An Adaptive Method for Lane Marking Detection Based on HSI Color Model

Lane marking detection is the problem of estimating the lane boundary of a road on the image captured by a camera. This paper proposed an adaptive method based on HSI color model to detect lane marking. First, we convert ROB-based image to its HSI-based image. However, HSI color model is improved by the change in the way to calculate the intensity (I) component from RGB color images. From observing the color images of the road scene in HSI color space, we utilized the limited range of color. Hence, H, S and I component are used in this method. Just simple operations, we can detect lane marking in various road images. By comparing the results of the proposed method with other methods using ROB color model and the same method in classical HSI color model which doesn't change the intensity component, the proposed method can label the location of lane marking accurately.