Edge map guided stereo matching in HSL color space for mobile robot navigation

Many robotic and computer-vision applications, such as robot navigation, demand accurate, reliable, and real-time stereo matching algorithm. However, stereo images captured from real sceneries usually have much noises, bad contrast, photometric distortions, and lack of textures, heavily affecting the stereo algorithms´ performance. This work proposes a new dissimilarity measure in HSL color space in order to substitute the established intensity-based ones. The proposed measure can be adopted by almost any of the existing stereo algorithms, enhancing it with its robust features. The performance of the dissimilarity measure is validated through experimentation with an edge map guided stereo matching algorithm, which is suitable for low texture stereo images. Experimental results for a variety of lighting conditions are gathered and compared to one of intensity-based algorithms. The proposed method outperforms the examined algorithm, exhibiting tolerance to illumination differentiations and lack of textures.