(18.207.240.230) 您好!臺灣時間:2020/07/09 10:31
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
本論文永久網址: 
line
研究生:馬依霖
研究生(外文):Nur Maulidiyah
論文名稱(外文):Analysis of the Performance of Different Classifiers for Cloud Detection Application
指導教授:鄭旭詠
指導教授(外文):Hsu-yung Cheng
學位類別:碩士
校院名稱:國立中央大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文出版年:2014
畢業學年度:102
語文別:英文
論文頁數:79
中文關鍵詞:云检测分类支持向量机随机森林分类回归树
外文關鍵詞:Cloud detectionClassifierSupport Vector MachineRandom ForestClassification And Regression Tree
相關次數:
  • 被引用被引用:0
  • 點閱點閱:109
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:5
  • 收藏至我的研究室書目清單書目收藏:0
云检测是一种提供必要的信息,如云层覆盖在许多应用中非常重要。的经典方法云检测是基于图像像素的红色和蓝色的比率的阈值。然而,很难选择合适的阈值对所有云的条件。另外,对于不同的全天空照相机所需的阈值是不同的。准确的云检测是一项具有挑战性的任务,因为云的颜色有时很容易地与天空或太阳的区域相混淆。
在这篇论文中,我们提出利用监督学习技术来进行云检测。云模型是通过培训过程中的经验教训。有许多分类器可以被用于此目的。我们认为,流行的分类,包括随机森林,分类回归树,和支持向量机。我们使用本地图像补丁,而不是只使用一个像素值的颜色信息。在一个本地图像补丁的像素的颜色值被设置为一个特征向量。该结果表明,该云检测使用本地图像补丁的颜色信息获得比使用一个像素值更好的精度。该结果还表明,支持向量机(SVM)具有最高的检测精度。采取由多个分类器提供的线索的优势,我们提出以投票方式相结合的多分类,以进一步提高检测精度。在实验中,我们已经表明,该方法能更准确区分云和非云像素。

Cloud detection is important for providing necessary information such as cloud cover in many applications. The classic method for cloud detection is based on thresholding of the red and blue ratio of an image pixel. However, it is difficult to select a suitable threshold for all cloud conditions. Also, the desired thresholds for different all-sky cameras are different. Accurate cloud detection is a challenging task since the colors of cloud sometimes easily be confused with sky or sun regions.
In this thesis, we propose to perform cloud detection using supervised learning techniques. The cloud models are learned through the training process. There are many classifiers that can be used for this purpose. We consider popular classifiers including random forest, classification and regression tree, and support vector machine. We use the color information of a local image patch instead of using only one pixel value. The color values of the pixels in a local image patch are arranged as a feature vector. The results show that the cloud detection using the color information of a local image patch get better accuracy than using one pixel value. The results also show that the support vector machine (SVM) has the highest detection accuracy. To take advantage of the clues provided by multiple classifiers, we propose a voting process to combine multiple classifiers to further increase the detection accuracy. In the experiments, we have shown that the proposed method can distinguish cloud and non-cloud pixels more accurately.
摘要 ......................................................................................................................................................... i
ABSTRACT........................................................................................................................................... ii
ACKNOWLEDGEMENT................................................................................................................... iii
LIST OF FIGURES ............................................................................................................................. iv
LIST OF TABLES ............................................................................................................................... vi
CHAPTER I ..........................................................................................................................................1
INTRODUCTION.............................................................................................................................1
1.1 Motivation................................................................................................................................1
1.2 Related Works.........................................................................................................................2
1.3 Thesis Organization................................................................................................................3
CHAPTER II.........................................................................................................................................4
REVIEW OF RELEVANT TECHNIQUES...................................................................................4
2.1 Cloud Detection.......................................................................................................................4
2.2 Thresholding Technique Using Red Blue Ratio ...................................................................5
2.2. Classification And Regression Tree (CART).....................................................................7
2.3. Random Forest...................................................................................................................11
2.4. Support Vector Machine (SVM).......................................................................................13
CHAPTER III.....................................................................................................................................18
PROPOSED SYSTEM ...................................................................................................................18
3.1 Features used.........................................................................................................................18
3.2 Classifier Cloud Detection...................................................................................................19
CHAPTER 4........................................................................................................................................25
DATA ANALYSIS AND RESULTS .............................................................................................25
4.1 System Environment and Device .........................................................................................25
4.2 Dataset....................................................................................................................................25
4.3 Accuracy, Recall, Precision, Miss Rate , and False Alarm Ratio .....................................28
4.4 Experimental results Using Red Blue Ratio (RBR) thresholding method.......................30
4.5 Experimental Results Using Classification And Regression Tree (CART)......................32
4.6 Experimental Results Using Random Forest......................................................................35
4.7 Experimental Results Using Support Vector Machine (SVM) .........................................37
4.8 Discussion of experimental results.......................................................................................40
4.9 Experimental Results Using Multiple Classifiers (Voting)...............................................45
CHAPTER 5........................................................................................................................................48viii
CONCLUSION ...................................................................................................................................48
REFERENCES....................................................................................................................................49
APPENDIX..........................................................................................................................................51
The result of cross validation C and gamma ................................................................................51
Tree of CART..................................................................................................................................51
Random Forest Tree .......................................................................................................................52
Ali J., Khan R., Ahmad N., Maqsood I. “Random Forests and Decision Trees.” International Journal of Computer Science Issues., 2012.
