臺灣博碩士論文加值系統

在這篇論文中，我們提出了兩個物體偵測方法，一個是對固定不變的影像樣式做搜尋，主要用於工業檢測中的自動影像檢測定位，另一個是無固定形狀的物體搜尋，我們把重點放在人臉偵測上。
在工業檢測上，我們發展一個能在不平均光源下做影像定位的演算法，在這個演算法中，我們先以邊緣偵測法取得影像中物體邊緣的強度，再取得區域中最大的邊緣強度值將整個影像中的邊緣強度做補償，所得到的影像稱為相對性的邊緣對映圖。我們的演算法中的兩個階段，以學習為基礎的樣式比對搜尋與以能量最小化為基礎的影像自動定位，皆使用由上述方法所取得的相對邊緣對映圖來做影像定位。我們所實做出來的系統除了擁有對光源不平均的強固性之外，另一優點為搜尋時非常快速。
我們的人臉偵測演算法分四個部份，分別是膚色過濾，人臉樣板比對，人臉確認與重疊人臉整合。在做偵測時，先利用色彩資訊將膚色的區域找出，再針對膚色的區域做搜尋。我們利用收集到的大量的人臉資料，取其影像特徵，再做群聚分類，每個群集的中心值向量即為我們的標準人臉樣板。將欲搜尋的影像區域取同樣的特徵向量，再與已訓練好的人臉樣板做比對，當兩者比對的值小於某個門檻值，此塊區域可能包含人臉。我們應用兩個步驟的人臉影像確認方法在可能包含人臉的影像區域上來做進一步的確認。最後，我們將重疊的人臉影像區塊做整合以得到最後的結果。

In this thesis, we proposed two new algorithms for object detection. One is the image localization algorithm for invariant or rigid objects, which are very common in automatic visual inspection. The other algorithm is used to detect variant objects. We focus on the problem of face detection.
In industrial inspection, we developed a new algorithm that can localize the object under non-uniform illuminations. In this algorithm, we first apply Sobel edge detection on an image to get the gradient map, and then we compute the local maximal gradient values to normalize the gradient map. The two stages of our alignment algorithm, i.e. the learning based pattern search and the energy minimization based alignment, perform alignment on this relative gradient map. The alignment system is not only robust against non-uniform illumination changes but also very efficient for image alignment.
Our face detection algorithm consists of four steps, i.e. the skin-color filtering, template matching, verification and overlap merging. We used skin-color information to extract skin-color region from input image, and then detected face on skin-color regions. Our face detection algorithm extract features from face images and classify the features into several clusters in the training process. The means of these clusters are face templates. Our system calculates the minimal distance between the feature vector extracted from an image region and the face templates to determine if it is a face candidate. We apply two verification steps on the detected face candidates. The first one is the complete face comparison verification and the second is the combination of criteria from local face region analysis. At last, we merge the overlapped face candidate regions to obtain the final results.

