Research on Calculation of Defect Area of Building Exterior Windows Based on Infrared Image Processing Technology
-
摘要: 将红外热成像与图像处理技术结合应用于建筑外窗缺陷的检测，提出一种外窗缺陷检测和面积计算方法。通过外窗缺陷检测实验，利用压差法进行外窗空气渗透检测，求出渗透的缺陷面积。将红外热成像仪采集的外窗红外图像进行图像的预处理、外窗缺陷的检测以及检测后的面积计算，并建立外窗缺陷红外图像检测模型。结果表明：利用加权平均法进行灰度化处理，中值滤波进行降噪处理、图像锐化和直方图均衡化进行图像增强处理，处理效果明显，可作为外窗红外图像的预处理方式；Roberts算法对预处理后外窗红外图像的检测与实验值差异最小，检测信息更接近实际缺陷位置；将处理方法和检测模型与建筑整体气密性检测结合，能够在现场对外窗气密性能等级进行初步判定。Abstract: A method for defect detection and area calculation of exterior windows of buildings is proposed by combining infrared thermal imaging technology and image processing technology. Using equipment for detection of building exterior window defects, the differential-pressure method was utilized to detect the air penetration of an exterior window, and the defective area of the air penetration of this window was calculated. Infrared images of the exterior window of the building collected by an infrared thermal imager were subjected to image preprocessing, exterior window defect detection, and area calculation after inspection. Then, an infrared-image detection model of exterior window defects was established. The results show that preprocessing can make use of the weighted average method for grayscale processing, the median filter for noise reduction, image sharpening, and histogram equalization for image enhancement processing. The outcome of the aforementioned approaches is evident. The detection of the pretreatment infrared image, which is obtained using the Roberts algorithm, minimizes the difference between the test and experimental values. This makes the detection information closer to the actual position of the defect. A primary assessment of the airtightness performance level of exterior windows can be achieved by comparing the results provided by the proposed infrared image processing technology with airtightness on-site tests.
-
0. 引言
随着科学技术的发展，人脸识别在刷脸支付、刷脸打卡等信息领域中具有广泛的应用。人脸识别系统主要由CCD、摄像镜头组和滤光片构成，滤光片是过滤非成像光波段，提高人脸识别的精确度的一个核心光学元件，需要工作在特定的入射角范围下。当入射角超出特定角度范围时，滤光片滤波特性会发生质的改变，导致拍摄人脸图像模糊，人脸识别精确度下降。
滤光片的角度效应指入射角变化对滤波特性的影响[1-4]，为了降低入射角变化对滤波特性的影响，科研人员对低角度效应滤光片进行了相关研究。2013年，K. D. Hendrix等[1]研制出一种近红外波段的低角度效应滤光片，通带透射率大于90%，入射角为0°~30°，通带偏移量为12.2 nm，薄膜厚度超过5 μm；2016年，毛克宁[2]使用氧化硅作为间隔层材料制备出一种具有低角度效应的滤光片，当入射角为0°~60°时，透过率峰的位置基本保持不变；2019年，刘冬梅[3]等使用Si-H、Si3N4及SiO2三种材料研制出61层具有低角度效应的虹膜识别滤光片，入射角为0°~38°，通带偏移量为19.2 nm；2020年，魏博洋等[4]使用Si-H和SiO2材料设计出71层用于3D成像的945 nm窄带滤光片，入射角为0°~38°，通带偏移量为13 nm。虽然这些滤光片都有较低的角度效应，但是设计膜层数多，制备困难。在光学薄膜中通常使用TiO2和SiO2（高匹配度的材料组合）进行膜系设计，然而目前对于低膜层数及对低角度效应的TiO2/SiO2窄带滤光片研究较少。
