[1]邓 权,林明星.基于改进 SSD 的海洋生物检测算法[J].计算机技术与发展,2022,32(04):51-56.[doi:10. 3969 / j. issn. 1673-629X. 2022. 04. 009]
　DENG Quan,LIN Ming-xing.Marine Organism Detection Algorithm Based on Improved SSD[J].,2022,32(04):51-56.[doi:10. 3969 / j. issn. 1673-629X. 2022. 04. 009]
点击复制

基于改进 SSD 的海洋生物检测算法()

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

卷:
32
期数:
2022年04期
页码:
51-56
栏目:
图形与图像
出版日期:
2022-04-10

文章信息/Info

Title:
Marine Organism Detection Algorithm Based on Improved SSD
文章编号:
1673-629X(2022)04-0051-06
作者:
邓 权1; 2; 3 ; 林明星1; 2; 3
1. 山东大学 机械工程学院,山东 济南 250061;
2. 高效洁净机械制造教育部重点实验室,山东 济南 250061;
3. 机械工程国家级实验教学示范中心,山东 济南 250061
Author(s):
DENG Quan1; 2; 3 ; LIN Ming-xing1; 2; 3
1. School of Mechanical Engineering,Shandong University,Jinan 250061,China;
2. Key Laboratory of High Efficiency and Clean Mechanical Manufacture,Jinan 250061,China;
3. National Demonstration Center for Experimental Mechanical Engineering Education,Jinan 250061,China
关键词:
SSD; 小目标; 特征融合; 特征增强; 实时性
Keywords:
SSD; small object; feature fusion; feature enhancement; real time
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 04. 009
摘要:
为了满足海洋生物检测对精度和实时性的要求,提出了一种基于改进 SSD 算法( single shot multibox detector) 的海洋生物检测算法。 针对 SSD 算法浅层特征层语义信息不足、小目标检测效果差等问题,设计了特征融合模块和特征增强模块。 特征融合模块通过融合不同特征层,丰富了浅层特征层的语义信息以及深层特征层的细节信息,综合上下文信息提高检测效果。 为了进一步提高浅层特征层的语义信息,提出了特征增强模块,通过引入空洞卷积以及多尺度的卷积核,综合不同感受野信息以提高改进算法对小目标的检测效果。 改进算法在仅增加少量计算量和参数量的情况下,全面提高了算法对海洋生物目标的检测准确率。 结果表明,改进算法在海洋生物数据集中的平均精度( mAP) 达 80. 8% ,比原始网络提高了 5% ,检测速度( FPS) 为 74,略低于 SSD 算法,但远高于其他改进算法。 改进算法能在保持实时性的同时取得较高的检测精度,能够满足海洋生物检测要求。
Abstract:
In order to meet the accuracy and real-time requirements of marine organism detection,a marine organism detection algorithm based on improved SSD algorithm ( single shot multibox detector) is proposed. For the problem of insufficient semantic information in the shallow feature layer of the SSD algorithm and poor detection of small targets,a feature fusion module and a feature enhancement module are designed. The feature fusion module enriches the semantic information of the shallow feature layer and the detailed information of the? ? ? deep feature layer by fusing different feature layers, and integrates context information to improve the detection accuracy. In order to further improve the semantic information of the shallow feature layer,a feature enhancement module is proposed,which integrates different receptive field information by introducing dilated convolution and multi-scale convolution kernels to improve the detection effect of the improved algorithm on small targets. The improved algorithm comprehensively improves the accuracy of the algorithm’s detection of marine organism while only adding a small amount of calculation and parameters. Experimental results show that the average accuracy ( mAP) of the improved algorithm in the marine organism data set is 80. 8% ,which is 5% higher than the original network,the detection speed ( FPS) is 74,slightly lower than the SSD algorithm,but much higher than other improved algorithms. The improved algorithm can achieve high detection accuracy while maintaining real-time performance,which can meet the requirements of marine organism detection.

相似文献/References:

[1]艾丽蓉,李凯.基于Postgres-XL 的数据管理优化技术研究[J].计算机技术与发展,2018,28(03):11.[doi:TP311]
　AI Li-rong,LI Kai.Study on Optimization Technology of Data Management Based on Postgres-XL[J].,2018,28(04):11.[doi:TP311]
[2]杜华,刘华春.云数据中心下重复数据删除技术研究[J].计算机技术与发展,2019,29(02):157.[doi:10．3969/j．issn．1673－629X．2019．02．033]
　DU Hua,LIU Huachun.Research on Deduplication of Data in Cloud Data Center[J].,2019,29(04):157.[doi:10．3969/j．issn．1673－629X．2019．02．033]
[3]马浩良,谢林柏.基于 SSD 的不平衡样本车辆检测与识别[J].计算机技术与发展,2019,29(12):135.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 024]
　MA Hao-liang,XIE Lin-bo.Vehicle Detection and Recognition in Unbalanced Samples Based on SSD[J].,2019,29(04):135.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 024]
[4]唐小敏,舒远仲,刘文祥,等.基于 SSD 深度网络的河道漂浮物检测技术研究[J].计算机技术与发展,2020,30(09):154.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 028]
　TANG Xiao-min,SHU Yuan-zhong,LIU Wen-xiang,et al.Research on River Floating Object Detection Technology Based on SSD Deep Network[J].,2020,30(04):154.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 028]
[5]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.[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(04):1.[doi:10. 3969 / j. issn. 1673-629X. 2022. 03. 001]
[6]石露露,廖光忠.改进 YOLOv5s 的明渠漂浮垃圾实时检测方法[J].计算机技术与发展,2023,33(09):83.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 013]
　SHI Lu-lu,LIAO Guang-zhong.Real-time Detection Method of Floating Garbage in Open Channels Based on Improved YOLOv5s[J].,2023,33(04):83.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 013]

更新日期/Last Update: 2022-04-10