Development of a method for detecting traffic violations based on artificial intelligence for a Telegram bot
Students Name: Pasichnyk Vasyl Vasylovych
Qualification Level: magister
Speciality: Information Technology Design
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2022-2023 н.р.
Language of Defence: ukrainian
Abstract: Pasichnyk V.V., Pleskanka N. M. (supervisor). Development of a method for detecting traffic violations based on artificial intelligence for a telegram bot. Master’s degree work – National University “Lviv Polytechnic”, Lviv, 2021. Extended abstract. The master’s qualification work is devoted to the development of a method for detecting traffic violations based on artificial intelligence for a telegram bot. The first section of the report describes the design and analysis of the requirements of the system under development. The problem of recording violations was identified [1] and given all the difficulties, this task could not be solved by classical programming methods, so the concepts of machine and deep learning were considered, with the help of which it was finally possible to make progress in the field of vehicle offense recognition. The report described the principles of artificial intelligence functioning, how the process of object recognition in machine learning takes place, gave an example of finding special features of an object and taking them into account to distinguish between two similar objects of different classes, and also demonstrated the principles of developing various telegram bots. The process of formation of neural networks from initial models to modern ones was shown. The first model of artificial intelligence "Perceptron" and its impact on the further development of this field was presented. The second section describes how artificial intelligence can improve the process of combating crime [2]. Thus, the expediency of creating an own system that would specialize in solving non-standard tasks was determined, and specific requirements for it were identified. Existing systems for fixing speeding violations are presented, their analysis and comparison are carried out. The third section describes the software implementation of the software product. Different algorithms for object recognition and classification in video are reviewed and analyzed, and the main algorithm for development is selected. The features of the development environment are presented, and the list of tools used is given. Object of research: method of detection of traffic violations. Subject of research: machine learning technology for detecting violations. The purpose of the study: to develop a method for detecting traffic violations based on artificial intelligence for a telegram bot. As a result, a telegram bot was developed that allows detecting speeding violations using artificial intelligence. List of used literary sources: machine learning, artificial intelligence, violation, Telegram, bot. Y. Mo, G. Han, H. Zhang, X. Xu, and W. Qu, “Highlight-assisted nighttime vehicle detection using a multi-level fusion network and label hierarchy,” Neurocomputing, ст. 13–23, 2019. Z. Liu, Y. Cai, H. Wang et al., “Robust target recognition and tracking of self-driving cars with radar and camera information fusion under severe weather conditions,” IEEE Transactions on Intelligent Transportation Systems, 2021.