[1]郭健忠,余腾飞,崔玉定,等.基于改进 SSD 的车辆小目标检测算法研究[J].计算机技术与发展,2022,32(03):1-7.[doi:10. 3969 / j. issn. 1673-629X. 2022. 03. 001]
　GUO Jian-zhong,YU Teng-fei,CUI Yu-ding,et al.Research on Vehicle Small Target Detection Algorithm Based on Improved SSD[J].,2022,32(03):1-7.[doi:10. 3969 / j. issn. 1673-629X. 2022. 03. 001]
点击复制

基于改进 SSD 的车辆小目标检测算法研究()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年03期
页码:
1-7
栏目:
人工智能
出版日期:
2022-03-10

文章信息/Info

Title:
Research on Vehicle Small Target Detection Algorithm Based on Improved SSD
文章编号:
1673-629X(2022)03-0001-07
作者:
郭健忠1 ; 余腾飞1 ; 崔玉定1 ; 周兴林2
1. 武汉科技大学 汽车与交通工程学院,湖北 武汉 430065;
2. 武汉科技大学 机械自动化学院,湖北 武汉 430065
Author(s):
GUO Jian-zhong1 ; YU Teng-fei1 ; CUI Yu-ding1 ; ZHOU Xing-lin2
1. School of Automobile and Traffic Engineering,Wuhan University of Science and Technology,Wuhan 430065,China;
2. School of Mechanical Automation,Wuhan University of Science and Technology,Wuhan 430065,China
关键词:
卷积神经网络; 目标检测; 小目标; 特征增强; 特征融合
Keywords:
convolutional neural network; object detection; small target; feature enhancement; feature fusion
分类号:
TP301. 6
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 03. 001
摘要:
针对 SSD( single shot multibox detector,单步多盒检测) 算法在车辆的自动紧急制动( AEB) 中对远方目标检测效果差、检测速度慢、对硬件资源需求高的问题,提出了一种基于 SSD 的改进算法。首先用 MobileNetv2 替换 SSD 中的 AGG-16作为检测网络,以减少参数数量和计算量,降低网络对硬件性能的需求;其次,提出了特征增强和融合的方法,反复挖掘目标信息,并把不同特征层的信息进行融合,以提高对小目标检测的能力;最后,对先验框解码过程进行改进,减少网络需要解码的先验框数量,再次减少计算量，提高网络检测速度，并调整先验框的尺寸，进一步增强小目标检测的能力。把改进后的网络和 SSD300、YOLO、 MobileNetv2-SSD 等网络在 KITTI 数据集上进行检测和对比分析,实验结果表明,改进后的网络对小目标识别的速度有所加快,鲁棒性更好,准确率更高,同时也降低了对硬件配置资源的需求。
Abstract:
In order to solve the problems of poor detection effect, slow detection speed and high demand for hardware resources, an improved algorithm based on SSD is proposed in automatic emergency braking ( AEB) of vehicles. Firstly,AGG-16? ? ?in SSD is replaced by MobileNetv2 as the detection network to reduce the number of parameters and the amount of computation, and reduce the demand of network hardware performance. Secondly,the method of feature enhancement and fusion is proposed,which excavates target information repeatedly and integrates the information of different feature layers to improve the ability of small target detection. Finally,the decoding process of the priori box is improved to reduce the number of priori boxes that need to be decoded,reduce the amount of computation again,improve the speed of network detection,and adjust the size of the priori box to further enhance the ability of small target detection.The improved network is detected and compared with SSD300,YOLO,MobilenetV2-SSD and other networks on KITTI dataset. The experimental results show that the improved network has higher speed, better robustness, higher accuracy, and reduces the demand for hardware configuration resources.

相似文献/References:

