Title: Technical evaluation on improved Hough transform for monitoring traffic sign images under depth algorithm

Authors: Jie Ding; Guotao Zhao

Addresses: College of Technology, Hubei Engineering University, 432000, Hubei, China ' School of Foreign Languages, Hubei Engineering University, 432000, Hubei, China

Abstract: The experiment drew the accuracy curve, training loss curve, and test loss curve of different traffic sign sizes and the network model after feature fusion, and compared the algorithm in this paper with the artificial neural network (ANN) algorithm and the random forest algorithm. The research results showed that the classification accuracy of the multi-scale model combining the three branch networks was as high as 99%. The classification accuracy of the extreme learning machine (ELM) classifier was better than that of the softmax and support vector machine (SVM) classifiers. Compared with the ANN algorithm and the random forest algorithm, the classification time of the algorithm in this paper was reduced by 125 ms and 155 ms, respectively. The classification accuracy reached 99%, which lays the foundation for the innovation and development of traffic sign image processing technology.

Keywords: traffic sign image; improved Hough transform; deep algorithm; multi-scale feature fusion; convolutional neural network.

DOI: 10.1504/IJDSDE.2025.146948

International Journal of Dynamical Systems and Differential Equations, 2025 Vol.14 No.1/2, pp.82 - 99

Received: 11 May 2024
Accepted: 02 Dec 2024
Published online: 27 Jun 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article