DF-SSD: A One-stage Small Target Detection Algorithm Based on Deconvolution and Feature Fusion

Abstract

Abstract: Aiming at the problem of the SSD model’s poor detection performance on small targets, the DF-SSD algorithm was proposed, its technical contributions include a one-stage detector method based on deconvolution and feature fusion and an improved default bounding boxes’ size calculation algorithm. Deconvolution and feature fusion can increase the semantic information of shallow feature layers. In DF-SSD algorithm, the improved default bounding boxes’ size calculation introduces the characteristics of the data set, which can effectively use each default bounding box for training and prediction. Compared with the improved R-SSD and RSSD models based on SSD, the DF-SSD method has higher detection accuracy. At the same time, DF-SSD’s detection overhead is only 1/2 of R-SSD and 1/5 of DSSD. The MAP of the DF-SSD on the VOC2007 and DIOR data sets is 1.4 and 3.6 percentage points higher than that of SSD respectively. Meanwhile, DF-SSD’s MAP of small targets of ship, vehicle, windmill, and cat increased 23.2, 12.6, 8 and 4.8 percentage points respectively. The results show DF-SSD effectively improves the detection accuracy of small targets and has a faster detection speed.

Key words: SSD model, deconvolution, feature fusion, small target detection, PASCAL VOC2007, DIOR

WANG Liang-wei, CHEN Mei, LI Hui, LI Huan-juan, SHI Ruo, DAI Zhen-yu. DF-SSD: A One-stage Small Target Detection Algorithm Based on Deconvolution and Feature Fusion[J]. Computer and Modernization, 2021, 0(06): 18-23.

References ［22］

［1］	寇大磊,权冀川,张仲伟. 基于深度学习的目标检测框架进展研究［J］. 计算机工程与应用, 2019,55(11):25-34.
［2］	张冬明,靳国庆,代锋,等. 基于深度融合的显著性目标检测算法［J］. 计算机学报, 2019,42(9):2076-2086.
［3］	周晓彦,王珂,李凌燕. 基于深度学习的目标检测算法综述［J］. 电子测量技术, 2017,40(11):89-93.
［4］	龙敏,佟越洋. 应用卷积神经网络的人脸活体检测算法研究［J］. 计算机科学与探索, 2018,12(10):1658-1670.
［5］	杨会成,朱文博,童英. 基于车内外视觉信息的行人碰撞预警方法［J］. 智能系统学报, 2019,14(4):752-760.
［6］	董晶,傅丹,杨夏. 无人机视频运动目标实时检测及跟踪［J］. 应用光学, 2013,34(2):255-259.
［7］	XIE L, AHMAD T, JIN L, et al. A new CNN-based method for multi-directional car license plate detection［J］. IEEE Transactions on Intelligent Transportation Systems, 2018,19(2):507-517.
［8］	GKIOXARI G, HARIHARAN B, GIRSHICK R, et al. R-CNNs for pose estimation and action detection［J］. Computer Science, 2014,12(8):1221-1229.
［9］	HE K M, GKIOXARI G, DOLLAR P. Mask R-CNN［C］// Proceedings of the IEEE International Conference on Computer Vision. 2017:2961-2969.
［10］	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks［C］// Proceedings of the 28th International Conference on Neural Information Processing Systems. 2015:91-99.
［11］	REDMON J, FARHADI A. YOLO9000: Better, faster, stronger［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition(CVP). IEEE, 2017:6517-6525.
［12］	LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single shot multibox detector［C］// Proceedings of the European Conference on Computer Vision. Springer, 2016:21-27.
［13］	LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2017:936-944.
［14］	LI Z X, ZHOU F Q. FSSD: Feature fusion single shot multibox detector［J］. Computer Vision and Pattern Recognition, 2018,36(7):356-366.
［15］	FU C Y, LIU W, RANGA A, et al. DSSD: Deconvolutional single shot detector［J］. Computer Vision and Pattern Recognition, arXiv:1701.06659.
［16］	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale imag recognition［J］. Computer Vision and Pattern Recognition, arXiv:1409.1556.
［17］	XING H Q, DU Z Q, SU B. Pedestrian detection method based on modifified SSD［J］. Computer Engineering, 2018,44(11):228-233.
［18］	TANG C, LING Y S, ZHENG K D, et al. Object detection method of multi-view SSD based on deep learning［J］. Infrared and Laser Engineering, 2018,47(1):302-310.
［19］	张文达,许悦雷,倪嘉成,等. 基于多尺度分块卷积神经网络的图像目标识别算法［J］. 计算机应用, 2016,36(4):1033-1038.
［20］	HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition［C］// Proceedings of the 13th European Conference on Computer Vision. 2014:346-361.
［21］	冯珂垚,饶鹏,陆福星,等. 基于神经网络的高分辨率快速目标检测方法［J］. 电子设计工程, 2018,26(22）:175-179.
［22］	LI K, WAN G, CHENG G, et al. Object detection in optical remote sensing images: A survey and a new benchmark［J］. ISPRS Journal of Photogrammetry and Remote Sensing, 2020,159:296-307.

