Abstract In recent years, there has been a significant expansion in the infrastructure for video surveillance, which has resulted in an increase in the number of intelligent surveillance systems that use computer vision and pattern recognition algorithms. In this article, we offer a unique intelligent surveillance system that is based on deep learning and is utilized for the management of road vehicles. In other words, the system is able to label any vehicle via the use of computer vision, and as a result, it is able to quickly distinguish cars that have visual tags. This capability allows the system to extract the vehicle visual tags that are present on urban roadways. The visual tags that are discussed in this article include the license plate number, the color of the car, and the kind of vehicle. These visual tags also contain a variety of other attributes, such as passing location and passing time. In this work the particular area video footage was retrieved. Then the formation of clustered hub was done for the fusion of the several footage data . The violation based vehicle details are grouped by using the Gaussian mac clustering approach. Then from the grouped features the specialized features according to the required event was isolated using the Metaheuristic chaos vortex optimization. Finally the proposed hyperbolic back propagate boltzmann neural network (HBPBNN) architecture detects the vehicle and its related violations in a precise manner. The whole experimentation was carried out in a real time database. On-road experimental findings show that the algorithm outperforms the most cutting-edge vehicle recognition algorithm in testing data sets. The findings of the comparative assessment indicated that the recommended model performed better than the other models in use.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|