局部二值模式(local binary pattern, LBP) 是電腦視覺領域中最常用的視覺特徵之一，它在物件偵測的架構中扮演很重要的角色。本論文針對LBP設計出對應的近似演算法，使得擷取LBP特徵這個步驟可較容易透過硬體架構進行加速，因此即便在高解析度影像中，也能有效率地擷取對應的LBP特徵，進而達到即時的物件辨識。此外，LBP特徵的擷取由兩個參數(R與P)分別控制每個像素點的特徵擷取範圍圓半徑以及圓上的取樣個數，各種組合的參數在不同的應用中會有不同的效果，因此本論文設計了一套可調變參數的LBP硬體架構，並以台積電0.18微米製程來實現此硬體。相較於現存的LBP硬體，本論文所設計出的LBP硬體使用要少的閘數(gate count)並且可達到500 MHz的效能。為了驗證使用近似演算法設計出的LBP硬體不影響實際的物件偵測準確度，本論文將所取出的LBP特徵套用於行人偵測應用，使用支持向量機對所取出的傳統LBP特徵與近似的LBP特徵進行分類後，證實兩者可達到類似的分類準確度。

In the field of computer vision, local binary pattern (LBP) is one of the most popular feature extraction method and has been used in many object detection frameworks. To efficiently extract LBP features in high-resolution images, a hardware architecture is needed to disperse CPU burden and to improve the entire object detection performance. In this thesis, a hardware implementation of an approximated LBP method with adjustable parameters is introduced. For simulation, Taiwan Semiconductor Manufacturing Company 0.18} micrometer technology is used to implement the LBP hardware, in which the hardware can achieve 500 MHz with lower gate count than the previous study. The proposed LBP circuit is applied to the pedestrian classification application and the evaluation results show that the approximated LBP values generated by our circuit can achieve comparable classification accuracy with the primitive LBP method. Additionally, the proposed LBP hardware provides adjustable parameters to fit different applications while requires fewer hardware costs as compared with the existing work.

[1] P. Y. Chen, C. C. Huang, C. Y. Lien, and Y. H. Tsai, “An efficient hardware implementation of hog feature extraction for human detection,” IEEE Trans. Intell. Transp. Syst., vol. 15, no. 2, pp. 656–662, April 2014.
[2] D. G. Lowe, “Distinctive image features from scale-invariant keypoints,” Int. J. Comput. Vision, vol. 60, no. 2, pp. 91–110, 2004.
[3] J. Yum, C. H. Lee, J. S. Kim, and H. J. Lee, “A novel hardware architecture with reduced internal memory for real-time extraction of sift in an hd video,” IEEE Trans. Circuits Syst. Video Technol., vol. 26, no. 10, pp. 1943–1954, Oct 2016.
[4] H.Cheng,N.Zheng,andJ.Qin,“Pedestriandetectionusingsparsegaborfilter and support vector machine,” in Proc. IEEE Intell. Veh. Symp., June 2005, pp. 583–587.
[5] M. Oren, C. Papageorgiou, P. Sinha, E. Osuna, and T. Poggio, “Pedestrian detection using wavelet templates,” in Proc. IEEE Conf. Comp. Vis. Pattern Recog., Jun 1997, pp. 193–199.
[6] T. Ojala, M. Pietikainen, and T. Maenpaa, “Multiresolution gray-scale and ro- tation invariant texture classification with local binary patterns,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 24, no. 7, pp. 971–987, Jul 2002.
[7] T. Ahonen, A. Hadid, and M. Pietikainen, “Face description with local binary patterns: Application to face recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 12, pp. 2037–2041, Dec 2006.
[8] K. C. Fan and T. Y. Hung, “A novel local pattern descriptor: Local vector pattern in high-order derivative space for face recognition,” IEEE Trans. Image Process., vol. 23, no. 7, pp. 2877–2891, July 2014.
[9] D. Huang, C. Shan, M. Ardabilian, Y. Wang, and L. Chen, “Local binary pat- terns and its application to facial image analysis: A survey,” IEEE Trans. Syst., Man, Cybern. C, vol. 41, no. 6, pp. 765–781, Nov 2011.
