UDC 004.85(075.8) USING NON-MAXIMUM SUPPRESSION ALGORITHM TO INCREASE ACCURACY AND PROCESSING SPEED OF REAL-TIME FACE RECOGNITION
Abstract
This paper discusses modern face recognition algorithms, which are conventionally classified into geometric and template methods. Template methods use statistical approaches such as SVM, PCA, LDA, and convolutional neural networks. Particular attention is paid to the Viola-Jones algorithm, which has a high detection rate, and the LBPH method, which provides accurate recognition. Also, the Non-Maximum Suppression (NMS) algorithm, which is used in object detection problems to eliminate redundant or overlapping predictions, is analyzed in detail. Mathematical formulas and a step-by-step implementation of the algorithm are presented, including the calculation of area, intersection, and overlap coefficient. NMS optimizes the final selection of bounding boxes and is widely used in computer vision systems.
About the Authors
List of references
MA Imran, MSU Miah, H Rahman , A Bhowmik, D Karmaker, 2015. Face Recognition Using Eigenfaces. International Journal of Computer Applications, vol. 118, no. 5
Dr. H S Prasantha. Detailed Survey Of Machine Learning Algorithms For Face Recognition. International Journal of Creative Research Thoughts (IJCRT). Volume 11, Issue 11 November 2023. 832-836 p.
Jan Hosang, Rodrigo Benenson, Bernt Schiele. Learning non-maximum suppression. Max Planck Institut für Informatik Saarbrücken, Germany.
Krishna, M., & Jawahar, C. V. (2021). Deep Learning for Drone and UAV Applications. In S. S. H. De Silva & B. B. Gupta (Eds.), Handbook of Smart Materials, Technologies, and Devices (pp. 75–93). Springer, Cham.
Xiyue Wang, Xinsheng Wang, Zhiquan Zhou, Yanhong Song. Fast Detection and Obstacle Avoidance on UAVs Using Lightweight Convolutional Neural Network Based on the Fusion of Radar and Camera. Applied Intelligence, August 2024
Zhu, X., Su, W., Lu, L., Li, B., Wang, X., & Dai, J. (2020). Deformable DETR: Deformable Transformers for End-to-End Object Detection.
Yanan Song, Quan-Ke- Pan , Liang Gao, Biao Zong (2019) Improved non-maximum suppression for object detection using harmony search algorithm
A review of non-maximum suppression algorithms for deep learning object detection, M. Gong, SPIE Proceedings, Vol. 11763, 2021
NMS Loss: Learning with Non Maximum Suppression for Crowded Pedestrian Detection (Luo Z. , 2021)
Multi Attribute NMS: An Enhanced Non-Maximum Suppression Algorithm for Pedestrian Detection in Crowded Scenes (Wang W. ., 2023)
Accelerating Non Maximum Suppression: A Graph Theory Approach (Si K.S. 2024)
How to Cite
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.