Histogram of Oriented Gradients with Cell Average Brightness for Human Detection

• Autorzy: Wójcikowski M.

Identyfikatory

DOI

10.1515/mms-2016-0012

• Języki publikacji: EN

Abstrakty

EN

A modification of the descriptor in a human detector using Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) is presented. The proposed modification requires inserting the values of average cell brightness resulting in the increase of the descriptor length from 3780 to 3908 values, but it is easy to compute and instantly gives ≈ 25% improvement of the miss rate at 10‒4 False Positives Per Window (FPPW). The modification has been tested on two versions of HOG-based descriptors: the classic Dalal-Triggs and the modified one, where, instead of spatial Gaussian masks for blocks, an additional central cell has been used. The proposed modification is suitable for hardware implementations of HOG-based detectors, enabling an increase of the detection accuracy or resignation from the use of some hardware-unfriendly operations, such as a spatial Gaussian mask. The results of testing its influence on the brightness changes of test images are also presented. The descriptor may be used in sensor networks equipped with hardware acceleration of image processing to detect humans in the images.

Słowa kluczowe

EN

digital image processing; object detection; human detection

• Wydawca: Komitet Metrologii i Aparatury Naukowej PAN
• Czasopismo: Metrology and Measurement Systems
• Rocznik: 2016
• Tom: Vol. 23, nr 1
• Strony: 27--36
• Opis fizyczny: Bibliogr. 22 poz., fot., rys., tab., wykr., wzory

Twórcy

autor

Wójcikowski M.

• Gdańsk University of Technology, Faculty of Electronics, Telecommunications and Informatics, G. Narutowicza 11/12, 80-233 Gdańsk, Poland

Bibliografia

• [1] Dalal, N., Triggs, B. (2005). Histograms of oriented gradients for human detection. Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., 886-893.
• [2] Viola, P., Jones, M. (2001). Rapid object detection using a boosted cascade of simple features. Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., 511-518.
• [3] Ma, Y., Chen, X., Jin, L., Chen, G. (2011). A monocular human detection system based on EOH and oriented LBP features. Proc. 7th Int. Conf. Adv. Visual Comput., I, 551-562.
• [4] Ma, Y., Deng, L., Chen, X., Guo, N. (2013). Integrating Orientation Cue With EOH-OLBP-Based Multilevel Features for Human Detection. IEEE Trans. Circuits Syst. Video Technol., 23(10), 1755‒1766.
• [5] Boser, B.E., Guyon, I., Vapnik, V. (1992). A training algorithm for optimal margin classifiers. Proc. Fifth Annual Workshop on Computational Learning Theory, ACM Press, 144‒152.
• [6] Cortes, C., Vapnik, V. (1995). Support-vector network. Machine Learning, 20, 273‒297.
• [7] Cui, J., Wang, Y., (2010). A novel approach of analog fault classification using a Support Vector Machines classifier. Metrol. Meas. Syst., 17(4), 561‒581.
• [8] Wojtowicz, B., Dobrowolski, A., Tomczykiewicz, K., (2015). Fall detector using discrete wavelet decomposition and SVM classifier. Metrol. Meas. Syst., 22(2), 303‒314.
• [9] Zhang, H., Bai, X., Zhou, J., Cheng, J., Zhao H. (2013). Object Detection via Structural Feature Selection and Shape Model. IEEE Trans. Image Process., 22(12), 4984‒4995.
• [10] Lowe, D.G. (2004). Distinctive Image Features from Scale-Invariant Keypoints. Int. Journal of Comput. Vision, 60(2), 91‒110.
• [11] Zhu, Q., Yeh, M.C., Cheng, K.T., Avidan, S. (2006). Fast human detection using a cascade of histograms of oriented gradients. Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., 2, 491‒1498.
• [12] Zeng, H.C., Huang, S.H., Lai, S.H. (2008). Real-time video surveillance based on combining foreground extraction and human detection. Proc. 14th Int. Multimedia Modeling Conf., MMM 2008, Kyoto, Japan, 70‒79.
• [13] Chen, Y.T., Chen, C.S. (2008). Fast human detection using a novel boosted cascading structure with meta stages. IEEE Trans. Image Process., 17(8), 1452‒1464.
• [14] Cheng, H.Y., Zeng, Y.J., Lee C.C., Hsu S.H. (2013). Segmentation of Pedestrians with Confidence Level Computation. Journal of Signal Processing Systems, 72(2), 87‒97.
• [15] Wang, X., Han, T. X., Yan, S. (2009). An HOG-LBP human detector with partial occlusion handling. Proc. IEEE Int. Conf. on Comput. Vision, ICCV 2009, Kyoto, 32‒39.
• [16] Geismann, P., Knoll, A. (2010). Speeding Up HOG and LBP Features for Pedestrian Detection by Multiresolution Techniques. Proc. 6th Int. Symposium Advances in Visual Computing, ISVC 2010, Las Vegas, NV, USA, 243‒252.
• [17] Zeng, C., Ma, H., Ming, A. (2010). Fast human detection using mi-sVM and a cascade of HOG-LBP features. Proc. 17th IEEE Int. Conf. on Image Processing (ICIP), 3845‒3848.
• [18] Crow, F. (1984). Summed-area tables for texture mapping. Proc. of SIGGRAPH, 18(3), 207‒212.
• [19] Takagi, K., Tanaka, K., Izumi, S., Kawaguchi, H., Yoshimoto, M. (2014). A Real-time Scalable Object Detection System using Low-power HOG Accelerator VLSI. Journal of Signal Processing Systems.
• [20] Jendernalik, W., Blakiewicz, G., Handkiewicz, A., Melosik, M. (2013). Analogue CMOS ASICs in image processing systems. Metrol. Meas. Syst., 20(4), 613‒622.
• [21] Everingham, M., Zisserman, A., Williams, C.K.I., Van Gool, L. (2006). The PASCAL Visual Object Classes Challenge 2006 (VOC 2006) Results. Technical Report, Univ. of Oxford.
• [22] Chang, C.C., Lin, C.J. (2011). LIBSVM: a library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2(3) 27:1‒27:27.

Uwagi

PL

Opracowanie ze środków MNiSW w ramach umowy 812/P-DUN/2016 na działalność upowszechniającą naukę.