[1]刘晓明 李毓蕙 高燕 郑华强.基于目标区域清晰显示的H.264编码策略[J].计算机技术与发展,2010,(06):29.
　LIU Xiao-ming,LI Yu-hui,GAO Yan,et al.A Coding Strategy of H.264 Based on High-definition Display of Target Region[J].,2010,(03):29.
[2]刘翔 吴谨 祝愿博 康晓晶.基于视频序列的目标检测与跟踪技术研究[J].计算机技术与发展,2009,(11):179.
　LIU Xiang,WU Jin,ZHU Yuan-bo,et al.A Study of Object Detecting and Tracking Based on Video Sequences[J].,2009,(03):179.
[3]曙光 张超 蔡则苏.基于改进的混合高斯模型的目标检测方法[J].计算机技术与发展,2012,(07):60.
　SHU Guang,ZHANG Chao,CAI Ze-su.Target Detection Method Based on Improved Gaussian Mixture Model[J].,2012,(03):60.
[4]刘洁,李目,周少武.一种混沌混合粒子群优化RBF神经网络算法[J].计算机技术与发展,2013,(08):181.
　LIU Jie[],LI Mu[],ZHOU Shao-wu[].An Algorithm of Chaotic Hybrid Particle Swarm Optimization Based on RBF Neural Network[J].,2013,(03):181.
[5]蒋翠清,孙富亮,吴艿芯. 基于相对欧氏距离的背景差值法视频目标检测[J].计算机技术与发展,2015,25(01):37.
　JIANG Cui-qing,SUN Fu-liang,WU Nai-xin. Video Object Detection of Background Subtraction Method Based on Relative Euclidean Distance[J].,2015,25(03):37.
[6]卢官明,衣美佳. 步态识别关键技术研究[J].计算机技术与发展,2015,25(07):100.
　LU Guan-ming,YI Mei-jia. Research on Critical Techniques in Gait Recognition[J].,2015,25(03):100.
[7]高翔,朱婷婷,刘洋. 多摄像头系统的目标检测与跟踪方法研究[J].计算机技术与发展,2015,25(07):221.
　GAO Xiang,ZHU Ting-ting,LIU Yang. Research of Target Detection and Tracking Method for Multi-camera System[J].,2015,25(03):221.
[8]崔凤焦.表情识别算法研究进展与性能比较[J].计算机技术与发展,2018,28(02):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
　CUI Feng-jiao.Ｒesearch and Performance Comparison of Facial Expression Ｒecognition Algorithm[J].,2018,28(03):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
[9]章文洁[][],黄旻[],张桂峰[]. 滤光片多光谱成像中运动目标场景误配准修正[J].计算机技术与发展,2016,26(01):18.
　ZHANG Wen-jie[][],HUANG Min[],ZHANG Gui-feng[]. Misregistration Correction for Moving Object Scene in Filter-type Multispectral Imaging[J].,2016,26(03):18.
[10]张丹丹,李雷. 基于PCANet-RF的人脸检测系统[J].计算机技术与发展,2016,26(02):31.
　ZHANG Dan-dan,LI Lei. Face Detection System Based on PCANet-RF[J].,2016,26(03):31.
[11]许必宵,宫 婧,孙知信.基于卷积神经网络的目标检测模型综述[J].计算机技术与发展,2019,29(12):87.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 016]
　XU Bi-xiao,GONG Jing,SUN Zhi-xin.A Survey of Object Detection Models Based on Convolutional Neural Networks[J].,2019,29(03):87.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 016]
[12]鲍润嘉,侯庆山,邢进生.一种改进的 SSD 网络车载图像检测方法[J].计算机技术与发展,2021,31(02):85.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 016]
　BAO Run-jia,HOU Qing-shan,XING Jin-sheng.An Improved SSD Network Vehicle Image Detection Method[J].,2021,31(03):85.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 016]
[13]张永福,宋海林,班 越,等.融合特征的深度学习遥感图像目标检测模型[J].计算机技术与发展,2021,31(09):48.[doi:10. 3969 / j. issn. 1673-629X. 2021. 09. 009]
　ZHANG Yong-fu,SONG Hai-lin,BAN Yue,et al.Fusion Features Based Deep Learning Remote Sensing Image Target Detection Model[J].,2021,31(03):48.[doi:10. 3969 / j. issn. 1673-629X. 2021. 09. 009]
[14]鲍先富,强赞霞,李丹阳,等.基于组卷积特征融合的 One-Stage 目标检测模型[J].计算机技术与发展,2021,31(11):86.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 015]
　BAO Xian-fu,QIANG Zan-xia,LI Dan-yang,et al.One-Stage Target Detection Model Based on Group ConvolutionFeature Fusion[J].,2021,31(03):86.[doi:10. 3969 / j. issn. 1673-629X. 2021. 11. 015]
[15]李骏峰,杨小军,张凯望.基于 YOLOX-L 算法的安全帽佩戴检测方法[J].计算机技术与发展,2022,32(09):100.[doi:10. 3969 / j. issn. 1673-629X. 2022. 09. 016]
　LI Jun-feng,YANG Xiao-jun,ZHANG Kai-wang.Safety Helmet Wearing Detection Method Based on YOLOX-L Algorithm[J].,2022,32(03):100.[doi:10. 3969 / j. issn. 1673-629X. 2022. 09. 016]
[16]曾碧卿,杨 睿,李一娴,等.基于卷积神经网络的零件圆检测方法[J].计算机技术与发展,2023,33(11):64.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 010]
　ZENG Bi-qing,YANG Rui,LI Yi-xian,et al.Part Circle Detection Method Based on Convolutional Neural Network[J].,2023,33(03):64.[doi:10. 3969 / j. issn. 1673-629X. 2023. 11. 010]
[17]韩 彤,曹铁勇,郑云飞,等.迷彩伪装目标检测的视觉特征偏好研究[J].计算机技术与发展,2023,33(12):193.[doi:10. 3969 / j. issn. 1673-629X. 2023. 12. 027]
　HAN Tong,CAO Tie-yong,ZHENG Yun-fei,et al.Research on Visual Feature Bias of Camouflaged Object Detection[J].,2023,33(03):193.[doi:10. 3969 / j. issn. 1673-629X. 2023. 12. 027]
[18]孙俊洋,符运来,吕晶,等.基于改进YOLOv7模型的海参苗计数方法研究[J].计算机技术与发展,2024,34(11):166.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0233]
　SUN Jun-yang,FU Yun-lai,LYU Jing,et al.Study on Counting Method of Sea Cucumber Seedlings Based on Improved YOLOv7 Model[J].,2024,34(03):166.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0233]

更新日期/Last Update: 2022-03-10