[1]	NING Juan, ZHOU Qing-hua, ZENG Xiao-wei. Application of Improved YOLOv7 Algorithm in Detection of Capping Defects of Vials [J]. Computer and Modernization, 2023, 0(12): 82-86.
[2]	CHEN Jun-yi. Health Event Prediction Model Based on Dynamic and Static Features of Graph Nodes [J]. Computer and Modernization, 2023, 0(10): 39-44.
[3]	XING Shi-shuai, LIU Dan-feng, WANG Li-guo, PAN Yue-tao, MENG Ling-hong, YUE Xiao-han. Image Super-resolution Reconstruction Based on Spatial Attention Residual Network [J]. Computer and Modernization, 2023, 0(10): 45-52.
[4]	CHEN Jia-min, ZHANG Bo-quan, MAI Hai-peng. Hippocampus Segmentation Based on Feature Fusion [J]. Computer and Modernization, 2023, 0(08): 1-6.
[5]	WANG Hong, GE Hong. Cross Modal Hash Retrieval Based on Attention Mechanism and Semantic Similarity [J]. Computer and Modernization, 2023, 0(08): 44-53.
[6]	HU Rui-jie, CHE Dou. Review of Infrared Small Target Detection [J]. Computer and Modernization, 2023, 0(08): 79-86.
[7]	WANG Jie, PAN Feng, ZHANG Yan-sha, TAN Mian, YAN Xiao-bo, WANG Lin, . Surface Defect Classification of Aluminum Profiles with Weighted Non-local Modules [J]. Computer and Modernization, 2023, 0(05): 86-92.
[8]	ZHU Li-qing, LI Xiang, . Vehicle Detection of Remote Sensing Images Based on Improved YOLOv5 Algorithm [J]. Computer and Modernization, 2023, 0(05): 117-121.
[9]	ZHANG Xiao-dong, WANG Xu-ying, QIN Zi-xuan. Pump Detection Period Predicting of Pump Well Based on Feature Fusion [J]. Computer and Modernization, 2022, 0(12): 60-66.
[10]	WANG Chang-jun, PENG Cheng, LI Yong. Multi-feature Fusion Ship Target Detection Algorithm in Complex Environment [J]. Computer and Modernization, 2022, 0(11): 81-88.
[11]	PENG Lu-lu, ZHU Yuan-yuan, JIN Wen-qian, WANG Xiao-mei. Surface Defect Detection of Automotive Steel Parts Based on Improved YOLOv4 [J]. Computer and Modernization, 2022, 0(09): 32-39.
[12]	LIU Li-ting, OU Yu-yi. DGA Domain Name Detection Combining Attention Mechanisms and Parallel Hybrid Network [J]. Computer and Modernization, 2022, 0(09): 119-126.
[13]	ZENG Yi-bin, GE Hong. Cross-modal Retrieval Based on Context Fusion and Multi-similarity Learning [J]. Computer and Modernization, 2022, 0(08): 50-56.
[14]	LI Yan, LU Zheng-song, LI Qing-yun , YANG Shi-hai, ZHANG Xiao-long. Mask Detection Algorithm Based on Lightweight Structure and Re-parameterized Network [J]. Computer and Modernization, 2022, 0(07): 40-46.
[15]	LIANG Zheng-you, GENG Jing-bang, SUN Yu. Traffic Sign Recognition Algorithm Based on Improved Residual Network [J]. Computer and Modernization, 2022, 0(04): 52-57.

DF-SSD: A One-stage Small Target Detection Algorithm Based on Deconvolution and Feature Fusion

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

References ［22］

Related Articles 15

Recommended Articles

Metrics

Comments