Abstract
According to the characteristics of YOLOV5, a method based on YOLOV5 is proposed. First, this method is used for the identification of small objects. Secondly, this paper conducts experiments on the improved YOLOV5 algorithm and conducts experiments on it, and compares the performance of the two methods. In modern detection technology, the most commonly used is the target detection technology. In recent years, small object detection has been an important research topic in industrial inspection. The small target has a small number of pixels, and its features are blurred. Larger targets have lower detection rates and higher false detection rates. In order to improve the accuracy and accuracy of target detection, this paper proposes a method based on depthwise convolution and K-means clustering, which realizes the clustering of the dimension and aspect ratio of the target image. On the basis of CCPD, the YOLOv5 algorithm is optimized and compared with the original YOLOv5 algorithm. Experiments show that the improved YOLOv5 can better detect smaller targets, so that the recall rate and average precision of small targets are significantly improved.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Leifeng, C., Lisa, H., Lihui, X., et al.: Target behavior knowledge discovery method based on deep learning. Comput. Digital Eng. 50(03), 532–537 (2022)
Shi, L., Jing, M.E., Fan, Y.B., et al.: Segmentation detection algorithm based on R-CNN algorithm. J. Fudan Univ. (Nat. Sci. Ed.) 59(04), 412–418 (2020)
Zhiyong, S., Jiahui, G.: Vehicle target detection based on improved fast R-CNN. Mod. Comput. 20, 74–79 (2021). https://doi.org/10.3969/j.issn.1007-1423.2021.20.015
Denig, W.F.: The international council for science world data system stewardship award 2014 presented to Dr. Robert Redmon of the national geophysical data center. Space Weather 13(1), 1 (2015). https://doi.org/10.1002/2014SW001136
Lei, G., Qiulong, W., Wei, X., et al.: Small target detection algorithm based on improved YOLOv5. J. Univ. Electron. Sci. Technol. 51(2), 251–258 (2022). https://doi.org/10.12178/1001-0548.2021235
Cheng, L.: Research on small target detection algorithm based on improved YOLOv5. Changjiang Inf. Commun. 34(9), 30–33 (2021). https://doi.org/10.3969/j.issn.1673-1131.2021.09.010
Deshuo, K., Wanmi, C., Nan, Y.: Target recognition algorithm of service robot based on improved YOLOv5. Ind. Control Comput. 35(2), 82–83 (2022). https://doi.org/10.3969/j.issn.1001-182X.2022.02.032
Shixiong, W., Mingwu, Y., Zhilian, G., et al.: A photoelastic data acquisition and processing system. J. Hefei Univ. Technol. (Nat. Sci. Ed.), (S1), 71–73 (1993)
Krebs, E.E.: VA state of the art (SOTA) conference on non-pharmacological approaches to chronic musculoskeletal pain management. Ann. Behav. Med. 51(suppl.1), S2364–S2365 (2017)
Wang, Z.L.: An effective model detection method based on LTL and petri nets. Comput. Appl. 26(10), 2490–2493 (2006)
Changming, L., Hongchen, Z., Chao, W., et al.: An efficient Yin Yang K-means clustering algorithm. J. Jilin Univ. (Sci. Ed.) 59(6), 1455–1460 (2021). https://doi.org/10.13413/j.cnki.jdxblxb.2020406
Liao, Q., Jiang, Y., Le, Q., et al.: Hot deformation behavior and processing map development of AZ110 alloy with and without addition of La-rich Mish Metal. Mater. Sci. Technol. (Engl. Version) 61(2), 1–15 (2021)
Fengdi, L., Weixing, S., Jiefang, W., et al.: Study on detection method of pine wood nematode disease tree based on YOLOv3 CIOU. J. Shandong Agric. Univ. (Nat. Sci. Ed.) 52(2), 224–233 (2021). https://doi.org/10.3969/j.issn.1000-2324.2021.02.012
Qifan, G., Lei, L., Lei, Z., et al.: Multi-scale feature fusion network based on feature pyramid. J. Eng. Math. 37(5), 521–530 (2020)
Yaqian, L., Chengyuan, G., Cunjun, X., et al.: Target detection network based on refined multi-scale depth features. Acta Electron. Sin. 48(12), 2360–2366 (2020). https://doi.org/10.3969/j.issn.0372-2112.2020.12.011
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zhao, Y., Shi, Y., Wang, Z. (2022). The Improved YOLOV5 Algorithm and Its Application in Small Target Detection. In: Liu, H., et al. Intelligent Robotics and Applications. ICIRA 2022. Lecture Notes in Computer Science(), vol 13458. Springer, Cham. https://doi.org/10.1007/978-3-031-13841-6_61
Download citation
DOI: https://doi.org/10.1007/978-3-031-13841-6_61
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-13840-9
Online ISBN: 978-3-031-13841-6
eBook Packages: Computer ScienceComputer Science (R0)