This paper presents a novel vehicle color classification technique for classifying vehicles into seven categories under different lighting conditions via color correction. First, to reduce lighting effects, a mapping function is built to minimize the color distortions between frames. In addition to color distortions, the effect of specular highlights can also make the window of a vehicle appear white and degrade the accuracy of vehicle classification. To reduce this effect, a window-removal task is performed to make vehicle pixels with the same color more concentrated on the analyzed vehicle. Thus, a vehicle can be more accurately classified into its corresponding category even when it is shone by strong sunlight. One major problem in vehicle color classification is that there are many shade colors; for example, white versus silver and black versus navy. Traditional methods lack the ability to classify vehicles with shade colors because a wrong classifier is designed by putting vehicles with the same label together even though their chromatic attributes are different. To treat this problem, a novel tree-based classifier is designed for classifying vehicles into chromatic/nonchromatic classes with their nonchromatic strengths and then into detailed color classes with their color features. The separation can significantly improve the accuracy of vehicle color classification even that vehicles are with various shade colors and captured under different lighting conditions. Index Terms-Vehicle color classification, color correction, SVM, vehicle window removal.

This paper proposes a novel approach for estimating vehicles' orientations from still images using "eigen color" and edge map through a clustering framework. To extract the eigen color, a novel color transform model is used for roughly segmenting a vehicle from its background. The model is invariant to various situations like contrast changes, background, and lighting. It does not need to be re-estimated for any new vehicles. In this eigen color space, different vehicle regions can be easily identified. However, since the problem of object segmentation is still ill-posed, only with this model, the shape of a vehicle cannot be well extracted from its background and thus affects the accuracy of orientation estimation. In order to solve this problem, the distributions of vehicle edges and colors are then integrated together to form a powerful but high-dimensional feature space. Since the feature dimension is high, the normalized cut spectral clustering (Ncut) is then used for feature reduction and orientation clustering. The criterion in Ncut tries to minimize the ratio of the total dissimilarity between groups to the total similarity within the groups. Then, the vehicle orientation can be analyzed using the eigenvectors derived from the Ncut result. The proposed framework needs only one still image and is thus very different to traditional methods which need motion features to determine vehicle orientations. Experimental results reveal the superior performances in vehicle orientation analysis.