Research on Recognition Method of Truck Colors Based on Convolutional Neural Network Model

Truck color is one of the important information of vehicle ownership in intelligent transportation system. Aiming at the problem that truck color recognition is easy to be affected by complex scene and illumination changes, this paper studied the truck recognition method based on convolutional neural network, constructed the truck color image set, transformed truck vehicle image samples into RGB, HSV, and LAB color spaces. A truck color recognition method based on convolutional neural network model is proposed and experimental research is proposed. The experimental results show that the performance of the proposed convolutional neural network model based on LAB color space is better than that of proposed convolutional neural network model based on HSV color space, and achieved a verification accuracy at 96.36%.