通过对人脸识别系统的了解，本文对人脸识别中入射角小于22°的940 nm窄带滤光片进行设计及制备。选择高匹配度的TiO2和SiO2材料，以法布里珀罗干涉原理设计窄带滤光片。通过改变间隔层材料，增加间隔层厚度，解决传统滤光片膜层数多及角度效应高的问题。最后使用电子束热蒸发沉积技术制备滤光片，对制备的滤光片进行了角度效应和耐性测试。
1. 设计
1.1 膜系材料选择
综合考虑膜层材料对薄膜的光学性能、机械及化学稳定性等的影响，选用具有高匹配度的TiO2（n＝2.2~2.3）和SiO2（n＝1.46）作为膜层材料。TiO2具有很高的折射率，它与低折射率材料一起使用时，能够提高截止带的截止率并适当降低膜系层数，减小膜系厚度；SiO2膜层牢固高、化学性质稳定，蒸镀技术成熟，容易控制[5-6]。
1.2 膜系设计
本文以法布里-玻罗干涉滤光片原理为基础进行膜系设计，因为单腔法布里-玻罗窄带滤光片透射率曲线的通带宽度、透射率峰的矩形度、陡度、截止透射率等性能均不理想，达不到使用要求，所以将多个单腔窄带滤光片组合起来构成多腔窄带滤光片，最终设计的窄带滤光片具有更好的光学性能[7]。
采用Essential Macleod软件进行膜系设计，探索多腔法布里-玻罗干涉滤光片膜系的干涉的级次，反射层层数以及多腔串置腔的个数在对膜系的截止区、半宽度、矩形度和陡度等因素产生的影响[8-10]。综合考虑膜系在后期的镀制条件，以及设计指标要求，本文采用两腔设计，干涉级次为一，反射层数为二，间隔层使用高折射率材料。
首先，以石英玻璃为基底进行基础膜系设计，设计结构为：A(2、1、1)LA(2、1、1)，基础膜系不同入射的角透射率曲线如图 1所示。由图 1可以看出，入射角为0°时，通带峰值透射率大于95%，截止透射率小于5%，通带半峰宽度为35 nm；入射角为15°时，940 nm处透射率大于90%，通带偏移量为11 nm。入射角为22°时，940 nm处透射率小于90%，通带偏移量为20 nm。入射角变大，通带向短波方向偏移。
![]()
图 1 基础膜系中不同入射角的透射率曲线Figure 1. Transmittance curves of different incident angles in the basic film system为使膜层数更少，膜系具有更低的角度效应，所以采用高折射率膜料替换膜系间隔层中的低折射率膜料，构成高折射率间隔腔层，提升腔层的等效折射率n∗[11]。在不影响中心波长透射率、通带半宽度、透射率峰的矩形度、陡度、峰值透射率和截止区的截止率等性能的条件下，大大减少膜层数，降低入射角灵敏度。优化后的膜系为：HL(6H)LHLHL(6H)LH，膜层数为11层。相对基础膜系，膜层数减少8层。
优化后的膜系在入射角为0°、15°和22°时透射率曲线如图 2所示。由图 2可以看出，在入射光垂直入射时中心波长的峰值透射率大于95%，截止区的截止透射率小于5%，半波宽为44 nm。当入射角为22°时通带的偏移量为14 nm，940 nm处透射率大于90%，满足设计要求。结合优化后的透射率曲线，使用HGLP-850颜色玻璃抑制可见光波段杂散光干扰[12]，将所设计的膜系基底换为HGLP-850颜色玻璃，完成最终的膜系设计。
![]()
图 2 优化后的膜系中不同入射角的透射率曲线Figure 2. Transmittance curves of the optimized film system at different incident angles2. 薄膜制备
本实验采用电子束热蒸发技术制备TiO2和SiO2薄膜[13-14]，所使用的设备是成都南光机器有限公司生产的型号为ZZS-800电子束蒸发镀膜机。在镀膜之前，将HGLP-850颜色玻璃基片放在无水乙醇中超声波清洗20 min。蒸发的膜料为高纯度TiO2和SiO2颗粒，电子枪预熔膜料前，腔体的本底真空度抽至7×10－4 Pa。电子束蒸发镀膜过程中，真空度为5.7×10－3 Pa，电子枪电压为8 kV，TiO2和SiO2的电子束流分别是95 mA与65 mA。TiO2和SiO2的电子束流大小与预熔时相同，沉积速率为0.1 nm/s。第3和第9层TiO2薄膜的物理厚度为626.67 nm，其余TiO2膜层物理厚度均为104.44 nm，SiO2膜层厚度均为104.44 nm。整个镀制过程由上海英福康公司生产的SQC310膜厚控制仪自动完成。镀制完成的滤光片如图 3所示。
3. 测试及分析
根据GJB 2485-95对光学薄膜附着力的测试标准，对TiO2/SiO2膜层，采用3M胶带进行测试，用胶带反复粘连膜面，对膜面撕扯20次，膜层未出现脱落、损伤现象。将样本放入沸水中煮30 min，煮后的样本无颜色变化、无脱膜现象，该滤光片在水汽和湿热环境下有很好的防水汽性能。
光谱测试设备使用日本岛津公司生产的IRPrestige-21型傅里叶变换红外光谱仪，制备的940 nm窄带滤光片的透射率光谱曲线如图 4所示。入射角为0°，窄带滤光片工作中心波长为940 nm，在截止区间（200~1100 nm）内，通带峰值透射率为83.4%，通带半宽度为45 nm，平均截止透射率小于1%。入射角为22°，通带向短波方向偏移量为14 nm，透射率大于80%。综上分析可得，940 nm窄带滤光片在入射角为0°~22°时，通带偏移量为14 nm，940 nm透射率大于80%，平均截止透射率小于1%。图 5为在800~1100 nm局部放大的透射率光谱曲线。