CHAPTER 1 INTRODUCTION 3
CHAPTER 2 PREVIOUS WORKS 6
CHAPTER 3 THE ALGORITHM OF IMAGE PATTERN LOCALIZATION 9
3.1 IMAGE MATCHING BASED ON RELATIVE GRADIENT 10
3.2 LEARNING-BASED PATTERN SEARCH 11
3.2.1 Sub-template Window 12
3.2.2 Feature Extraction 13
3.2.3 Transformation parameters 14
3.2.4 Hierarchical Nearest-Neighbor Network 14
3.2.5 Multi-Resolutional Search 16
3.3 ENERGY--MINIMIZATION BASED MATCHING 17
CHAPTER 4 THE ALGORITHM OF FACE DETECTION 19
4.1 OUTLINE 19
4.2 SKIN-COLOR FILTER 20
4.3. PATTERN SEARCH WITH HIERARCHICAL NEAREST NEIGHBOR NETWORK 22
4.3.1 Training Phase 23
4.3.2 Execution Phase 26
4.4 FACE VERIFICATION 27
4.4.1 Complete Face Comparison 28
4.4.2 Combination of Local Face Region Analysis 29
4.5 MERGING OVERLAPPING DETECTIONS 33
CHAPTER 5 EXPERIMENT RESULTS OF IMAGE PATTERN LOCALIZATION ALGORITHM 35
5.1 INTENSITY FEATURE 35
5.1.1 Rotation Experiments 35
5.1.2 Shift in x direction 37
5.2.3 Shift in y direction 38
5.2 RELATIVE GRADIENT FEATURE 40
5.2.1 Test 1: CROSS 41
5.2.2 Test 2: KeyPad 42
5.2.3 Test 3: PCB 43
CHAPTER 6 EXPERIMENT RESULTS OF FACE DETECTION ALGORITHM 44
6.1 TEST SET 45
6.2 THE INTERMEDIATE RESULTS OF FACE DETECTION 46
6.3 SOME RESULTS OF FACE DETECTION 48
6.4 FACE DETECTION MISTAKES 50
CHAPTER 7 CONCLUSION 52
REFERENCES 54

[1] R. T. Chin, “Automatic visual inspection: 1981 to 1987,” Computer Vision, Graphics, and Image Processing, Vol. 41, No. 3, pp. 346-381, 1988.
[2] T. S. Newman and A. K. Jain, “A survey of automatic visual inspection,” Computer Vision and Image Understanding, Vol. 61, No. 2, pp. 231-262, 1995.
[3] S. Manickam, S. D. Roth, and T. Bushman, “Intelligent and optimal normalized correlation for high-speed pattern matching”, Technical Report, Datacube Inc.
[4] S. Ando, “Image Field Categorization and Edge/Corner Detection from Gradient Covariance”, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, No. 2, February 2000, pp. 179-190.
[5] H. Wolfson and I. Rigoutsos, “Geometric hashing: an overview”, IEEE Magazine on Computational Science & Engineering, pp. 10-21, Oct. 1997.
[6] S.-H. Lai and M. Fang, “A FLASH system for fast and accurate pattern localization”, Proceedings of SPIE Conf. on Machine Vision Applications in Industrial Inspection VII, Vol. 3652, pp. 164-173, San Jose, California, Jan. 25-27, 1999.
[7] J. McNames, “A Fast Nearest-Neighbor Algorithm Based on a Principal Axis Search Tree”, IEEE Trans. Pattern Analysis Machine Intelligence, Vol. 23, No. 9, pp. 964-976, 2001.
[8] S. A. Nene and S. K. Nayar, “A Simple Algorithm for Nearest-Neighbor Search in High Dimensions”, IEEE Trans. Pattern Analysis Machine Intelligence, Vol. 19, No. 9, pp. 989-1003, 1997.
[9] S.-H. Lai “Robust image matching under partial occlusion and spatially varying illumination change”, Computer Vision Image Understanding, Vol. 78, pp. 84-98, 2000.
[10] G.A.F. Seber and C. J. Wild, Nonlinear Regression, John Wiley & Sons, 1989.
[11] Ming-Hsuan Yang, David Kriegman and Narendra Ahuja, ”Detecting Faces in Images: A Survey”, IEEE Trans. Pattern Analysis Machine Intelligence, vol. 24, no. 1, pp. 34-58, January 2002.
[12] Erik Hjelmas, “Face Detection: A Survey”, Computer Vision and Image Understanding 83, 236-274(2001)
[13] K.-K. Sung and T. Poggio, Example-based learning for view-based human face detection, IEEE Trans. Pattern Analysis Machine Inteilligence 20, 39-51, 1998.
[14] H. Rowley, S. Baluja, and T. Kanada “Neural Network-Based Face Detection” IEEE Trans. Pattern Analysis Machine Intelligence, vol. 20, no. 1, January 1998.
[15] P. Viola and M. Jones, Robust Real-time Object Detection, ICCV 2001 Workshop on Statistical and Computation Theories of Vision