A Real-time system for the automatic identification of motorcycle - using Artificial Neural networks

This paper presents a real-time system for the automatic identification of moving motor cycle. The system uses the latest image processing techniques for the processing of images and artificial neural networks is used for the recognition purpose. The proposed system is intended for the automatic control and monitoring of a motor cycle parking lot. One digital camera is to be fixed suitably in the inward passage of the parking area and another digital camera is to be fixed suitably in the outward passage of the parking area to capture the image of the motor cycle with cyclists. Form the first image, the vehicle number from number plate of the motor cycle and the head features of the cyclist are extracted and are stored in the real-time distributed database. From the image of the return trip of the motor cycle, the same details are again extracted and the corresponding image features are retrieved from the database. These features are then compared and provide valuable information to the parking area controller. The proposed system is an extension of the works presented in previous papers. The system works well under various lighting conditions and can be employed as Real-time Vehicle Monitoring and Information System.