![]()
图 4 入射角为0°和22°时透射率光谱曲线Figure 4. The transmittance spectrum curves when the incident angle is 0° and 22°![]()
图 5 局部放大的通带区间透射率光谱曲线Figure 5. Partially amplified passband transmittance spectrum curves4. 结论
本文以法布里-珀罗干涉滤光片原理为基础，利用Essential Macleod软件设计低角度效应的人脸识别窄带滤光片，并使用电子束热蒸发沉积技术制备了低角度效应的人脸识别窄带滤光片。通过实验分析，使用TiO2作为膜系间隔层材料，提升膜系等效折射率n*，使滤光片的膜层数降低，改善角度效应。光在0°~22°入射时，滤光片通带透过率大于80%，偏移量为14 nm，截止率小于1%，数据能满足人脸识别窄带滤光片的技术指标。采用TiO2和SiO2进行设计和制备，使膜层附着力良好，不仅极大减少了膜层数量，降低了膜层厚度，而且还降低了薄膜制备的工艺要求。人脸识别系统中，低角度效应窄带滤光片不仅能缓解人脸识别系统角度受限的问题，而且还能增强系统对杂散光的抗干扰能力，提升人脸识别系统的识别准确性。
-
![]()
图 1 红外热成像仪采集建筑外窗图像原理
Figure 1. Schematic diagram of infrared thermal imager collecting images of building windows
1. Thermal imager; 2. Tripod; 3. Building wall; 4. Building windows
![]()
图 3 灰度化处理图(a)~(h)与灰度分布直方图(a1)~(h1)
Figure 3. Gray processing diagram(a)-(h) and gray distribution histogram(a1)-(h1)
![]()
图 4 图像增强结果对比(a)、(f)、(k)灰度化处理；(b)、(g)、(l)中值滤波处理；(c)、(h)、(m)拉普拉斯锐化；(d)、(i)、(n)直方图均衡化；(e)、(j)、(o)两种方式混合
Figure 4. Comparison of image enhancement results: (a), (f), (k) Gray scale processing; (b), (g), (l) Median filtering; (c), (h), (m) Laplacian sharpening; (d), (i), (n) Histogram equalization; (e), (j), (o) Two methods mixed
![]()
图 5 外窗红外图像预处理结果：(a)、(e)、(i)原始红外图像；(b)、(f)、(j)灰度化处理；(c)、(g)、(k)中值滤波；(d)、(h)、(l)图像增强结果
Figure 5. Preprocessing results of infrared image of windows: (a), (e), (i) Infrared imagery; (b), (f), (j) Gray scale processing; (c), (g), (k) Median filtering; (d), (h), (l)Image enhancement
![]()
图 6 各方法进行外窗缺陷检测结果：(a)~(f) Roberts、Sobel、Prewitt、Canny、Log算法、阈值分割法
Figure 6. Testing results of window defects by each method: (a)-(f)Roberts, Sobel, Prewitt, Canny, Log, threshold segmentation
![]()
图 7 各方法检测后形态学处理结果：(a)~(f)Roberts、Sobel、Prewitt、Canny、Log算法、阈值分割法
Figure 7. Morphological processing results after detection by each method: (a)-(f)Roberts, Sobel, Prewitt, Canny, Log, Threshold segmentation
![]()
图 8 外窗缺陷实验值与图像处理值的散点图
Figure 8. Scatter diagram of experimental values and image processing values of window defects
表 1 外窗缺陷面积对比
Table 1 Defect area comparison table of windows
