用於即時交通標誌辨識的高資源效能之硬體實作

簡易檢索 / 詳目顯示

回結果列表

研究生：	翁槐懋 Weng, Hui-Mao
論文名稱：	用於即時交通標誌辨識的高資源效能之硬體實作 Resource Efficient Hardware Implementation For Real-Time Traffic Sign Recognition
指導教授：	許雅三 Hsu, Yar-Sun
口試委員:	邱瀞德 Chiu, Ching-Te 鐘太郎 Jong, Tai-Lang
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2017
畢業學年度：	106
語文別：	英文
論文頁數：	74
中文關鍵詞：	路標辨識
外文關鍵詞：	Traffic sign recognition
相關次數：	點閱：1 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

近年來，車載電子成為一個熱門的領域。其中的駕駛輔助系統 (Advanced Driver Assistance System, ADAS) 更是現代車輛中常見的配備。交通標誌辨識則是駕駛輔助系統其中一項可以改進的技術。
針對交通標誌辨識中最重要的議題，也就是即時性與節約資源，本論文提出一個高效的硬體架構以達到這些目標。即時性是任何車載系統的首要條件，只有速度夠快才得以應對道路上多變的情況。另一方面，一個設計良好的系統必須節約能源的使用並有較低的成本，也就是必須節省資源。
在本論文中，交通標誌辨識採用兩階段的方法，分別為偵測與識別。在交通標誌偵測的階段，我們採用色域標準化轉換 (Normalized RGB Transform)，再以單次連通性分析 (Single-Pass Connected Component Labeling) 尋找可能為標通標誌的區域。在交通標誌識別的階段，我們採用方向梯度直方圖 (Histogram of oriented gradient, HOG) 得出區域的特徵向量，再以支持向量機 (Support Vector Machine, SVM) 進行分類識別。此交通標誌辨識方法使用GTSDB的資料進行測試，得到96.61%的偵測率與90.85%的辨識率。
在硬體實作上，我們提出的架構節省了連通性分析所需的記憶體大小、簡化了方向梯度直方圖的計算複雜度。於典型條件下，CCL的主要記憶體比最先進的設計減少20%的大小。使用TSMC 90nm的製程下，提出的硬體操作頻率為105MHz，每秒可處理135張1360x960大小的影像。晶片面積約為1mm2且功耗僅約8mW，因此達到了即時性與節約資源的要求。

Recently, car electronic systems become more popular. Advanced Driver Assistance System (ADAS) is a common device in modern cars. Traffic sign recognition (TSR, or Road Sign Recognition, RSR) is one of the ADAS techniques that can be improved.
To concern the most important issues, which are real-time and computational resources, we propose a high efficiency hardware implementation for traffic sign recognition. Real-time is a restriction for any car electronic systems. Only fast systems are able to react to the changing traffic conditions. Furthermore, a well-designed system should be energy saving as well as low cost, which means resource efficiency.
In this work, we divide the TSR into two stages, detection and recognition. In the detection stage, we use Normalized RGB color transform and Single-Pass Connected Component Labeling (CCL) to find the potential traffic signs. In the recognition stage, Histogram of Oriented Gradient (HOG) is used to generate the descriptor of the signs, and we classify the signs with Support Vector Machine (SVM). The proposed method achieves 96.61% detection rate and 90.85% recognition rate while testing with GTSDB dataset.
Our hardware implementation reduces the storage of CCL, and simplifies the HOG computation. Main CCL storage size is reduced by 20% comparing to the most advanced design under typical condition. By using TSMC 90nm technology, the proposed design operates at 105 MHz clock rate and processes in 135 fps with image size of 1360x960. The chip size is about 1mm2 and power consumption is close to 8mW. Therefore, this work is resource efficient and achieves real-time requirement.

