融合不降维局部跨通道交互策略的双路径金字塔检测算法

doi:10.3778/j.issn.1002-8331.2307-0206

摘要/Abstract

摘要： 针对道路障碍物目标检测任务中多尺度目标检测精度低，以及在不同场景下检测鲁棒性和泛化能力差等问题。改进算法基于传统特征金字塔网络，提出一种自上而下和自下而上结合的双路径特征金字塔网络模块，通过特征拼接和融合操作保留预测特征层中更多的浅层和深层次语义信息。在此基础上，提出一种空间和通道机制串联的注意力网络模块，通过采用不降维的局部跨通道交互策略，进一步提升网络模型检测性能。经实验验证，改进算法相较于原始算法目标检测准确率提升4.6个百分点；小目标检测准确率提升11.76个百分点；中目标检测准确率提升5.78个百分点；大目标检测准确率提升3.7个百分点。

关键词: 计算机视觉, 目标检测, 深度学习, 特征金字塔网络, 注意力机制

Abstract: In the object detection task for road obstacles, the detection accuracy of multi-scale objects is low, and the detection robustness and generalization ability are poor in different scenes. Based on the traditional feature pyramid network, the improved algorithm proposes a double path feature pyramid network that combines top-down and bottom-up features, and retains more shallow and deep semantic information in the prediction feature layer through feature concat and fusion methods. On this basis, a mechanism attention module with spatial and channel mechanisms in series is proposed, which further improves the network model detection performance by adopting a local cross-channel interaction strategy without dimension reduction. The experimental results show that the accuracy of object detection of the improved algorithm is 4.6?percentage points higher than that of the original algorithm. The accuracy of small object detection increased by 11.76 percentage points. The accuracy of medium object detection is increased by 5.78?percentage points. The accuracy rate of large object detection is increased by 3.7?percentage points.

Key words: computer vision, object detection, deep learning, feature pyramid network, attention mechanism

焦博文, 王玉林, 王鹏, 王洪昌, 于奕轩, 沈正坤. 融合不降维局部跨通道交互策略的双路径金字塔检测算法[J]. 计算机工程与应用, 2024, 60(12): 91-100.

JIAO Bowen, WANG Yulin, WANG Peng, WANG Hongchang, YU Yixuan, SHEN Zhengkun. Double Path Feature Pyramid Network Detection Algorithm Based on Integrating Non-Dimensionality-Reducing Locally Cross-Channel Interaction Strategy[J]. Computer Engineering and Applications, 2024, 60(12): 91-100.

参考文献

[1] KRIAN B R, SOBH I, TALPAER V, et al. Deep reinforcement learning for autonomous driving: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(6): 4909-4926.
[2] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), 2005: 886-893.
[3] ZHAO W L, NGO C W. Flip-invariant SIFT for copy and object detection[J]. IEEE Transactions on Image Processing, 2012, 22(3): 980-991.
[4] VIOLA P, JONES M. Rapid object detection using a boosted cascade of simple features[C]//Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2001.
[5] TENG Z, ZHANG B, KIM O, et al. Regional SVM classifiers with a spatial model for object detection[C]//2014 International Conference on Computer Vision Theory and Applications (VISAPP), 2014: 372-379.
[6] MA J, PAN Q, HU J, et al. Small object detection with random decision forests[C]//2017 IEEE International Conference on Unmanned Systems (ICUS), 2017: 566-571.
[7] DING J, ZHANG J, ZHAN Z, et al. A precision efficient method for collapsed building detection in post-earthquake UAV images based on the improved NMS algorithm and faster R-CNN[J]. Remote Sensing, 2022, 14(3): 663.
[8] 张婷, 张兴忠, 王慧民, 等. 基于图神经网络的变电站场景三维目标检测[J]. 计算机工程与应用, 2023, 59(9): 329-336.
ZHANG T, ZHANG X Z, WANG H M, et al. 3D object detection in substation scene based on graph neural network[J]. Computer Engineering and Applications, 2023, 59(9): 329-336.
[9] 王鹏, 王玉林, 焦博文. 基于雾天复杂场景下的道路目标检测研究[J]. 青岛大学 (工程技术版), 2023, 38(2): 37-45.
WANG P, WANG Y L, JIAO B W. Research on road object detection based on complex scenarios in foggy weather[J]. Journal of Qingdao University (Engineering & Technology Edition), 2023, 38(2): 37-45.
[10] 张诗慧, 罗晖, 裴莹玲, 等. 基于改进RetinaNet的高铁无砟轨道板表面裂缝检测[J]. 计算机工程与应用, 2023, 59(6): 310-317.
ZHANG S H, LUO, H, PEI Y L, et al. Surface crack detection in ballastless slab track of high-speed railway based on improved RetinaNet[J]. Computer Engineering and Applications, 2023, 59(6): 310-317.
[11] 王鹏, 王玉林, 焦博文, 等. 基于YOLOv5的道路目标检测算法研究[J]. 计算机工程与应用, 2023, 59(1): 117-125.
WANG P, WANG Y L, JIAO B W, et al. Research on road target detection algorithm based on YOLOv5[J]. Computer Engineering and Applications, 2023, 59(1): 117-125.
[12] XIE S, GIRSHICK R, DOLLA P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 1492-1500.
[13] ZHU Q Q, LIU S, GUO W M. Research on vehicle appearance component recognition based on mask R-CNN[C]//2019 3rd International Conference on Computer Graphics and Digital Image Processing (CGDIP 2019), Rome, Italy, 25-27 July 2019.
[14] ZHANG Y, ZHUO L, MA C, et al. Abnormal object detection in x-ray images with self-normalizing channel attention and efficient data augmentation[C]//International Workshop on Advanced Imaging Technology (IWAIT2022), 2022: 108-112.
[15] LI H, CHENG J Z, CHOU Y H, et al. AttentionNet: learning where to focus via attention mechanism for anatomical segmentation of whole breast ultrasound images[C]//2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), 2019: 1078-1081.
[16] ZENG N Y, WU P S, WANG Z D, et al. A small-sized object detection oriented multi-scale feature fusion approach with application to defect detction[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-14.
[17] LI S S, LI Y J, LI Y, et al. YOLO-FIRI: improved YOLOv5 for infrared image object detection[J]. IEEE Access, 2021 99: 1.
[18] LENG J X, REN Y H, JIANG W, et al. Realize your surroundings: exploiting context information for small object detection[J]. Neurocomputing, 2021, 433(8).
[19] YAN D, LI G, LI X, et al. An improved faster R-CNN method to detect tailings ponds from high-resolution remote sensing images[J]. Remote Sensing, 2021, 13(11): 2052.
[20] LIN T Y, DOLLA P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
[21] FAN B B, YANG H. Multi-scale traffic sign detection model with attention[J]. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2021, 235(2/3): 708-720.
[22] YU H, CHENG X, LI Z, et al. Disease recognition of apple leaf using lightweight multi-scale network with ECANet[J]. CMES-Computer Modeling in Engineering & Sciences, 2022, 132(3).
[23] WANG Q, WU B, ZHU P, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11534-11542.
[24] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[25] CUI Y, XU Y, PENG R, et al. Layer normalization for TSK fuzzy system optimization in regression problems[J]. IEEE Transactions on Fuzzy Systems, 2022, 31(1): 254-264.
[26] KUMAR S, CHOUDHARY M, KUMARI K, et al. Intelligent driving system at opencast mines during foggy weather[J]. International Journal of Mining Reclamation and Environment, 2022, 36(3): 196-217.