An Efficient Method For Dark License Plate Detection

License plate detection from images is an important step in vehicle license plate recognition for automated transport system. Dark images and complex background make license plate detection problems challenging. In this paper, a novel method of based on rectangle window was proposed to resolve license plate detection problems. The proposed method firstly enhances images using traditional method. Secondly input enhanced images are converted from RGB color space to HSI color space, then we extract I component from HSI color space and get vertical edge images using Soble operator. Thirdly a novel method was proposed to remove most of the background curves and noise edges. Fourthly, we propose an effective algorithm search the license plate region by a rectangle window in the residual edge. Finally segment the license plate detection from the original dark license plate images. Experimental results show that the license plate detection by our method is better than other conventional methods. The proposed method also shows the effectiveness and robustness.