Windows Value of experiment/cm2 Roberts/cm2 Sobel/cm2 Prewitt/cm2 Canny/cm2 Log/cm2 Threshold value segmentation/cm2 C0407(1) 0.90 0.96 1.39 1.39 1.70 1.62 0.92 C0407(2) 1.50 1.50 1.80 1.80 1.40 2.10 1.50 C0709(1) 3.20 3.28 4.07 3.94 3.49 5.86 3.68 C0709(2) 3.20 3.21 4.03 3.84 3.46 4.48 3.65 C0814(1) 4.30 5.19 11.32 11.07 6.58 11.61 4.70 C0814(2) 4.10 4.53 11.93 11.81 5.59 12.64 4.30 C1218(1) 11.20 11.50 14.44 14.40 15.44 7.20 11.52 C1218(2) 10.50 10.95 18.13 17.89 11.73 14.47 11.32 C1218(3) 10.30 11.03 14.22 14.02 10.89 14.76 10.43 C1218(4) 11.10 11.30 14.07 13.82 12.94 7.93 11.21 C1716(1) 9.20 9.30 11.92 11.90 9.32 4.96 9.30 C1716(2) 14.50 15.01 16.79 16.62 14.83 9.57 14.24 C2114(1) 15.80 14.86 25.42 24.88 22.93 23.33 16.87 C2114(2) 11.50 13.80 17.20 17.00 13.80 16.30 13.90 C2418(1) 22.10 24.50 30.92 30.50 36.28 26.65 25.66 C2418(2) 20.90 24.60 31.80 31.40 23.40 26.90 24.60 Error of mean - 7.23% 56.01% 53.76% 26.18% 61.60% 7.67% 
下载: 导出CSV
-
[1] Mathur U, Damle R. Impact of air infiltration rate on the thermal transmittance value of building envelope[J]. Journal of Building Engineering, 2021, 40: 102302. DOI: 10.1016/j.jobe.2021.102302
[2] Lai Y, Ridley I, Brimblecombe P. Blower-door estimates of PM 2.5 deposition rates and penetration factors in an idealized room[J]. Indoor and Built Environment, 2022, 31(8): 2064-2082. DOI: 10.1177/1420326X20944427
[3] 中国建筑科学研究院有限公司. 建筑外门窗气密、水密、抗风压性能检测方法: GB/T 7106-2019[S]. 国家市场监督管理总局, 国家标准化管理委员会, 2019. China Academy of Building Research, Test methods of air permeability, water tightness, wind load resistance performance for building external windows and doors: GB/T 7106-2019[S]. State Administration for Market Regulation; Standardization Administration, 2019.
[4] 张小愉. 建筑门窗三性检测方法探究[J]. 中国建材科技, 2019, 28(6): 12-13. https://www.cnki.com.cn/Article/CJFDTOTAL-JCKJ201906006.htm ZHANG Xiaoyu. Exploration on three-character detection method of architectural doors and windows [J]. China Building Materials Science & Technology, 2019, 28(6): 12-13. https://www.cnki.com.cn/Article/CJFDTOTAL-JCKJ201906006.htm
[5] 张浚泉. 建筑门窗工程质量检测探讨[J]. 门窗, 2017(2): 23-24. https://www.cnki.com.cn/Article/CJFDTOTAL-MENC201803010.htm ZHANG Junquan. Discussion on quality inspection of building door and window engineering[J]. Doors & Windows, 2017(2): 23-24. https://www.cnki.com.cn/Article/CJFDTOTAL-MENC201803010.htm
[6] Figuli L, Papan D, Papánová Z, et al. Experimental mechanical analysis of traditional in-service glass windows subjected to dynamic tests and hard body impact[J]. Smart Structures and Systems, 2021, 27(2): 365-378.
