以物理學為基礎之雨滴去除法應用於移動物件偵測

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：56

、訪客IP：3.14.251.103

姓名

陳冠志(Guan-jhih Chen) 查詢紙本館藏

畢業系所

資訊工程學系

論文名稱

以物理學為基礎之雨滴去除法應用於移動物件偵測
(Physics-based Rain Removal for Moving Object Detection)

相關論文

★ 使用視位與語音生物特徵作即時線上身分辨識	★ 以影像為基礎之SMD包裝料帶對位系統
★ 手持式行動裝置內容偽變造偵測暨刪除內容資料復原的研究	★ 基於SIFT演算法進行車牌認證
★ 基於動態線性決策函數之區域圖樣特徵於人臉辨識應用	★ 基於GPU的SAR資料庫模擬器：SAR回波訊號與影像資料庫平行化架構 (PASSED)
★ 利用掌紋作個人身份之確認	★ 利用色彩統計與鏡頭運鏡方式作視訊索引
★ 利用欄位群聚特徵和四個方向相鄰樹作表格文件分類	★ 筆劃特徵用於離線中文字的辨認
★ 利用可調式區塊比對並結合多圖像資訊之影像運動向量估測	★ 彩色影像分析及其應用於色彩量化影像搜尋及人臉偵測
★ 中英文名片商標的擷取及辨識	★ 利用虛筆資訊特徵作中文簽名確認
★ 基於三角幾何學及顏色特徵作人臉偵測、人臉角度分類與人臉辨識	★ 一個以膚色為基礎之互補人臉偵測策略

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

在影片或監控系統下偵測移動物件時，常有物理雜訊干擾物件偵測。將雜訊去除以
提高物件偵測識別率一直是物件偵測上一個重要的課題。在不同天氣下常有不同雜訊去除的議題必須克服：雨，由於它在空間上的隨機分佈與時間上的快速移動，是相當具有挑戰性的雜訊。
這篇論文建立且改良幾位學者所提出偵測雨滴與背景估測的方法與相關策略，然而
若只是根據雨滴在亮度上變化的性質，初步所偵測的像素位置並非完全屬於雨滴。為了將誤判的區域去除，我們使用正確率高的特徵分析，在多層感知類神經網路(Multilayer Perceptron Neural Network, MLP)的訓練下，將相關環境的特徵屬性建構出來。分析特徵之一是在 YCbCr通道下對每一個像素(Pixel)計算其於CbCr的變化量，此變化量大所對應的像素位置存在移動物件的可能性較高；特徵之二是Sobel邊緣偵測，通常有雨滴經過的區域其邊緣資訊會比沒有雨滴經過的較小；特徵之三則是像素的飽和度資訊，有雨區域存在一飽和度範圍區間。
此分類的結果可用於判斷分析，除了初步去除誤判區域，運用於色彩變化相關性的
特質，更可以將移動物所位於的誤判像素去除。我們的策略包含使用雨滴在時間與空間上的相關性，標示出雨滴所存在的區域並將這些去域範圍內的雨滴去除；在時空相關性較低的戶外條件下，仍可用類神經與色彩變化相關性來去除可能誤判的移動物。
我們將此雨滴雜訊去除的架構應用於背景相減，找出正確的前景。將找出的前景做
前景區域的正確率分析，所測試環境為正常光照下固定式攝影機的影片檔。實驗的結果將顯示我們的方法改良之前的幾篇研究並有效率的偵測出前景。實驗結果顯示此一架構的確有助於雨天天候下的前景偵測，我們得知雨滴去除雖然不會提升偵測正確率，但大幅降低偵測誤判率，在背景相減之前景物偵測仍相當有幫助。

摘要(英)

Physical noises, such as rain, frequently affect the detection performance of moving-objects when they present in a film or a monitoring system. The elimination of physical noises is thus a prerequisite to uplift the detection accuracy. There are different kinds of noises needed to be eliminated under different weather conditions. The elimination of rain drops, due to the random spatial distribution and fast motion in short time, is hence
admitted as a challenging problem.
In this thesis, we improve the detection method of rain drops and construct the estimation on the intensity of background. Extensive studies were conducted in analyzing
these algorithms. However, not all of the detected pixels contain real rain drops if merely find them based on the property of intensity change. To remove the false alarm, we use three effective features in our work and construct the environmental parameters of features by MLPNN (Multilayer Perceptron Neural Network) training. One of the features is the changed value of color from Cb and Cr. The probability of a pixel containing moving objects is higher if the changed value is high. The second feature is the information obtained by the sobel edge detector. Generally, the lower value in edge will exist in the rain area. The last feature is the saturation value of a pixel. A range of saturation values exists in the rain area.
The result of classified method can be used for making decisions on false alarm, or we can use it on the algorithm in constructing the color-correlation to map more false alarms. The color-correlation is very useful for removing the false alarm in moving objects.
The proposed method is implemented on object detection to find real foreground. We analyze the detecting accuracy with and without the rain removal system in the normal illumination cameras or films. The results show that the proposed system is indeed helpful on moving-object detection in the rainy weather. It really reduces the false alarm although the accuracy is not raised with the rain removal system.

