Abstract Pedestrian detection is one of the important areas in computer vision. This work is about detecting the multi- directional pedestrian’s left, right, and the front movements. On recognizing the direction of movement, the system can be alerted depending on the environmental circumstances. Since multiple pedestrians moving in different directions may be present in a single image, Convolutional Neural Network (CNN) is not suitable for recognizing the multi-directional movement of the pedestrians. Moreover, the Faster R-CNN (FR-CNN) gives faster response output compared to other detection algorithms. In this work, a modified Faster Recurrent Convolutional Neural Network (MFR-CNN), a cognitive approach is proposed for detecting the direction of movement of the pedestrians and it can be deployed in real-time. A fine-tuning of the convolutional layers is performed to extract more information about the image contained in the feature map. The anchors used in the detection process are modified to focus the pedestrians present within a range, which is the major concern for such automated systems. The proposed model reduced the execution time and obtained an accuracy of 88%. The experimental evaluation indicates that the proposed novel model can outperform the other methods by tagging each pedestrian individually in the direction in which they move.

References

[1] Baek J., Kim J., and Kim E., “Fast and Efficient Pedestrian Detection via the Cascade Implementation of an Additive Kernel Support Vector Machine,” IEEE Transactions on Intelligent Transportation Systems, vol. 18, no. 4, pp. 902-916, 2017.

[2] Bin W. and Shiru Q., “A study on Occluded Pedestrian Detection Based on Block-Based Features And Ensemble Classifier,” in Proceedings of 34th Chinese Control Conference, Hangzhou, pp. 4710-4715, 2015.

[3] Cao J., Pang Y., and Li X., “Learning Multilayer Channel Features for Pedestrian Detection,” IEEE Transactions on Image Processing, vol. 26, no. 7, pp. 3210-3220, 2017.

[4] Chen E., Tang X., and Fu B., “A Modified Pedestrian Retrieval Method Based on Faster R- CNN with Integration of Pedestrian Detection and Re-Identification,” in Proceedings of International Conference on Audio, Language and Image Processing, Shanghai, pp. 63-66, 2018.

[5] Chen Y., Xie H., and Shin H., “Multi-layer fusion Techniques Using A cnn for Multispectral Pedestrian Detection,” IET Computer Vision, vol. 12, no. 8, pp.1179-1187, 2018.

[6] Dominguez-Sanchez A., Cazorla M., and Orts- Escolano S., “Pedestrian Movement Direction Recognition Using Convolutional Neural Networks,” IEEE Transactions on Intelligent Transportation Systems, vol. 18, no. 12, pp. 3540- 3548, 2017.

[7] Elmezain M. and Al-Hamadi A., “Vision-Based Human Activity Recognition Using LDCRFs,” The International Arab Journal of Information Technology, vol. 15, no. 3, pp. 389-395, 2018.

[8] Jeong M., Ko B., and Nam J., “Early Detection of Sudden Pedestrian Crossing for Safe Driving During Summer Nights,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 27, no. 6, pp.1368-1380, 2017.

[9] Kharjul R., Tungar V., Kulkarni Y., Upadhyay S., and Shirsath R., “Real-Time Pedestrian Detection Using SVM and AdaBoost,” in Proceedings of International Conference on Energy Systems and Applications, Pune, pp.740-743, 2015.

[10] Li Y., Lu W., Wang S., and Ding X., “Local Haar- Like Features in Edge Maps for Pedestrian Detection,” in Proceedings of 4th International Congress on Image and Signal Processing, Shanghai, pp. 1424-1427, 2011.

[11] Lin C., Lin S., Hwang C., and Chen Y., “Real- Time Pedestrian Detection System with Novel Thermal Features At Night,” in Proceedings of IEEE International Instrumentation and Measurement Technology Conference Proceedings, Montevideo, pp. 1329-1333, 2014.

[12] Nam W., Dollar P., and Han J., “Local 'HFRUUHODWLRQIRU,PSURYHG'HWHFWLRQ´arxiv.org, abs/1406.1134, 2014.

[13] Orozco C., Buemi M., and Berlles J., “New Deep Convolutional Neural Network Architecture for Pedestrian Detection,” in Proceedings of 8th International Conference of Pattern Recognition Systems, Madrid, pp. 1-6, 2017.

