Research on Adaptive Algorithms for Optimization of Cross-Country Vehicle Configuration Comparison using Deep Learning

• Published in: 2024 International Conference on Computers, Information Processing and Advanced Education (CIPAE) pp. 794 - 797
• Main Authors: Liu, Lei, Ding, Qian, Li, Hongqing, Zhang, Fan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.08.2024
• Subjects: Accuracy; Adaptation models; adaptive algorithm; Automobiles; Convolutional neural networks; Deep learning; Feature extraction; Image recognition; License plate recognition; multi-column deep convolutional network; multinational vehicle configuration comparison; Optimization; Training; vehicle type recognition
• DOI: 10.1109/CIPAE64326.2024.00150

This study aims to optimize adaptive algorithms for cross-country vehicle configuration comparison by using deep learning techniques, especially multi-column Deep Convolutional network (MDCNN) vehicle recognition models. The algorithm can automatically identify and compare vehicle configuration differences in different countries and provide consumers with accurate vehicle information. Firstly, a database of automobile images containing various brands and models is constructed, and MDCNN model is used to extract and recognize the features of the images. Through extensive training and verification, the model is able to accurately identify the vehicle model and its configuration differences in different countries. The research further developed an adaptive algorithm, which can automatically select the vehicle with specific configuration according to the needs of users, and carry out cross-country configuration comparison. The experimental results show that the accuracy of the proposed MDCNN model is as high as 95%, which is far superior to the traditional license plate recognition and vehicle classification methods. Adaptive algorithms show a high degree of flexibility and accuracy when dealing with complex cross-country configuration comparisons, which can effectively help users make purchase decisions, and also provide market positioning insights for automobile manufacturers.