تشخیص ناهنجاری در مسیرخودرو با استفاده از از آموزش لغت نامه و بازسازی تنک
محورهای موضوعی : مهندسی برق و کامپیوترریحانه تقی زاده خانکوک 1 , عباس ابراهیمی مقدم 2 , مرتضی خادمی 3
1 - دانشگاه فردوسی،دانشکده مهندسی
2 - دانشگاه فردوسی،دانشکده مهندسی
3 - دانشگاه فردوسی،دانشکده مهندسی
کلید واژه: آموزش لغت‌نامه, استخراج ویژگی, بازسازی تنک تشخیص ناهنجاری,
چکیده مقاله :
در سامانه‌های کنترل ترافیک و ثبت تخلفات وسایل نقلیه همواره دستیابی به سامانه‌ای که بتوان با استفاده از آن به طور خودکار رفتارهای ناهنجار رانندگان را شناسایی کرد، چالشی اساسی به شمار می‌آید. در این تحقیق سامانه‌ای با مشخصات مذکور برای تشخیص ناهنجاری مسیر خودروها پیشنهاد گردیده که در آن ابتدا به استخراج ویژگی‌های زمانی- مکانی و تشکیل یک طبقه‌بند با کمک لغت‌نامه حاصل از آن ویژگی‌ها پرداخته می‌شود. طبقه‌بند از پردازش‌هایی چون خوشه‌بندی بهینه‌شده با الگوریتم جفت‌گیری زنبور عسل و پردازش تنک روی ویژگی‌های زمانی- مکانی حاصل از داده‌های آموزشی تشکیل می‌گردد. طبقه‌بند طراحی‌شده روی داده‌های آزمون، به منظور تشخیص ناهنجاری اعمال می‌شود. وجه تمایز این پژوهش نسبت به پژوهش‌های پیشین علاوه بر شیوه نوین در پیش‌پردازش صورت‌گرفته به منظور ایجاد ماتریس لغت‌نامه، تشخیص ناهنجاری بر پایه ارزیابی ماتریس حاصل از تعلق داده‌ها به هر طبقه است که منجر به دقت بالاتر روش پیشنهادی نسبت به سایر روش‌های رقیب می‌شود. برای ارزیابی بهتر روش پیشنهادی، ابتدا آن را روی پایگاه داده UCSD و سپس روی دنباله‌های ویدئویی استخراج‌شده از عبور و مرور خودروها در ضلع شمالی دانشگاه فردوسی مشهد اعمال نموده و سپس نتایج حاصل، با نتایج سایر پژوهش‌های شناخته‌شده در این حوزه مقایسه می‌گردد.
In traffic control and vehicle registration systems a big challenge is achieving a system that automatically detects abnormal driving behavior. In this paper a system for detection of vehicle anomalies proposed, which at first extracts spatio-temporal features form clusters then creates dictionary from these features. This classification stage consists of processes such as, optimized clustering with the bee mating algorithm and sparse processing on spatiotemporal features derived from the training data. Finally the trained classifier is applied to the test data for anomaly detection. The distinction of this study from previous research is using new method of pre-processing to create a dictionary matrix and anomaly detection based on evaluation of matrix that related to each class dependency, which leads to higher accuracy of the proposed method compared to other leading methods. To evaluate the proposed method, UCSD database and video sequences recorded from vehicle traffic on Vakilabad Boulevard at the north side of Ferdowsi University of Mashhad are used and the performance of the proposed method is compare to other competing methods in this field. By analyzing the evaluation standards, we find that the proposed method performance is better than other methods.
[1] J. Wrigh, A. Y. Yang, A. Ganesh, S. Shankar Sastry, and Y. Ma, "Robust face recognition via sparse representation," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 31, no. 2, pp. 210-227, Feb. 2009.
[2] Y. Chong, W. Chen, et al., "Method for preceding vehicle type classification based on sparse representation," Transportation Research Record, J. of the Transportation Research Board, vol. 2243, no. 1, pp. 74-80, 2011.
[3] A. G. Okyere and N. Attoh-Okine, "Traffic sign recognition using sparse representations and active contour models," Transportation Research Record: J. of the Transportation Research Board, vol. 2463, no. 1, pp. 35-45, 2014.
[4] Y. Cong, J. Yang, and J. Liu, "Abnormal event detection in crowded scenes using sparse representation," Pattern Recognition, vol. 46, no. 7, pp. 1851-1864, Jul. 2013.
[5] R. Chartrand, "Nonconvex regularization for shape preservation," in Proc. IEEE Int. Conf. Image Process., vol. 1, pp. 293-296, San Antonio, CA, USA, 16-19 Sept. 2007.
[6] R. Chartrand, "Exact reconstruction of sparse signals via nonconvex minimization," IEEE Signal Processing Letters, vol. 14, no. 10, pp. 707-710, Oct. 2007.
[7] J. Fan and R. Li, "Variable selection via nonconcave penalized likelihood and its oracle properties," J. of the American Statistical Association, vol. 96, no. 456, pp. 1348-1360, Dec. 2001.
[8] M. Nikolova, "Analysis of the recovery of edges in images and signals by minimizing nonconvex regularized least-squares," Multiscale Modeling & Simulation, vol. 4, no. 3, pp. 960-991, 2005.
