Paper
14 October 2022 Improved site truck gate detection algorithm based on deep learning
Chang Liu, Yutong Cui, Taizhou Yang, Guopeng Zhang, Wei Ren, Dawei Jiang, Wei Chen
Author Affiliations +
Proceedings Volume 12343, 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022); 123432M (2022) https://doi.org/10.1117/12.2651112
Event: 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022), 2022, Qingdao, China
Abstract
At present, with the rapid development of industrialization, construction and real estate industries have also been developed, but the verification and entrance control of vehicles carrying large equipment and cement also need to be checked manually, causing a lot of artificial burden. In this paper, an improved algorithm based on deep learning is proposed to realize unmanned gate detection. The distance between the vehicle and the gate can be judged by real-time detection of whether there is a vehicle passing by, so as to make the instruction of lifting the gate or lowering the gate. Compared with the traditional gate detection, the use of a lower cost of ordinary cameras can be completed, so as to save a lot of cost, and can detect the vehicle information in real time, consistent with China's industrial intelligence strategy. In this paper, YoLoV4 algorithm is used and mobilnet-V3 algorithm is used to replace the original main network structure, which greatly reduces the number of parameters of the algorithm structure and improves the running speed of the algorithm, so that it can reach the real-time state. VOC data sets were used for training and detection, and the experiment proved that the speed was greatly improved and real-time vehicle detection could be realized.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chang Liu, Yutong Cui, Taizhou Yang, Guopeng Zhang, Wei Ren, Dawei Jiang, and Wei Chen "Improved site truck gate detection algorithm based on deep learning", Proc. SPIE 12343, 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022), 123432M (14 October 2022); https://doi.org/10.1117/12.2651112
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Convolution

Detection and tracking algorithms

Evolutionary algorithms

Feature extraction

Target detection

Neural networks

Computer vision technology

RELATED CONTENT

Research on coal gangue detection algorithm based on YOLOv4
Proceedings of SPIE (December 12 2021)
An improved MDNet target tracking algorithm
Proceedings of SPIE (August 04 2022)
Improved tongue detection algorithm based on YOLO V4
Proceedings of SPIE (October 19 2023)
A joint mechanism algorithm for object detection
Proceedings of SPIE (May 06 2019)

Back to Top