Vigilate technology is one of the best ANPR camera for detect classification of vehicle type
Detect vehicle type
Automatic vehicle classification has been more important in intelligent transportation systems and visual traffic surveillance systems in recent decades. It is critical to restrict vehicle movement as much as possible in countries that have imposed a lock down (mobility constraints help reduce the spread of COVID-19). It is critical to detect vehicles from photos and classify them into distinct classes for an effective visual traffic surveillance system (e.g., bus, car, and pickup truck).
Most prior research projects have solely concentrated on increasing the percentage of correct predictions, which has poor real-time performance and uses more computational resources. This research study proposes a new technique for vehicle type classification to highlight the issues with classifying imbalanced data.
The data is first gathered from the centeral traffic Camera dataset. Adaptive histogram equalization and the Gaussian mixture model are also used to improve the quality of collected vehicle photos and to recognize cars in denoised photographs. The feature vectors from the identified cars are then extracted using the Steerable Pyramid Transform and the Weber Local Descriptor.
Finally, the collected characteristics are fed into a vehicle classification algorithm using an ensemble deep learning approach. The proposed ensemble deep learning approach achieved 99.13 percent and 99.28 percent classification accuracy on the Traffic Camera dataset in the simulation phase. On both datasets, the achieved results are effective when compared to existing benchmark methodologies.