Application and Evaluation of Colour Constancy in Visual Surveillance

Abstract

The problem of colour constancy in the context of visual surveillance applications is addressed in this paper. We seek to reduce the variability of the surface colours inherent in the video of most indoor and outdoor surveillance scenarios to improve the robustness and reliability of applications which depend on reliable colour descriptions e.g. content retrieval. Two well-known colour constancy algorithms -the Grey-World and Gamut-Mapping -are applied to frame sequences containing significant variations in the colour temperature of the illuminant. We also consider the problem of automatically selecting a reference image, representative of the scene under the canonical illuminant. A quantitative evaluation of the performance of the colour constancy algorithms is undertaken.

Edge-Based Color Constancy

“…Obtaining color constancy is of importance for many computer vision applications, such as image retrieval, image classification, color object recognition and object tracking [2], [3], [4].…”

A general method for human activity recognition in video

“…Xiang and Gong have addressed the important issue of how to effectively recognise action in a surveillance context when there is a sparsity of example data [27] and what rô le the high-level labelling of trajectories plays in this situation and in the general case [28]. Renno et al have expended considerable effort in creating solutions for a deployable, wide-area visual surveillance system and have addressed a variety of issues including themes such as colour constancy [19] and learning semantic models [17] in addition to the more common problems associated with surveillance, such as tracking. Notable work includes the investigation of how to track through blind regions, i.e., areas between camera views which cannot be seen in any view [2].…”

Vision-based production of personalized video

“…Some of the most popular ones are the Grey-world assumption, which assumes that the spatial average of surface reflectances in a scene is achromatic and the Gamut-mapping, which recovers the transform that best projects the measured gamut into that of a canonical (see e.g., [10]). In [11] it is shown that both algorithms improve the color constancy for the visual surveillance scenario. The Grey-world method can be implemented in real time while the Gamut-mapping not.…”