[9] M. Izadi, Z. Azimifar, and G. H. Jowkar, "Abnormal event detection in indoor video using feature coding," in Proc. IEEE Artificial Intelligence and Signal Processing Conf., pp. 151-155, Shiraz, Iran, 25-27 Oct. 2017.
[10] S. Li, C. Liu, and Y. Yang, "Anomaly detection based on sparse coding with two kinds of dictionaries," Signal, Image and Video Processing, vol. 12, no. 5, pp. 983-989, Jul. 2018.
[11] S. Biswas and V. Gupta, "Abnormality detection in crowd videos by tracking sparse components," Machine Vision and Applications, vol. 28, no. 1, pp. 35-48, Feb. 2017.
[12] X. Chen, F. Xu, and Y. Ye, "Lower bound theory of nonzero entries in solutions of \ell_2-\ell_p minimization," SIAM J. on Scientific Computing, vol. 32, no. 5, pp. 2832-2852, 2010.
[13] A. Adam, E. Rivlin, I. Shimshoni, and D. Reinitz, "Robust real-time unusual event detection using multiple fixed-location monitors," Pattern Analysis and Machine Intelligence, IEEE Trans. on, vol. 30, no. 3, pp. 555-560, Mar. 2008.
[14] R. Mehran, A. Oyama, and M. Shah, "Abnormal crowd behavior detection using social force model," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 935-942, Miami, FL, USA, 20-25 Jun. 2009.
[15] J. Kim and K. Grauman, "Observe locally, infer globally: a space-time MRF for detecting abnormal activities with incremental updates," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 2921-2928, FL, USA, 20-25 Jun. 2009.
[16] T. Hospedales, S. Gong, and T. Xiang, "A Markov clustering topic model for mining behaviour in video," in Proc. IEEE 12th Int. Conf. on Computer Vision, pp. 1165-1172, Kyoto, Japan, 29 Sept.- 2 Oct. 2009.
[17] V. Kaltsa, A. Briassouli, I. Kompatsiaris, L. J. Hadjileontiadis, and M. G. Strintzis, "Swarm intelligence for detecting interesting events in crowded environments," IEEE Trans. on Image Processing, vol. 24, no. 7, pp. 2153-2166, Jul. 2015.
[18] S. Zhou, et al., "Spatial-temporal convolutional neural networks for anomaly detection and localization in crowded scenes," Signal Processing, Image Communication, vol. 47, pp. 358-368, Sept. 2016.
[19] K. Doshi and Y. Yilmaz, "An efficient approach for anomaly detection in traffic videos," in Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition, pp. 4236-4244, Nashville, TN, USA, 19-25 Jun. 2021.
[20] M. Liatsikou, S. Papadopoulos, L. Apostolidis, and Y. Kompatsiaris, "A denoising hybrid model for anomaly detection in trajectory sequences," in Proc. EDBT/ICDT Workshops, 23-26 Mar. 2021.
[21] H. Mohimani, M. Babaie-Zadeh, and C. Jutten, "A fast approach for overcomplete sparse decomposition based on smoothed l0 norm," IEEE Trans. on Signal Processing, vol. 57, no. 1, pp. 289-301, Jan. 2008.
[22] H. Zayyani, M. Babaie-Zadeh, and C. Jutten, "An iterative Bayesian algorithm for sparse component analysis in presence of noise," IEEE Trans. on Signal Processing, vol. 57, no. 11, pp. 4378-4390, Jun. 2009.
[23] Y. Bosi, Y. Liu, and Q. Sun, "Low-rank approximation based abnormal detection in the video sequence," in Proc. IEEE Int. Conf. on Digital Signal Processing, pp. 129-133, Beijing, China, 16-18 Oct. 2016.
[24] A. Afshar, O. Bozorg Haddad, M. A. Marino, B. J. Adams, "Honey-bee mating optimization (HBMO) algorithm for optimal reservoir operation," J. of the Franklin Institute, vol. 344, no. 5, pp. 452-462, Aug. 2007.
[25] C. Yang, J. Yuan, and J. Liu, "Sparse reconstruction cost for abnormal event detection," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 3449-3456, Colorado Springs, CO, USA, 20-25 Jun. 2011.
[26] L. Cewu, J. Shi, and J. Jia, "Abnormal event detection at 150 fps in MATLAB," in Proc. of the IEEE Int. Conf. on Computer Vision, pp. 2720-2727, Sydney, Australia, 1-8 Dec. 2013.
[27] C. Zhi-Jun, et al., "Vehicle behavior learning via sparse reconstruction with l2 – lp minimization and trajectory similarity," IEEE Trans. on Intelligent Transportation Systems, vol. 18, no. 2, pp. 236-247, Feb. 2017.
[28] UCSD Anomaly Detection Dataset, http://www.svcl.ucsd.edu/projects/anomaly/dataset.htm.
[29] B. Auslander, K. Moy Gupta, and D. W. Aha, "A comparative evaluation of anomaly detection algorithms for maritime video surveillance," Proceedings Volume 8019, Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense X, Article ID: 801907, 2011.