關鍵字(中)

★ 雨滴去除
★ 移動物件
★ 物件偵測
★ 物理學

關鍵字(英)

★ Object Detection
★ Moving Object
★ Rain Removal
★ Physics

論文目次

摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 x
符號說明 xi
一、緒論 1
1-1 研究動機與目的 1
1-2 研究範圍 2
1-3 主要挑戰 3
1-4 系統架構 5
1-5 論文架構 8
二、相關研究 9
三、方法與解 14
3-1 特徵擷取 14
3-2 候選雨滴偵測 19
3-3 移動物件移除 35
3-4 雨滴偵測決定 42
3-5 雨滴移除 44
四、實驗與討論 46
4-1 實驗建置 46
4-2 實驗影片 48
4-3 實驗結果 49
4-4 實驗分析 56
五、結論與未來工作 57
5-1 結論 57
5-2 未來工作 57
參考文獻 58

參考文獻

[1] K. Garg and S.K. Nayar, “Detection and removal of rain from videos,” in Proc. CVPR 2004, vol. 1, pp. 528–535.
[2] Website: Videos, Slide. http://www.cs.columbia.edu/CAVE/projects/rain_detection/
[3] Kshitiz Garg and Shree K. Nayar, “Vision and Rain,” IJCV 2007, vol. 75, no. 3-27.
[4] Y. Matsushita, et al. “Full frame video stabilization,” in Proc. CVPR 2005., vol. 1, pp. 50–57.
[5] J.R. Bergen, et al. “Hierarchical model-based motion estimation,” in Proc. ECCV 1992, pp. 237–252.
[6] Birchfield, S. 2005. KLT: An Implementation of the Kanade-Lucas-Tomasi Feature Tracker.
[7] Lucas, B.D. and Kanade, T. 1981. An iterative image registration technique with an application to stereo vision. In International Joint Conference on Artificial Intelligence.
[8] Xiaopeng Zhang, et al. “Rain removal in video by combining temporal and chromatic properties,” IEEE Conference on ICME 2006., pp. 461-464.
[9] McCartney, E.J. 1975. Optics of the Atmosphere: Scattering by molecules and particles. John Wiley and Sons.
[10] Middleton, W.E.K. 1952. Vision Through the Atmosphere. University of Toronto Press.
[11] Narasimhan, S.G. and Nayar, S.K. 2002. Vision and the Atmosphere. IJCV, 48(3):233–254.
[12] Cozman, F. and Krotkov, E. 1997. Depth from scattering. CVPR, 31:801–806.
[13] Narasimhan, S.G. and Nayar, S.K. 2003. Contrast restoration of weather degraded images. IEEE Trans. on PAMI, 25(6).
[14] Oakley, J.P. and Satherley, B.L. 1998. Improving image quality in poor visibility conditions using a physical model for degradation. IEEE Trans. on Image Processing, 7.
[15] Schechner, Y.Y., Narasimhan, S.G., and Nayar, S.K. 2001. Instant dehazing of images using polarization. CVPR.
[16] Tan, K. and Oakley, J.P. 2000. Enhancement of color images in poor visibility conditions. ICIP, 2.
[17] Beard, K.V. and Chuang, C.H. 1987. A new model for the equilibrium shape of raindrops. Journal of Atmospheric Science, 44(11):1509–1524.
[18] Kubesh, R.J. and Beard, K.V. 1993. Laboratory measurements of spontaneous oscillations of moderate size raindrops. Journal of the Atmospheric Sciences, 50:1089–1098.
[19] Tokay, A. and Beard, K.V. 1996. A field study of raindrop oscillations. part I. Journal of Applied Meteorology, 35.
[20] Gunn, R. and Kinzer, G.D. 1949. Terminal velocity for water droplet in stagnant air. Journal of Meterology, 6:243–248.
[21] Manning, R.M. 1993. Stochastic Electromagnetic Image Propogation. McGraw-Hill, Inc.
[22] K. Garg and S.K. Nayar, “When does a camera see rain?,” in Proc. ICCV 2005., vol. 2, pp. 1067–1074.
[23] Andsager, K., Beard, K.V., and Laird, N.F. 1999. Laboratory measurements of axis ratios for large raindrops. Journal of the Atmospheric Sciences, 56:2673–2683.
[24] Bradley, S.G., StowC.D., and Lynch-Blosse, C.A. 2000. Measurements of rainfall properties using long optical path imaging. American Meteorological Society, 17:761–772.
[25] Loffler-Mang, M. and Joss, J. 2000. An optical disdrometer for measuring size and velocity of hydrometeors. Journal of Atmospheric and Oceanic Technology, 17:130–139.
[26] Marshall, J.S. and Palmer,W.M.K. 1948. The distribution of raindrops with sizes. Journal of Meterology, 5:165–166.