[14] Qiu D. and Liu D., “The Optimal Pedestrian Detection Algorithm Based on Dynamic Adaptive Region Convolution Model,” Chinese Automation Congress (CAC), Jinan, pp. 7808-7910, 2017.

[15] Sun W., Zhu S., Ju X., and Wang D., “Deep Learning Based Pedestrian Detection,” Chinese 252 The International Arab Journal of Information Technology, Vol. 20, No. 2, March 2023 Control and Decision Conference (CCDC), Shenyang, pp. 1007-1011, 2018.

[16] Wang A., Dai S., Yang M., and Iwahori Y., “A Novel Human Detection Algorithm Combining HOG with LBP Histogram Fourier,” in Proceedings of 10th International Conference on Communications and Networking in China, Shanghai, pp. 793-797, 2015.

[17] Wang S., Cheng J., Liu H., Wang F., and Zhou H., “Pedestrian Detection via Body Part Semantic and Contextual Information with DNN,” IEEE Transactions on Multimedia, vol. 20, no. 11, pp. 3148-3159, 2018.

[18] Wang Y. and Liu F., “A New Pedestrian Detection Algorithm Used for Advanced Driver-Assistance System with One Cheap Camera,” in Proceedings of International Conference on Mechatronic Sciences, Electric Engineering and Computer, Shengyang, pp.1315-1318, 2013.

[19] Weixing L., Haijun S., Feng P., Qi G., and Bin Q., “A fast Pedestrian Detection Via Modified HOG Feature,” in Proceedings of 34th Chinese Control Conference, Hangzhou, pp. 3870- 3873, 2015.

[20] Yang Z., Li J., and Li H., “Real-Time Pedestrian Detection for Autonomous Driving,” in Proceedings of International Conference on Intelligent Autonomous Systems, Singapore, pp. 9- 13, 2018.

[21] Yu B., Ma Y., and Li J., “Fast Pedestrian Detection with Multi-Scale Classifiers,” in Proceedings of International Conference on Computing Intelligence and Information System, Nanjing, pp. 225-230, 2017.

[22] Zhang C. and Kim J., “Multi-Scale Pedestrian Detection Using Skip Pooling and Recurrent Convolution,” Multimedia Tools and Applications, vol. 78, pp. 1719-1736, 2018.

[23] Zhang H., Du Y., Ning S., Zhang Y., Yang S., and Du C., “Pedestrian Detection Method Based on Faster R-CNN,” in Proceedings of 13th International Conference on Computational Intelligence and Security, Hong Kong, pp. 427- 430, 2017.

[24] Zhang J., Xiao J., Zhou C., and Peng C., “A Multi- Class Pedestrian Detection Network for Distorted Pedestrians,” in Proceedings of 13th IEEE Conference on Industrial Electronics and Applications, Wuhan, pp. 1079-1083, 2018.

[25] Zhang S. and Wang X., “Human Detection and Object Tracking based on Histograms of Oriented Gradients,” in Proceedings of 9th International Conference on Natural Computation, Shenyang, pp. 1349-1353, 2013. Jayachitra Virupakshipuram Panneerselvam received the B.E. degree in computer science and engineering from university of Madras, Chennai, India, in 2000 and the M.E degree in Computer Science and Engineering and Ph.D. degree in Information and communication engineering from Anna University, Chennai, India, in 2008 and 2017 respectively. Currently, she is an assistant professor in Department of Computer Technology, MIT campus, Anna University, Chennai, India. Her research interest includes Wireless Sensor Networks, Machine learning and Internet of Things. Bharanidharan Subramaniam received the Bachelors of Engineering degree in Computer Science and Engineering from Anna University, India in 2019. He is currently pursuing a career in the software industry as lead software engineer at Samsung. His areas of interest include Data structures and Machine Learning. Mathangi Meenakshisundaram received the Bachelors of Engineering degree in Computer Science and Engineering at Department of Computer Technology from Anna University, India, graduated in the year 2020. She is currently pursuing a career in the software industry as software engineer at Fidelity Investments. Her areas of interests include Machine Learning, Big Data, Database Management and Internet of Things.