An Overview of Vehicle Target Detection on Highway

This paper provides a comprehensive literature review in the field of highway vehicle target detection, aiming to summarize and analyze the evolution from traditional methods to deep learning-based methods, classify and analyze the standard techniques of highway vehicle target detection, and explore their performance characteristics, applicable scenarios, and future development directions. Vehicle target detection on highways has always been the core of the intersection of multiple disciplines, and highway vehicle target detection is of great significance in traffic flow monitoring, intelligent driver assistance systems, and traffic accident prevention. Its impact has been widely reflected in social, scientific, and economic fields. First, this paper summarizes and introduces traditional target detection, deep learning-based target detection (which includes single-stage and two-stage target detection), and other target detection methods. In addition, through an in-depth combing and analysis of domestic and international literature, this paper summarizes analyzes, and compares the relevant data, and puts forward the advantages and shortcomings of each method, as well as the direction and possible trends of future research.