1.School of Mechanical, Electronic and Control Engineering, Beijing Jiaotong University, Beijing 100044, China
2.Key Laboratory of Vehicle Advanced Manufacturing, Measuring and Control Technology, Beijing Jiaotong University, Ministry of Education, Beijing 100044, China
3.Safety Supervision Brigade, China Railway Lanzhou Group Co. , Ltd. , Lanzhou Gansu 730000, China
Accurately identifying human and large livestock intruding within the perimeter is a key focus of intelligent video analysis technology for railway perimeter intrusion. It is of great significance for ensuring railway safety operations. However, existing object detection algorithms struggle to handle significant scale variations of intrusion objects in railway monitoring scenarios. Therefore, a Multiple Input Double Output Network (MIDO-Net) and a multi-scale feature perception algorithm based on adaptive weighted fusion are proposed. Firstly, the MIDO-Net extracts richer multi-scale feature information of image objects through its multi-level cascaded multiple input and double output network structure. Secondly, based on the multi-stage characteristics of the backbone network, the multi-level features are sampled up to unified resolution and then weighted using attention modules and adaptive parameters. Then, the features are input into the detection head to complete the recognition of railway perimeter intrusion. Finally, the algorithm is validated using the Visual Object Classes (VOC) public dataset and a self-made dataset of railway foreign object intrusion in multiple scenes and scales. The results show that the proposed multi-scale feature perception algorithm achieves a detection accuracy of 83.3% in the VOC public dataset and 91.1% in the dataset of railway foreign object intrusion in multiple scenes and scales. The average recall rate is 56.2%, which is superior to various widely used feature extraction backbone networks. The algorithm detection rate is 45 frames per second (fps), surpassing similar backbone networks, and can meet the requirements for pedestrian real-time monitoring in railway scenarios.
ZHANGJing. Brief Introduction of Scientific and Technological Achievements in China Academy of Railway Sciences in 2015 (Continued 2) [J]. China Railway Science, 2017, 38 (3): 24, 39, 46, 76, 82, 109, 123. in Chinese
[3]
ALVAREZF J, UREFIAJ, MAZOM, et al. Ultrasonic Sensor System for Detecting Falling Objects on Railways [C]// IEEE Intelligent Vehicles Symposium. New York: IEEE Press, 2004: 866-871.
WANGQianxuan, LIANGXifeng, LIUYinglong, et al. Visual Detection Method for the Invasion of Slowly Changing Foreign Matters to Railway Lines [J]. China Railway Science, 2014, 35 (3): 137-143. in Chinese
DONGHonghui, SUNZhiyuan, GEDawei, et al. Target Recognition Method of Railway Invasion Based on Gaussian Mixture Model [J]. China Railway Science, 2011, 32 (2): 131-135. in Chinese
[8]
李荣增.基于视频识别的铁路入侵检测研究与实现[D].武汉:武汉理工大学,2016:48-53.
[9]
LIRongzeng. Research and Implementation of Railway Intrusion Detection Based on Computer Vision [D]. Wuhan: Wuhan University of Technology, 2016: 48-53. in Chinese
QUZhijian, ZHOURuilin, SUNXubing, et al. Scale Adaptive PSA-Kcf Dimension Reduction Tracking Method for Railway Foreign Body Intrusion [J]. Journal of the China Railway Society, 2019, 41 (5): 71-81. in Chinese
SHIHongmei, CHAIHua, WANGYao, et al. Study on Railway Embedded Detection Algorithm for Railway Intrusion Based on Object Recognition and Tracking [J]. Journal of the China Railway Society, 2015, 37 (7): 58-65. in Chinese
[14]
GIRSHICKR, DONAHUEJ, DARRELLT, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation [C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA. New York: IEEE Press, 2014: 580-587.
[15]
GIRSHICKR. Fast R-CNN [C]// 2015 IEEE International Conference on Computer Vision. Santiago, Chile. New York: IEEE Press, 2015: 1440-1448.
[16]
RENS Q, HEK M, GIRSHICKR, et al. Faster R-CNN: towards Real-Time Object Detection with Region Proposal Networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137-1149.
[17]
LIUW, ANGUELOVD, ERHAND, et al. SSD: Single Shot MultiBox Detector [M]. Computer Vision - ECCV 2016. Cham: Springer, 2016: 21-37.
REDMONJ, DIVVALAS, GIRSHICKR, et al. You Only Look Once: Unified, Real-Time Object Detection [C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA New York: IEEE Press, 2016: 779-788.
[21]
REDMONJ, FARHADIA. Yolo9000: Better, Faster, Stronger [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, USA. New York: IEEE Press, 2017: 6517-6525.
