Handheld call detection of driver based on improved Faster RCNN
1. School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China; 2. Department of Information Technology, Jiangsu Union Technical Institute, Xuzhou, Jiangsu 221008, China
Abstract：To solve the problem of high false positive rate of existing driver call behavior recognition, an improved Faster RCNN was proposed based on driver behavior detection method for detecting the illegal handheld call of driver. An optimization strategy for the region proposal network (RPN) and the loss function was introduced, and the robustness of the network in detecting targets with different sizes was enhanced by applying multiscale training, increasing the number of anchor points and introducing the residual expansion network on the original Faster RCNN. The simulation experiments of the proposed method were conducted with the images of driver behavior collected on an invehicle platform. The results show that compared with original Faster RCNN, RPN and Faster RCNN can realize efficient target detection by alternatively optimizing the shared feature extraction network part with 38% improvement in detection precision and better adaptation to the environment.
WEN H Q, QING L B，JIN R L, et al. Emotion recognition based on fusion of expression and posture[J]. Journal of Sichuan University (Natural Science Edition), doi: 10.19907/j.04906756.2021.043002. (in Chinese)
［2］
TARABAY R, ABOUZEID M. Assessing the effects of auditoryvocal distraction on driving performance and physiological measures using a driving simulator[J]. Transportation Research Part F, 2018, 58:351-364.
［3］
GU J X, WANG Z H, KUEN J, et al. Recent advances in convolutional neural networks[J]. Pattern Recognition, 2018, 77: 354-377.
［4］
SUREZPANIAGUA V, SEGURABEDMAR I. Evaluation of pooling operations in convolutional architectures for drugdrug interaction extraction[J]. BMC Bioinformatics,doi: 10.1186/s1285901821951.
XIA H S, SHEN H, HU W. Detecting distraction of drivers using human pose keypoints[J]. Computer Technology and Development, 2019, 29(7):1-5. (in Chinese)
［6］
ZAIDI S S A, ANSARI M S, ASLAM A, et al. A survey of modern deep learning based object detection models[J]. Digital Signal Processing: A Review Journal, doi: 10.1016/j.dsp.2022.103514.
［7］
WANG Y, BASHIR S M A, KHAN M, et al. Remote sensing image superresolution and object detection: benchmark and state of the art[J]. Expert Systems with Applications,doi: 10.1016/j.eswa.2022.116793.
YUAN M, QUAN T F, HUANG J, et al. Smoke detection based on convolutional neural network[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2020,32(4):620-629. (in Chinese)
［9］
HAN G X, HUANG S Y, MA J W, et al. Meta faster RCNN: towards accurate fewshot object detection with attentive feature alignment[C]∥Proceedings of the 36th AAAI Conference on Artificial Intelligence. \[S.l.\]: Association for the Advancement of Artificial Intelligence,2022.
LI P S, LI J D, NI T Y, et al. Convolution kernel initialization method based on image features[J]. Journal of Jilin University (Science Edition), 2021,59(3):587-594. (in Chinese)
CHI T, WANG Y, CHEN M. Hyperspectral image classification of multilayer local perceptual convolutional neural networks[J]. Journal of Sichuan University (Natural Science Edition),2020,57(1):103-112. (in Chinese)