[7] JI Y, Duanmu L. Airtightness field tests of residential buildings in Dalian, China[J]. Building and Environment, 2017, 119(7): 20-30. https://www.sciencedirect.com/science/article/pii/S0360132317301610
[8] Barreira E, Almeida R M S F, Moreira M. An infrared thermography passive approach to assess the effect of leakage points in buildings[J]. Energy & Buildings, 2017, 140(4): 224-235. https://www.sciencedirect.com/science/article/pii/S0378778817304206
[9] Freitas S S D, Freitas V P D, Barreira E. Detection of fade plaster detachments using infrared thermography – a nondestructive technique[J]. Construction & Building Materials, 2014, 70(11): 80-87.
[10] Ibarra-Castanedo C, Sfarra S, Klein M, et al. Solar loading thermography: time-lapsed thermographic survey and advanced thermographic signal processing for the inspection of civil engineering and cultural heritage structures[J]. Infrared Physics & Technology, 2017, 82: 56-74. https://www.sciencedirect.com/science/article/pii/S135044951630620X
[11] Mahmoodzadeh M, Gretka V, Wong S, et al. Evaluating patterns of building envelope air leakage with infrared thermography[J]. Energies, 2020, 13(14): 3545. DOI: 10.3390/en13143545
[12] CHEN Y F, SANG N. Attention-based hierarchical fusion of visible and infrared images[J]. Optik, 2015, 126(23): 4243-4248. DOI: 10.1016/j.ijleo.2015.08.120
[13] Lourenço Tomás, Matias Luís, Faria P. Anomalies detection in adhesive wall tiling systems by infrared thermography[J]. Construction and Building Materials, 2017, 148: 419-428. DOI: 10.1016/j.conbuildmat.2017.05.052
[14] Hiasa S H, Birgul R, Catbas N, et al. A data processing methodology for infrared thermography images of concrete bridges[J]. Computers & Structures, 2017, 190(10): 205-218. https://www.sciencedirect.com/science/article/pii/S0045794916312238
[15] Omar T, Nehdi M L, Zayed T. Infrared thermography model for automated detection of delamination in RC bridge decks[J]. Construction & Building Materials, 2018, 168: 313-327. DOI: 10.12096/j.2096-4528.pgt.2018.048
[16] LV C, WANG K, GU G, et al. Accurate full-edge detection and depth measurement of internal defects using digital speckle pattern interferometry[J]. NDT & E International, 2019, 102(3): 1-8. https://www.sciencedirect.com/science/article/pii/S0963869518302445
[17] Mohan A, Poobal S. Crack detection using image processing: A critical review and analysis[J]. Alexandria Engineering Journal, 2017: 787-798. https://www.sciencedirect.com/science/article/pii/S1110016817300236
[18] HAN Q, YIN Q, ZHENG X, et al. Remote sensing image building detection method based on mask R-CNN[J]. Complex & Intelligent Systems, 2022, 8(1): 1847-1835.
[19] Gehri N, J Mata-Falcón, Kaufmann W. Automated crack detection and measurement based on digital image correlation[J]. Construction and Building Materials, 2020, 256: 119383. DOI: 10.1016/j.conbuildmat.2020.119383
[20] Thusyanthan I, Blower T, Cleverly W. Innovative uses of thermal imaging in civil engineering[J]. Proceedings of the Institution of Civil Engineers, 2017, 170(CE2): 81-87. DOI: 10.1680/jcien.16.00014
[21] Ostańska Anna. Thermal imaging for detection of defects in envelopes of buildings in use: qualitative and quantitative analysis of building energy performance[J]. Periodica Polytechnica Civil Engineering, 2018, 62(4): 939-946.
[22] PAN N H, Tsai C H, CHEN K Y, et al. Enhancement of external wall decoration material for the building in safety inspection method[J]. Journal of Civil Engineering and Management, 2020, 26: 216-226.
