MNNP: Design and Development of Traffic Sign Identification and Recognition System to Support Smart Vehicles using Modified Neural Network Principles
Anitha G. S., R. Hemalatha, N Sakthisaravanan., K R Kavitha, Mani Tamilselvi, Ata Kishore Kumar · 2023
In recent times, the acceleration of automotive and computer vision technologies has sparked significant interest in autonomous vehicles. These vehicles' capacity to operate securely and effectively hinges on their adeptness at accurately identifying traffic signs. Consequently, traffic sign recognition has emerged as a pivotal element within autonomous driving systems. Scientists have been hard at work looking at various approaches to accomplishing this recognition, mostly drawing on machine learning and deep learning techniques. To address this issue, we offer a Modified Neural Network approach that combines the Haar cascade algorithm with a MobileNet model classifier. Our developed model is rigorously trained on the GTSRB dataset and then put to the test on a wide variety of traffic sign types. The culmination of our efforts yields a remarkable testing accuracy rate of 97.25%. This finding highlights the possible effectiveness of our technique in improving autonomously vehicle traffic sign identification, which in turn helps to improve the security and effectiveness of autonomous driving systems.