NEWELLA, HUANGZ A, DENGJ. Associative Embedding: End-to-End Learning for Joint Detection and Grouping [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS 2017. Long Beach, California, USA. New York: ACM, 2017: 2274-2284.
[24]
GEZ, LIUS, WANGF, et al. Yolox: Exceeding Yolo Series in 2021 [EB/OL]. arXiv: 2107.08430, 2021 (2021-8-6) [2022-10-18].
[25]
SIMONYANK, ZISSERMANA. Very Deep Convolutional Networks for Large-Scale Image Recognition [EB/OL]. arXiv: 1409.1556, 2014 (2015-4-10) [2022-10-18].
[26]
HEK M, ZHANGX Y, RENS Q, et al. Deep Residual Learning for Image Recognition [C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA. New York: IEEE Press, 2016: 770-778.
[27]
HUANGG, LIUZ, MAATENL V D, et al. Densely Connected Convolutional Networks [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, USA. New York: IEEE Press, 2017: 4700-4708.
[28]
SUNK, XIAOB, LIUD, et al. Deep High-Resolution Representation Learning for Human Pose Estimation [C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA. New York: IEEE Press, 2019: 5693-5703.
[29]
JINY M, FUY S, WANGW Q, et al. Multi-Feature Fusion and Enhancement Single Shot Detector for Traffic Sign Recognition [J]. IEEE Access, 2020, 8: 38931-38940.
[30]
HANC Y, ZHUJ L, LIF, et al. Design of Lightweight Pedestrian Detection Network in Railway Scenes [C]// 2020 5th International Conference on Intelligent Computing and Signal Processing (ICSP). Bristol: IOP Publishing, 2020: 12053-12061.
[31]
CUIL S, LÜP, JIANGX H, et al. Context-Aware Block Net for Small Object Detection [J]. IEEE Transactions on Cybernetics, 2022, 52 (4): 2300-2313.
[32]
QIUY, LIUY, YANGH, et al. A Simple Saliency Detection Approach via Automatic Top-Down Feature Fusion [J]. Neurocomputing, 2020, 388: 124-134.
[33]
LIY Y, HUANGQ, PEIX, et al. Cross-Layer Attention Network for Small Object Detection in Remote Sensing Imagery [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 2148-2161.
[34]
WUQ F, FENGD Q, CAOC Q, et al. Improved Mask R-CNN for Aircraft Detection in Remote Sensing Images [J]. Sensors, 2021, 21 (8): 2618-2630.
YET, WANGB C, SONGP, et al. Automatic Railway Traffic Object Detection System Using Feature Fusion Refine Neural Network under Shunting Mode [J]. Sensors, 2018, 18 (6): 1916-1934.
[37]
YET, ZHANGX, ZHANGY, et al. Railway Traffic Object Detection Using Differential Feature Fusion Convolution Neural Network [J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22 (3): 1375-1387.
[38]
DUANK W, BAIS, XIEL X, et al. CenterNet: Keypoint Triplets for Object Detection [C]// 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South). New York: IEEE Press, 2019: 6569-6578.
[39]
DAIJ F, LIY, HEK M, et al. R-FCN: Object Detection via Region-Based Fully Convolutional Networks [C]// Proceedings of the 30th International Conference on Neural Information Processing Systems (NIPS 2016). Barcelona, Spain. New York: ACM, 2016: 379-387.
[40]
FUC Y, LIUW, RANGAA, et al. DSSD: Deconvolutional Single Shot Detector [EB/OL]. arXiv: 1701.06659, 2017 (2017-1-23) [2022-10-18].
[41]
ZHANGS F, WENL Y, LEIZ, et al. RefineDet++: Single-Shot Refinement Neural Network for Object Detection [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 31 (2): 674-687.
[42]
TIANZ, SHENC, CHENH, et al. FCOS: Fully Convolutional One-Stage Object Detection [C]// 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South). New York: IEEE Press, 2019: 9627-9636.
[43]
NEWELLA, YANGK Y, DENGJ. Stacked Hourglass Networks for Human Pose Estimation [M]. Computer Vision - ECCV 2016. Cham: Springer International Publishing, 2016: 483-499.
[44]
YUF, WANGD, SHELHAMERE, et al. Deep Layer Aggregation [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2018: 2403-2412.
[45]
ZHANGX Y, ZHOUX Y, LINM X, et al. ShuffleNet: an Extremely Efficient Convolutional Neural Network for Mobile Devices [C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT. New York: IEEE Press, 2018: 6848-6856.
[46]
SANDLERM, HOWARDA, ZHUM L, et al. MobileNetV2: Inverted Residuals and Linear Bottlenecks [C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT. New York: IEEE Press, 2018: 4510-4520.