Amit Y., Geman D. “Shape quantization and recognition with randomized trees.” Neural Computation 9 (August 1996): 1545-1588.
Benhur A., Weston J. “A User’s guide to Support Vector Machines.” Biological Data Mining, 2009.
Breiman L., Friedman J.H. Classification and regression trees. Monterey, CA: Wadsworth &; Brooks/Cole Advanced Books &; Software, 1984.
Breiman, L. “Bagging predictors.” Machine Learning 24 (1996): 123-140.
—. Using adaptive bagging to debias regressions. Statistics Dept. UCB, 1999.
Buch, K. A., Sun J., Thorne L. R. "Cloud classification using whole-sky imager data." Proc. Fifth Atmospheric Radiation Measurement Science Team Meeting. San DIego, 1995. 35–39.
Calbo J., Sabburg J. "Feature extraction from whole-sky ground-based images for cloud-type recognition." (Journal of Athmoshpheric and Oceanic Technology) 25 (2008).
Dietterich, T.G. “An Experimental Comparison of Three Methods Constructing Ensembles of Decision Trees: Bagging, Boosting and Randomization.” Machine Learning, 1999: 1-22.
Duncan D. L., Donoho M. R. “Digital curvelet transform: Strategy, implementation and experiments.” Proc. SPIE. 2000. 12-29.
Heinle A., Macke A. “Automatic cloud classification of whole sky Images.” Atmospheric Measurement Technique, 2010: 557-567.
Ho, T. “The random subspace method for constructing decision forests.” IEEE Trans. on Pattern Analysis and Machine Intelligence, 1998: 832-844.
Johnson R.W., Koehler T.L. “A multi-station set of whole sky imagers and a preliminary assessment of the emerging data base.” Proc. Cloud Impacts on DOD Operations and Systems, 1988: 159-162.
Koehler T.L., Johnson R.W.,Shields J.E. “Status of The Whole Sky Imager Database.” Proceedings of the Cloud Impacts on DOD Operations and Systems, 1991.
Kohavi R., Provost F. “Glossary of terms: Special issue on applications of machine learning and the knowledge discovery process.” MAchine Learning 30, 1998: 271-274.
Lewis, R. “An Introduction to Classification and Regression Tree (CART) Analysis.” Annual Meeting of the Society for Academic Emergency Medicine. San Francisco, California, 2000.
Li Q., Lu W., Yang J. “A Hybrid Thresholding Algorithm for Cloud Detection on Ground-Based.” Journal of Athmoshperic and Oceanic Technology, 2011: 1286-1296.
Long C. N., Sabburg J. M., Calbo J., Pages D. "Retrieving cloud characteristics from ground-based daytime color all-sky images." (Journal of Athmoshpheric and Oceanic Technology) 23 (2006).
Luiz S., Wangenheinmen A.V., Bueno E. “The Use of Euclidean Geometric Distance on RGB Color Space for the Classification of Sky and Cloud Patterns.” Journal of Athmoshperic and Oceanic Technology, 2010: 1504-1517.
Oshiro T.M., Perez P.S., Baranauskas J.A. “How Many Trees in a Random Forest?” Springer-Verlag Berlin Heidelberg (Department of Computer Science and Mathematics Faculty of Philosophy, Sciences and Languages at Ribeirao Preto), 2012: 154-168.
Quinlan, J. R. “Induction of Decision Trees.” Machine Learning 1 (Kluwer Academic Publishers), 1986: 81-106.
Rokach L., Maimon O. Data mining with decision trees: theory and applications. Singapore: World Scientific Pub Co Inc, 2008.
Shawe-Taylor J., Cristianini N. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods. Cambridge: Cambridge University Press, 2000.
Slater, D.W., Long C. N. , Tooman T. P. "Total sky imager whole sky imager cloud fraction comparison." Atlanta: Proc. 11th ARM Science Team Meeting Proc, 2001.
Souza-Echer M.P., Pereira E. B., Bins L. S. , Andrade M.A.R. “A Simple Method for the Assessment of the Cloud Cover State in High-Latitude Regions by a Ground-Based Digital Camera.” Journal of Athmoshperic and Oceanic Technology 23 (2006): 437-447.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊
 
系統版面圖檔 系統版面圖檔