基于运动与模糊特征的红外热成像烟雾检测

李咸静; 郝争辉

基于运动与模糊特征的红外热成像烟雾检测

李咸静^1,,
郝争辉²

1.
长治学院计算机系, 山西长治 046000
2.
山西国科晋云信息产业有限公司, 山西太原 030002

基金项目:

山西省高等学校科技创新计划项目 2023L323

详细信息

作者简介:
李咸静（1992-），女，山西长治人，博士，讲师，研究方向为计算机视觉及图像处理。E-mail：760072875@qq.com

中图分类号: TP391
计量
- 文章访问数: 39
- HTML全文浏览量: 15
- PDF下载量: 10
- 被引次数: 0
出版历程
- 收稿日期: 2022-12-03
- 修回日期: 2023-07-10
- 刊出日期: 2024-03-20

Infrared Thermal Imaging Smoke Detection Based on Motion and Fuzzy Features

LI Xianjing^1
,,
HAO Zhenghui²

1.
Department of Computer Scienice, Changzhi University, Changzhi 046000, China
2.
Shanxi Guoke Jinyun Information Industry Co., Taiyuan 030002, China

摘要

摘要: 在焦化企业生产过程中会产生大量烟雾，排放、泄漏会对自然环境造成污染危及生命生产安全。针对热成像视频对比度低、纹理度差等特点，本文采用烟雾具有运动与模糊特性来进行检测。提出计算每帧图像的噪点度来改进Vibe检测算法的固定阈值，从而可更完整地将移动目标区域检测出来。将整幅图像划分为块区域图像；结合运动区域来提取该区域内的模糊与噪点比值、FFT计算模糊度所计算特征训练生成烟雾分类器；对实验视频进行检测，平均准确率达到94.53%。结果表明，本文所提算法对焦化企业红外热成像视频烟雾检测的准确性和实时性，具有良好的抗干扰能力。
- 改进Vibe模糊烟雾特征 /
- 红外热成像 /
- 烟雾检测
Abstract: The production process of coking enterprises generates abundant smoke. Their discharge and leakage can pollute the natural environment, endangering the safety of life and production. Considering the low contrast and poor texture of thermal imaging videos, this study detected smoke with motion and fuzzy characteristics. The noise degree of each frame image can be calculated to replace the fixed threshold of the Vibe detection algorithm so that the moving target area can be completely detected. First, the image was divided into block area images; then, the fuzzy-to-noise ratio in this area was extracted by combining the motion area, the features calculated when the fast fourier transform (FFT) was used to calculate the ambiguity were trained to generate a smoke classifier, and finally, the experimental video detection, with an average accuracy rate of 94.53%. The results show that the proposed algorithm is accurate, operates in real-time for smoke detection in infrared thermal imaging videos of coking enterprises, and has good anti-interference ability.
- improved Vibe fuzzy smoke features /
- infrared thermal imaging /
- smoke detection

HTML全文

图 1 ViBe像素分类示意图

Figure 1. Schematic diagram of ViBe pixel classification

下载: 全尺寸图片幻灯片

图 2 ViBe背景模型的随机更新

Figure 2. Stochastic update of the ViBe background mode l

下载: 全尺寸图片幻灯片

图 3 Vibe检测目标与改进后的算法结果

Figure 3. Vibe detection target and improved algorithm results

下载: 全尺寸图片幻灯片

图 4 有烟雾和无烟雾块区域图像所提取的纹理特征变化

Figure 4. Variation of texture features extracted from images with and without smoke block region

下载: 全尺寸图片幻灯片

图 5 本文算法提取特征

Figure 5. Features extracted by the algorithm of this paper

下载: 全尺寸图片幻灯片

图 6 部分移动干扰目标下的烟雾检测

Figure 6. Smoke detection under partially moving interference targets

下载: 全尺寸图片幻灯片

参考文献(12)