Chapter 1    Introduction    1
1    Motivation    1
2    Goals and Contributions    2
3    Thesis Organization    3
Chapter 2    Related Work    4
1    Overview    4
2    Image-Processing (Traditional Computer-Vision) Based Method    5
3    Machine Learning Based Method    6
4    Hybrid Method    7
Chapter 3    Preliminary    9
1    Color Segmentation    9
2    Connected Component Labeling (CCL)    13
2.1    Traditional Two-Pass Algorithm    14
2.2    Single-Pass Algorithm    15
3    Histogram of Oriented Gradient Descriptor (HOG)    21
4    Support Vector Machine Classification (SVM)    24
Chapter 4    Proposed method    26
1    Detection Procedure    27
1.1    Reducing the Searching Space    27
1.2    Improved Color Segmentation    28
1.3    Single-Pass Connected Component Labeling    29
2    Recognition Procedure    30
2.1    Preparing the ROIs    30
2.2    Simplification of HOG Calculation    30
2.3    SVM Classification    32
3    Performance Evaluation    33
3.1    Testing Dataset and Environment Settings    33
3.2    The Influence of Color Segmentation Thresholds    35
3.3    The Influence of Reducing HOG Calculation Bits    37
3.4    Time Consumption of Software Implementation    40
Chapter 5    Hardware Implementation    41
1    Overall Processing Flow    41
2    TSR Top Module    43
3    Detection Module    44
3.1    NRB Transform Module    46
3.2    Connected Component Labeling Module    47
4    Recognition Module    55
4.1    Histogram of Gradient Module    56
4.2    Specification and Cycles for Recognition Module    59
5    Timing and Speed of Hardware Implementation    61
6    Implementation Result    62
7    Comparison with Other Works    63
7.1    Comparison of CCL Hardware Implementation    63
7.2    Comparison of TSR Hardware Implementation    67
Chapter 6    Conclusions and Future Work    69
1    Conclusions    69
2    Future Work    70
Bibliography    71

                                

