The thesis is conducted at the Ternopil National Technical University named after Ivan Puluj, specifically at the Faculty of Computer Information Systems and Software Engineering, Department of Computer Sciences. The research focuses on the development of software for detecting and recognizing road signs using convolutional neural networks (CNNs). The work is carried out by a student of the 6th course, group SNM-61, major in Computer Sciences. The supervisor is Prof. Kunańets N. Y., and the thesis is approved by the head of the Department of Computer Sciences. The research aims to develop an algorithm and software tool for detecting and recognizing road signs. The study analyzes existing systems and algorithms for road sign detection and recognition, comparing their technical characteristics and performance. It identifies the most effective algorithms for detecting and recognizing road signs, such as the Gabor filter for shape detection and the Maximum Stable Extreme Regions (MSER) method. The research also explores the use of artificial and convolutional neural networks for road sign recognition. The study uses the GTSRB and GTSDB datasets for training and testing. The proposed software implementation, based on the Keras and OpenCV libraries, demonstrates a high identification rate for road signs. The research highlights the importance of using multiple algorithms together to improve the robustness and accuracy of road sign detection and recognition. The study also discusses the practical applications of the developed software in various fields, including traffic monitoring and video surveillance. The research concludes that the combination of the Gabor filter and MSER method is the most effective for road sign detection and recognition, and the proposed software shows a high level of accuracy and stability.The thesis is conducted at the Ternopil National Technical University named after Ivan Puluj, specifically at the Faculty of Computer Information Systems and Software Engineering, Department of Computer Sciences. The research focuses on the development of software for detecting and recognizing road signs using convolutional neural networks (CNNs). The work is carried out by a student of the 6th course, group SNM-61, major in Computer Sciences. The supervisor is Prof. Kunańets N. Y., and the thesis is approved by the head of the Department of Computer Sciences. The research aims to develop an algorithm and software tool for detecting and recognizing road signs. The study analyzes existing systems and algorithms for road sign detection and recognition, comparing their technical characteristics and performance. It identifies the most effective algorithms for detecting and recognizing road signs, such as the Gabor filter for shape detection and the Maximum Stable Extreme Regions (MSER) method. The research also explores the use of artificial and convolutional neural networks for road sign recognition. The study uses the GTSRB and GTSDB datasets for training and testing. The proposed software implementation, based on the Keras and OpenCV libraries, demonstrates a high identification rate for road signs. The research highlights the importance of using multiple algorithms together to improve the robustness and accuracy of road sign detection and recognition. The study also discusses the practical applications of the developed software in various fields, including traffic monitoring and video surveillance. The research concludes that the combination of the Gabor filter and MSER method is the most effective for road sign detection and recognition, and the proposed software shows a high level of accuracy and stability.