[1]	张斌, 魏维, 何冰倩. 基于多特征融合的早期野火烟雾检测[J]. 成都信息工程大学学报, 2018, 33(4): 408-412. https://www.cnki.com.cn/Article/CJFDTOTAL-CDQX201804010.htm ZHANG Bin, WEI Wei, HE Bingqian. Early wildfire smoke detection based on multi-feature fusion[J]. Journal of Chengdu University of Information, 2018, 33(4): 408-412. https://www.cnki.com.cn/Article/CJFDTOTAL-CDQX201804010.htm
[2]	刘通, 程江华, 华宏虎, 等. 结合Y^dU^aV^a颜色模型和改进MobileNetV3的视频烟雾检测方法[J]. 国防科技大学学报, 2021, 43(5): 80-85. https://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ202105009.htm LIU Tong, CHENG Jianghua, HUA Honghu, et al. Video smoke detection method combining Y^d U^a V^a color and improved MobileNetV3[J]. Journal of National University of Defense, 2021, 43(5): 80-85. https://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ202105009.htm
[3]	宋少杰. 基于GMM改进算法的森林火灾检测研究[D]. 南京: 南京邮电大学, 2021. DOI: 10.27251/d.cnki.gnjdc.2021.000473. SONG Shaojie. Research on Forest Fire Detection Based on proved GMM Algorithm[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2021. DOI: 10.27251/d.cnki.gnjdc.2021.000473.
[4]	Ojala T, Pietikäinen M, Harwood D. A comparative study of texture measures with classification based on featured distributions[J]. Pattern Recognition, 1996, 29(1): 51-59. doi: 10.1016/0031-3203(95)00067-4
[5]	YUAN F, SHI J, XIA X, et al. Co-occurrence matching of local binary patterns for improving visual adaption and its application to smoke recognition[J]. IET Computer Vision, 2019, 13(2): 178-187. doi: 10.1049/iet-cvi.2018.5164
[6]	YUAN F N, XIA X, SHI J T. Holistic learning-based high-order feature descriptor for smoke recognition[J]. International Journal of Wavelets, Multiresolution and Information Processing, 2019, 17(2): 194005.
[7]	邓实强, 丁浩, 杨孟, 等. 基于视频图像的公路隧道火灾烟雾检测[J]. 隧道建设, 2022, 42(2): 291-302. https://www.cnki.com.cn/Article/CJFDTOTAL-JSSD202202015.htm DENG Shiqiang, DING Hao, YANG Meng, et al. Fire smoke detection in highway tunnels based on video images[J]. Tunnel Construction, 2022, 42(2): 291-302. https://www.cnki.com.cn/Article/CJFDTOTAL-JSSD202202015.htm
[8]	WANG Y, HAN Q, LI Y, et al. Video smoke detection based on multi-feature fusion and modified random forest[J]. Engineering Letters, 2021, 29(3): 38-45.
[9]	殷梦霞, 王理, 孙连营. 基于多特征融合的自适应烟雾检测算法[J]. 建筑科学, 2019, 35(9): 26-31. DOI:10.13614/j.cnki.11-1962/tu.2019. 09.005. YIN Mengxia, WANG Li, SUN Lianying. Adaptive smoke detection algorithm based on multi-feature fusion[J]. Building Science, 2019, 35(9): 26-31. DOI: 10.13614/j.cnki.11-1962/tu.2019.09.005.
[10]	王媛彬. 煤矿烟雾的计算机视觉检测方法[J]. 辽宁工程技术大学学报: 自然科学版, 2016, 35(11): 1230-1234. https://www.cnki.com.cn/Article/CJFDTOTAL-FXKY201611007.htm WANG Yuanbin. Smoke detection based on computer vision in coal mine[J]. Journal of Liaoning Technical University: Natural Science, 2016, 35(11): 1230-1234. https://www.cnki.com.cn/Article/CJFDTOTAL-FXKY201611007.htm
[11]	马永杰, 陈梦利, 刘培培, 等. ViBe算法鬼影抑制方法研究[J]. 激光与光电子学进展, 2020, 57(2): 105-112. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202002009.htm MA Yongjie, CHEN Mengli, LIU Peipei, et al. Research on ViBe algorithm ghost suppression method[J]. Laser & Optoelectronics Progress, 2020, 57(2): 105-112. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202002009.htm
[12]	汤旻安, 王晨雨. 基于改进ViBe算法的静态场景运动目标检测[J]. 激光与光电子学进展, 2021, 58(14): 216-224. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202114021.htm TANG Minan, WANG Chenyu. Moving object detection in static scene based on improved ViBe algorithm[J]. Laser & Optoelectronics Progress, 2021, 58(14): 216-224. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202114021.htm