022,35(11):2554-2563.
    [15] 王乃芯. 多分类支持向量机的研究[D]. 华东师范大学,2020.
    [16] Peignier S,Lacotte V,Duport G M, et al. Detection of Aphids on Hyperspectral Images Using One-Class SVM and Laplacian of Gaussians [J]. Remote Sensing, 2023, 15 (8):2103.
    [17] 曾宝国,尹文刚.基于SIFT与SVM的应急救援图像检测方法研究[J].中国安全生产科学技术,2020,16(08):186-192.
    [18] 赵玉清,杨慧丽,张悦,等. 基于特征组合与SVM的小粒种咖啡缺陷生豆检测[J]. 农业工程学报, 2022, 38 (14): 295-302.
    [19] 刘晓玉,李灯熬,赵菊敏.基于多核SVM的AdaBoost心力衰竭死亡率评估模型[J].太原理工大学学报,2023,54(05):804-811.
    [20] 陈逊龙,孙一铭,郭仕杰,等.应用无人机可见光影像和面向对象的随机森林模型对城市树种分类[J].东北林业大学学报,2024,52(03):48-59.
    [21] 李佳,郭剑毅,刘艳超,等.基于多分类器加权投票法的越南语组合歧义消歧[J].计算机科学,2018,45(01):167-172.
    [22] Ping Y,Jian F,Junting X, et al.The Optimization of Multi-classifier Ensemble Method Based on Dynamic Weighted Voting[J].Journal of Physics: Conference Series,2022,2185(1):1-8.
    [23] Dipen S,Rachit G,Rahul M, et al.HFRAS: design of a high-density feature representation model for effective augmentation of satellite images[J].Signal, Image and Video Processing,2023,18(2):1393-1404.
    [24] Abbas Q .基于HOG、LBP和多分类SVM的板球视频事件识别方法研究[D].天津大学,2020.(Abbas Q .Cricket Video Events Recognition using HOG, LBP And Multi-class SVM[D].Tianjin University,2020.)
    [25] Mukku C,Santhosh M .Tri-stage offline Telugu character recognition system based on fusion of HOG and ULBP[J].Measurement: Sensors,2024,32101059-.
    [26] Amanpreet K,Munish K,K. M J .Cattle identification system: a comparative analysis of SIFT, SURF and ORB feature descriptors[J].Multimedia Tools and Applications,2023,82(18):27391-27413.
    [27] Li S .Dynamic Interactive Design of Graphic Design Elements Based on Image Feature Extraction Algorithm[J].Applied Mathematics and Nonlinear Sciences,2024,9(1):1-14
    [28] Li W .Texture feature extraction of a landscape design image based on the contour wave transform[J].International Journal of Data Science,2023,8(1):39-51.
    [29] 郑凤仙,王夏黎,何丹丹等.单幅图像去雾算法研究综述[J].计算机工程与应用,2022,58(03):1-14.
    [30] 石冬阳,黄胜,刘焕淋,等.具有透射率先验与亮度感知的图像去雾算法[J].红外与激光工程,2024,53(07):279-292.
    [31] 韩昊男,钱锋,吕建威等.改进暗通道先验的航空图像去雾[J].光学精密工程,2020,28(06):1387-1394.第一作者简介:陈旗(2000年),性别(男),江苏盐城,硕士,研究方向为视觉导航。(联系地址:江苏省南京市江宁区秣陵街道将军路29号南京航空航天大学将军路校区自动化学院;电话:15751106839;邮箱:chenqichance@nuaa.edu.cn)。基金项目:中央高校基本科研业务费专项资金资助(NJ2024012)。
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文
分享
文章指标
  • 点击次数:2
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2024-11-15
  • 最后修改日期:2025-01-02
  • 录用日期:2024-11-29
文章二维码