[23] JIANG M. Edge enhancement and noise suppression for infrared image based on feature analysis[J]. Infrared Physics & Technology, 2018, 91: 142-152. https://www.sciencedirect.com/science/article/pii/S1350449517308617
[24] Kim C, Choi J S, Jang H, et al. Automatic detection of linear thermal bridges from infrared thermal images using neural network[J]. Applied Sciences, 2021, 11(3): 931.
[25] CHEN L, WANG Y, JIA S, et al. Development of panoramic infrared images for surface temperature analysis of buildings and infrastructures[J]. Energy and Buildings, 2020, 232(1): 110660.
[26] D Antón, Amaro-Mellado J L. Engineering graphics for thermal assessment: 3D thermal data visualisation based on infrared thermography, GIS and 3D point cloud processing software[J]. Symmetry, 2021, 13(2): 335.
[27] Mun J, Lee J, Kim M. Estimation of infiltration rate (ACH Natural) using blower door test and simulation[J]. Energies, 2021, 14(4): 912.
[28] LI X, ZHOU W, LIN D. Research on air infiltration predictive models for residential building at different pressure[J]. Building Simulation, 2020: 1-12.
[29] DI Yue, LI Meiyan, QIAO Tong, et al. Edge detection and mathematic fitting for corneal surface with Matlab software[J]. International Journal of Ophthalmology, 2017, 10(3): 336-342. http://www.ijo.cn/gjyken/ch/reader/view_abstract.aspx?file_no=20170302&flag=1
[30] Thenkabail P S, Biradar R M, Noojipady P, et al. Sub-pixel area calculation methods for estimating irrigated areas[J]. Sensors, 2007, 7(11): 2519. https://www.mdpi.com/1424-8220/7/11/2519
[31] 李峰, 周雷, 苗刚中. 基于单目视觉的目标测距方法[J]. 中国科学技术大学学报, 2012, 42: 93-98. https://www.cnki.com.cn/Article/CJFDTOTAL-WJFZ202102016.htm LI Feng, ZHOU Lei, MIAO Gangzhong. Target ranging method based on monocular vision[J]. Journal of University of Science and Technology of China, 2012, 42: 93-98. https://www.cnki.com.cn/Article/CJFDTOTAL-WJFZ202102016.htm
[32] 杨玉忠, 李丽梅, 陈刚, 等. 利用红外热像仪测试窗洞口热工缺陷的研究[J]. 建筑科学, 2016, 32(8): 111-114. https://www.cnki.com.cn/Article/CJFDTOTAL-JZKX201608019.htm YANG Yuzhong, LI Limei, CHEN Gang, et al. Study on the thermodynamic disfigurement testing for window opening by infrared thermography[J]. Building Science, 2016, 32(8): 111-114. https://www.cnki.com.cn/Article/CJFDTOTAL-JZKX201608019.htm
[33] 冯力强, 王欢祥, 晏大玮, 等. 红外热像法检测建筑外墙饰面层内部缺陷试验研究[J]. 土木工程学报, 2014, 47(6): 51-56. https://www.cnki.com.cn/Article/CJFDTOTAL-TMGC201406010.htm FENG Liqiang, WANG Huanxiang, YAN Dawei, et al. Experimental study on internal defects detection of exterior wall finish coat by infrared thermography[J]. China Civil Engineering Journal, 2014, 47(6): 51-56. https://www.cnki.com.cn/Article/CJFDTOTAL-TMGC201406010.htm
[34] 田鹏飞, 周骥平, 朱兴龙, 等. 基于红外热成像的气密性检测技术探讨[J]. 扬州大学学报(自然科学版), 2013, 16(2): 45-48. https://www.cnki.com.cn/Article/CJFDTOTAL-YZDZ201302012.htm TIAN Pengfei, ZHOU Jiping, ZHU Xinglong, et al. Research on air tightness detection based on infrared thermal imaging[J]. Journal of Yangzhou University(Natural Science Edition), 2013, 16(2): 45-48. https://www.cnki.com.cn/Article/CJFDTOTAL-YZDZ201302012.htm
计量
- 文章访问数: 121
- HTML全文浏览量: 30
- PDF下载量: 31
