Target recognition technology based on improved YOLOV7

Lei Huang; Chao Wang; Zhiyuan Li

doi:10.1117/12.2674669

28 April 2023 Target recognition technology based on improved YOLOV7

Lei Huang, Chao Wang, Zhiyuan Li

Proceedings Volume 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022); 126261I (2023) https://doi.org/10.1117/12.2674669
Event: International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 2022, Zhuhai, China

Abstract

As the environment of special targets is complex and constantly changing, better requirements are put forward for rapid and high-precision target detection and recognition. In this paper, the improved YOLOV7 algorithm is adopted. First, Kmeans is used to match the new anchor coordinates, and multiple detection scales are added to improve the detection accuracy; Secondly, the attention mechanism module is integrated into the feature extraction network Darknet-53 to obtain important features; Then, taking advantage of the lightweight technology of Ghost module, Ghost BottleNeck composed of Ghost modules is introduced to replace the Neck module in YOLOV7, which greatly reduces the parameters and computation of the network model; Finally, IOU_ Nms is modified to DIOU_ Nms is used to optimize the loss function. experiments show that the accuracy and real-time performance of the algorithm are improved.

Citation Download Citation

Lei Huang, Chao Wang, and Zhiyuan Li "Target recognition technology based on improved YOLOV7", Proc. SPIE 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 126261I (28 April 2023); https://doi.org/10.1117/12.2674669

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available