[27] Schonhuber, M., et al. 1994. Measurements of precipitation characteristics of new disdrometer. In Proceedings of Atmospheric Physics and Dynamics in the Analysis and Prognosis of Precipitation Fields.
[28] Wang, T. and Clifford, S.F. 1975. Use of rainfall-induced optical scintillations to measure path-averaged rain parameters. Journal of the Optical Society of America, 8:927–937.
[29] Flori, J.P. 1990. Mouillage et Schage D’une Facade Verticale: Analyse Experimentale. Technical Report EN-CLI 90.2 L, Centre Scientifique et Technique du B, timen, 1990.
[30] Habib, E., Krajewski, W.F., and Kruger, A. 2001. Sampling errors of tipping-bucket rain gauge measurements. Journal of Hydrology Engeering, 6:159.
[31] Borovoy, A.G., KabanovM.V., and Saveliev, B.A. 1975. Intensity fluctuations of optical radiation in scattering media. Applied Optics, 14(11):2731–2739.
[32] Chu, T.S. and Hogg, D.C. 1996. Effects of precipitation on propagation at 0.63, 3.5 and 10.6 microns. Bell System Technical Journal.
[33] Deepak, A. and Box, M.A. 1978. Forward scattering corrections for optical extinction measurements in aerosol media. 2: Poly-Dispersions. Applied Optics, 17:3169–3176.
[34] Choi, E. 1995. Numerical modelling of gust effect on wind-driven rain. In 9th International Conference onWind Engineering, vol. 9, no. 13, pp. 214–215.
[35] Dorsey, J., Pedersen, H., and Hanrahan, P. 1996. Flow and changes in appearance. In Proceedings of Siggraph, vo. 79, no. 30.
[36] Mulvin, L., and Lewis, J. 1994. Architectural detailing, weathering and stone decay. Building and Environment, 29(1):113–138 .
[37] Langer, M.S., et al. 2004. A spectral-particle hybrid method for rendering falling snow. In Rendering Techniques, pp. 217–226.
[38] Lomas, A. 2005. The Matrix Revolutions. Personal Communication.
[39] Starik, K. and Werman, M. 2003. Simulation of rain in videos. Texture Workshop, ICCV.
[40] Wang, N. and Wade, B. 2004. Rendering falling rain and snow. SIGGRAPH (sketches 0186).
[41] Garg, K. and Nayar, S.K. 2006. Photorealistic rendering of rain streaks. ACM Trans. on Graphics (also Proc. of ACM SIGGRAPH), 25:996–1002.
[42] Horn, B.K.P. 1986. Robot Vision. The MIT Press.
[43] Website: http://www.comp.nus.edu.sg/~cs4243/demo/rainremoval/rainremoval.html
[44] K.A. Patwardhan and G. Sapiro, “Projection based image and video inpainting using wavelets,” in Proc. ICIP 2003., vol. 1, pp. 857–860.
[45] K.A. Patwardhan, G. Sapiro, and M. Bertalmio, “Video inpainting of occluding and occluded objects,” in Proc. ICIP 2005., vol. 2, pp. 69–72.
[46] Zeidenberg, Matthew, “Neural Networks in Artificial Intelligence, ” Ellis Horwood Limited. 1990., ISBN 0-13-612185-3
[47] 蘇木春、張孝德，機器學習：類神經網路、模糊系統以及基因演算法則，全華科技圖書股份有限公司，初版，1997，二版，1999，三版，2004。
[48] desJardins, M., and Gordon, D.F. (1995). Evaluation and selection of biases in machine learning. Machine Learning Journal, 5:1--17, 1995.
[49] Mitchell, T.M. (1980). The need for biases in learning generalizations. CBM-TR 5-110, Rutgers University, New Brunswick, NJ.
[50] Utgoff, P.E. (1984). Shift of bias for inductive concept learning. Doctoral dissertation, Department of Computer Science, Rutgers University, New Brunswick, NJ.
[51] Zoran Zivkovic, “Improved Adaptive Gaussian Mixture Model for Background Subtraction,” In Proc. ICPR, 2004., vol. 2, pp. 28–31.
[52] Tianci Huang, et al. “Motion Detection Based On Background Modeling And Performance Analysis For Outdoor Surveillance,” IEEE Trans. on ICCM, 2009.
[53] Jacinto C. Nascimento and Jorge S. Marques. “Performance Evaluation of Object Detection Algorithms for Video Surveillance, ” IEEE Trans. on multimedia, VOL. 8, NO. 4, 2006.

指導教授

范國清(Kuo-chin Fan)

審核日期

2010-7-27

推文