[1] A. Mammeri, A. Boukerche and M. Almulla, "Design of traffic sign detection, recognition, and transmission systems for smart vehicles," in IEEE Wireless Communications, vol. 20, no. 6, pp. 36-43, December 2013.
[2] H. Gomez-Moreno, S. Maldonado-Bascon, P. Gil-Jimenez and S. Lafuente-Arroyo, "Goal Evaluation of Segmentation Algorithms for Traffic Sign Recognition," in IEEE Transactions on Intelligent Transportation Systems, vol. 11, no. 4, pp. 917-930, Dec. 2010.
[3] Á. Gonzalez et al., "Automatic Traffic Signs and Panels Inspection System Using Computer Vision," in IEEE Transactions on Intelligent Transportation Systems, vol. 12, no. 2, pp. 485-499, June 2011.
[4] M. Liang, M. Yuan, X. Hu, J. Li and H. Liu, "Traffic sign detection by ROI extraction and histogram features-based recognition," The 2013 International Joint Conference on Neural Networks (IJCNN), Dallas, TX, 2013, pp. 1-8.
[5] J. F. Khan, S. M. A. Bhuiyan and R. R. Adhami, "Image Segmentation and Shape Analysis for Road-Sign Detection," in IEEE Transactions on Intelligent Transportation Systems, vol. 12, no. 1, pp. 83-96, March 2011.
[6] A. Ruta, Y. Li, and X. Liu, “Real-Time Traffic Sign Recognition from Video by Class-Specific Discriminative Features,” Pattern Recognition, vol. 43, no. 1, pp. 416-430, 2010.
[7] Overett, G., Tychsen-Smith, L., Petersson, L. et al., “Creating robust high-throughput traffic sign detectors using centre-surround HOG statistics”, Machine Vision and Applications, vol. 25, no. 3, pp. 713-726, 2014.
[8] Y. W. Seo, J. Lee, W. Zhang and D. Wettergreen, "Recognition of Highway Workzones for Reliable Autonomous Driving," in IEEE Transactions on Intelligent Transportation Systems, vol. 16, no. 2, pp. 708-718, April 2015.
[9] K. Zhang, Y. Sheng and J. Li, "Automatic detection of road traffic signs from natural scene images based on pixel vector and central projected shape feature," in IET Intelligent Transport Systems, vol. 6, no. 3, pp. 282-291, September 2012.
[10] J. Jin, K. Fu and C. Zhang, "Traffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks," in IEEE Transactions on Intelligent Transportation Systems, vol. 15, no. 5, pp. 1991-2000, Oct. 2014.
[11] J. Greenhalgh and M. Mirmehdi, "Real-Time Detection and Recognition of Road Traffic Signs," in IEEE Transactions on Intelligent Transportation Systems, vol. 13, no. 4, pp. 1498-1506, Dec. 2012.
[12] Z. Chen, X. Huang, Z. Ni and H. He, "A GPU-based real-time traffic sign detection and recognition system," 2014 IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems (CIVTS), Orlando, FL, 2014, pp. 1-5.
[13] Fatin Zaklouta, Bogdan Stanciulescu, “Real-time traffic sign recognition in three stages,” Robotics and Autonomous Systems, vol. 62, no. 1, pp.16-24, 2014.
[14] X. Yuan, X. Hao, H. Chen and X. Wei, "Robust Traffic Sign Recognition Based on Color Global and Local Oriented Edge Magnitude Patterns," in IEEE Transactions on Intelligent Transportation Systems, vol. 15, no. 4, pp. 1466-1477, Aug. 2014.
[15] Y. Yang, H. Luo, H. Xu and F. Wu, "Towards Real-Time Traffic Sign Detection and Classification," in IEEE Transactions on Intelligent Transportation Systems, vol. 17, no. 7, pp. 2022-2031, July 2016.
[16] C. Liu, F. Chang, Z. Chen and D. Liu, "Fast Traffic Sign Recognition via High-Contrast Region Extraction and Extended Sparse Representation," in IEEE Transactions on Intelligent Transportation Systems, vol. 17, no. 1, pp. 79-92, Jan. 2016.
[17] S. Houben, J. Stallkamp, J. Salmen, M. Schlipsing and C. Igel, "Detection of traffic signs in real-world images: The German traffic sign detection benchmark," The 2013 International Joint Conference on Neural Networks (IJCNN), Dallas, TX, 2013, pp. 1-8.
[18] A. Rosenfeld and J. L. Pfaltz, "Sequential operations in digital picture processing," Journal of the ACM, Vol. 13, pp. 471-494, Oct. 1966.
[19] A. AbuBaker, R. Qahwaji, S. Ipson and M. Saleh, "One Scan Connected Component Labeling Technique," 2007 IEEE International Conference on Signal Processing and Communications, Dubai, 2007, pp. 1283-1286.
[20] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), San Diego, CA, USA, 2005, pp. 886-893 vol. 1.
[21] C. Cortes, V. Vapnik, “Support-vector networks”, Machine Learning, vol. 20, no. 3, pp. 273-297, 1995.
[22] The OpenCV library; http://opencv.org.
[23] Chang, Chih-Chung and Lin and Chih-Jen, “LIBSVM : a library for support vector machines”, ACM Transactions on Intelligent Systems and Technology, vol. 2, no. 3, pp. 27:1-27:27, 2011.
[24] X. D. Yang, "Design of fast connected components hardware," Computer Vision and Pattern Recognition, 1988. Proceedings CVPR '88., Computer Society Conference on, Ann Arbor, MI, 1988, pp. 937-944.
[25] C. T. Johnston and D. G. Bailey, "FPGA implementation of a Single Pass Connected Components Algorithm," 4th IEEE International Symposium on Electronic Design, Test and Applications (delta 2008), Hong Kong, 2008, pp. 228-231.
[26] N. Ma, D. G. Bailey and C. T. Johnston, "Optimised single pass connected components analysis," 2008 International Conference on Field-Programmable Technology, Taipei, 2008, pp. 185-192.
[27] M. J. Klaiber, D. G. Bailey, Y. O. Baroud and S. Simon, "A Resource-Efficient Hardware Architecture for Connected Component Analysis," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 26, no. 7, pp. 1334-1349, July 2016.
[28] Chokri Souani, Hassene Faiedh and Kamel Besbes, “Efficient algorithm for automatic road sign recognition and its hardware implementation,” Journal of Real-Time Image Processing, vol. 9, no. 1, pp. 79-93, March 2014.
[29] S. Waite and E. Oruklu, "FPGA-Based Traffic Sign Recognition for Advanced Driver Assistance Systems," Journal of Transportation Technologies, vol. 3 no. 1, pp. 1-16, 2013.
[30] N. Aguirre-Dobernack, H. Guzmán-Miranda and M. A. Aguirre, "Implementation of a machine vision system for real-time traffic sign recognition on FPGA," IECON 2013 - 39th Annual Conference of the IEEE Industrial Electronics Society, Vienna, 2013, pp. 2285-2290.
[31] J. Zhao, B. Thörnberg, Y. Shi and A. Hashemi, "Color segmentation on FPGA using minimum distance classifier for automatic road sign detection," 2012 IEEE International Conference on Imaging Systems and Techniques Proceedings, Manchester, 2012, pp. 516-521.
[32] Y. Zhou, Z. Chen and X. Huang, "A pipeline architecture for traffic sign classification on an FPGA," 2015 IEEE International Symposium on Circuits and Systems (ISCAS), Lisbon, 2015, pp. 950-953.
[33] J. Park, J. Kwon, J. Oh, S. Lee, J. Y. Kim and H. J. Yoo, "A 92-mW Real-Time Traffic Sign Recognition System With Robust Illumination Adaptation and Support Vector Machine," in IEEE Journal of Solid-State Circuits, vol. 47, no. 11, pp. 2711-2723, Nov. 2012.

簡易檢索 / 詳目顯示

相關論文