[10] Y.Cao,S.Pranata,andH.Nishimura,“Localbinarypatternfeaturesforpedes- trian detection at night/dark environment,” in 18th IEEE Int. Conf. on Image Process, Sept 2011, pp. 2053–2056.
[11] F. Tajeripour and E. Kabir, “Defect detection in patterned fabrics using modi- fied local binary patterns,” in Int. Conf. Computational Intell. and Multimedia Applicat., vol. 2, Dec 2007, pp. 261–267.
[12] Y. Guo, G. Zhao, Z. Zhou, and M. Pietikäinen, “Video texture synthesis with multi-frame lbp-top and diffeomorphic growth model,” IEEE Trans. Image Process., vol. 22, no. 10, pp. 3879–3891, Oct 2013.
[13] M. Cote and A. B. Albu, “Robust texture classification by aggregating pixel- based lbp statistics,” IEEE Signal Process. Lett., vol. 22, no. 11, pp. 2102– 2106, Nov 2015.
[14] L. Nanni and A. Lumini, “Local binary patterns for a hybrid fingerprint matcher,” Pattern Recogn., vol. 41, no. 11, pp. 3461–3466, Nov. 2008. [Online]. Available: http://dx.doi.org/10.1016/j.patcog.2008.05.013
[15] S. Liao and A. C. S. Chung, “Texture classification by using advanced local binary patterns and spatial distribution of dominant patterns,” in IEEE ICASSP, vol. 1, April 2007, pp. I–1221–I–1224.
[16] P. Hurney, P. Waldron, F. Morgan, E. Jones, and M. Glavin, “Night-time pedestrian classification with histograms of oriented gradients-local binary patterns vectors,” IET Intell. Transp. Syst., vol. 9, no. 1, pp. 75–85, 2015.
[17] J. Boutellier, I. Lundbom, J. Janhunen, J. Ylimäinen, and J. Hannuksela, “Application-specific instruction processor for extracting local binary pat- terns,” in Conf. on Design and Architectures for Signal and Image Process. (DASIP), Oct 2012, pp. 1–8.
[18] N. J. Wang, S. C. Chang, and P. J. Chou, “A real-time multi-face detection system implemented on fpga,” in Int. Symp. on Intell. Signal Process. and Commun. Syst. (ISPACS), Nov 2012, pp. 333–337.
[19] O. Lahdenoja, M. Laiho, and A. Paasio, “Reducing the feature vector length in local binary pattern based face recognition,” in IEEE Int. Conf. on Image Process 2005, vol. 2, Sept 2005, pp. II–914–17.
[20] W. Ge, W. Quan, and C. Han, “Face description and identification using his- togram sequence of local binary pattern,” in Seventh Int. Conf. on Advanced Computational Intell. (ICACI), March 2015, pp. 415–420.
[21] J. Janhunen, P. Salmela, O. Silvén, and M. Juntti, “Fixed- versus floating- point implementation of mimo-ofdm detector,” in IEEE Int. Conf. on Acoust., Speech and Signal Processing (ICASSP), May 2011, pp. 3276–3279.
[22] G. Zhao, T. Ahonen, J. Matas, and M. Pietikainen, “”rotation-invariant image and video description with local binary pattern features”,” IEEE Trans. Image Process., vol. 21, no. 4, pp. 1465–1477, April 2012.
[23] Y. Fang, J. Luo, and C. Lou, “Fusion of multi-directional rotation invariant uniform lbp features for face recognition,” vol. 2, Nov 2009, pp. 332–335.
[24] N. Dalal and B. Triggs, “Histograms of oriented gradients for human de- tection,” in IEEE Comput. Soc. Conf. on Comput. Vis. and Pattern Recogn. (CVPR’05), vol. 1, June 2005, pp. 886–893 vol. 1.
[25] M. C. Hu, K. S. Ng, P. Y. Chen, Y. J. Hsiao, and C. H. Li. Datasets used in local binary pattern circuit generator with adjustable parameters for feature extraction. [Online]. Available: https://goo.gl/KvYB4d
[26] C.-C. Chang and C.-J. Lin, “Libsvm: A library for support vector machines,” ACM Trans. Intell. Syst. Technol., vol. 2, no. 3, pp. 27:1–27:27, may 2011. [Online]. Available: http://doi.acm.org/10.1145/1961189.1961199