Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications
Hardware implementation of a scale and rotation invariant object detection algorithm on FPGA for real-time applications
A hardware implementation of a computationally light, scale, and rotation invariant method for shape detection on FPGA is devised. The method is based on histogram of oriented gradients (HOG) and average magnitude difference function (AMDF). AMDF is used as a decision module that measures the similarity/dissimilarity between HOG vectors of an image in order to classify the object. In addition, a simulation environment implemented on MATLAB is developed in order to overcome the time-consuming and tedious process of hardware verification on the FPGA platform. The simulation environment provides specific tools to quickly implement the proposed methods. It is shown that the simulator is able to produce exactly the same results as those obtained from FPGA implementation. The results indicate that the proposed approach leads to a shape detection method that is computationally light, scale, and rotation invariant, and, therefore, suitable for real-time industrial and robotic vision applications.
___
- [1] Pedrino EC, Morandin O, Kato ERR, Roda VO. Intelligent FPGA based system for shape recognition. In: IEEE 2011 Southern Conference on Programmable Logic; 1315 April 2011; C´ordoba, Spain. New York, NY, USA: IEEE. pp. 197-202.
- [2] Chen LD, Zhang MJ, Xiong ZH. Series-parallel pipeline architecture for high-resolution catadioptric panoramic unwrapping. IET Image Process 2010; 4: 403-412.
- [3] Geninatti SR, Ben´ıtez JIB, Calvino MH, Mata NG, Luna JG. FPGA implementation of the generalized hough transform. In: IEEE 2009 International Conference on Reconfigurable Computing and FPGAs; 911 December 2009; Quintana Roo, Mexico. New York, NY, USA: IEEE. pp. 172-177.
- [4] Kim D, Jin S, Nguyen DD, Jeon JW. Real-time binary shape matching system based on FPGA. In: IEEE 2008 International Conference on Robotics and Biomimetics; 2225 February 2009; Bangkok, Thailand. New York, NY, USA: IEEE. pp. 1194-1199.
- [5] Karakaya F, Altun H, Cavuslu M. Implementation of HOG algorithm for real time object recognition applications on FPGA based embedded system. In: IEEE 2009 Conference on Signal Processing and Communications Applications; 911 April 2009; Antalya, Turkey. New York, NY, USA: IEEE. pp. 508-511.
- [6] Koo JJ, Evans AC, Gross WJ. 3-D brain MRI tissue classification on FPGAs. IEEE T Image Process 2009; 18: 2735-2746.
- [7] Gao C, Lu S-LL, Suh T, Lim H. Field programmable gate array-based Haar classifier for accelerating face detection algorithm. IEEE T Image Process 2010; 4: 184-194.
- [8] Miyoshi T, Shibata T. A hardware-friendly object detection algorithm based on variable-block-size directional-edge histograms. In: IEEE 2010 World Automation Congress; 1923 September 2010; Kobe, Japan. New York, NY, USA: IEEE. pp. 1-6.
- [9] Zhang L, Nevatia R. Efficient scan-window based object detection using GPGPU. In: IEEE 2008 Conference on Computer Vision and Pattern Recognition Workshops; 2328 June 2008; Anchorage, AK, USA. New York, NY, USA: IEEE. pp. 1-7.
- [10] Gu Q, Takaki T, Ishii I. Fast FPGA-based multiobject feature extraction. IEEE T Circuits Syst Vid 2013; 23: 30-45.
- [11] Kotoulas L, Andreadis I. Colour histogram content-based image retrieval and hardware implementation. IEEE T Circ Syst 2003; 150: 1350-2409.
- [12] Yeo JC, Guo JI. Efficient hierarchical chaotic image encryption algorithm and its VLSI realisation. IEEE T Signal Proces 2000; 147: 167-175.
- [13] Kadota R, Sugano H, Hiromoto M, Ochi H, Miyamoto R, Nakamura Y. Hardware architecture for HOG feature extraction. In: IEEE 2009 Conference on Intelligent Information Hiding and Multimedia Signal Processing; 1214 September 2009; Kyoto, Japan. New York, NY, USA: IEEE. pp. 1330-1333.
- [14] Peker M, Altun H, Karakaya F. Hardware emulation of HOG and AMDF based scale and rotation invariant robust shape detection. In: IEEE 2012 International Conference on Engineering and Technology; 1011 October 2012; Cairo, Egypt. New York, NY, USA: IEEE. pp. 1-5.
- [15] Arslan O, Demirci B, Altun H, Tunaboylu NS. A novel rotation-invariant template matching based on HOG and AMDF for industrial laser cutting applications. In: IEEE 2013 International Symposium on Mechatronics and its Applications; 911 April 2013; Amman, Jordan. New York, NY, USA: IEEE. pp. 1-5.
- [16] Peker M, Altun H, Karakaya F. Hardware emulation of HOG and AMDF based scale and rotation invariant robust shape detection. In: IEEE 2012 International Conference on Engineering and Technology; 1011 October 2012; Cairo, Egypt. New York, NY, USA: IEEE. pp. 1-5.
- [17] Shashua A, Gdalyahu Y, Hayun G. Pedestrian detection for driving assistance systems: single-frame classification and system level performance. In: IEEE 2004 Intelligent Vehicles Symposium; 1417 June 2004; Parma, Italy. New York, NY, USA: IEEE. pp. 1-6.
- [18] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: IEEE 2005 Computer Society Conference on Computer Vision and Pattern Recognition; 25 June 2005; San Diego, CA, USA. New York, NY, USA: IEEE. pp. 886-893.
- [19] Muhammad G. Extended average magnitude difference function based pitch detection. Int Arab J Inf Technol 2011; 8: 197-203.
- [20] Andraka R. A survey of CORDIC algorithms for FPGA based computers. In: ACM/SIGDA 1998 International Symposium on Field Programmable Gate Arrays; 2225 February 1998; Monterey, CA, USA. New York, NY, USA:ACM. pp. 191-200.