Chapter 16: Object and Scene Recognition
Image category recognition is important to access visual information on the level of objects (buildings, cars, etc.) and scene types (outdoor, vegetation, etc.). 1In general, systems for category recognition on images [8–11] and video [12, 13] use machine learning based on image descriptions to distinguish object and scene categories. In Chapter 13, methods have been discussed to detect salient points that are invariant to translation, rotation, and scale. Further, different color descriptors are presented in Chapter 14. Because there are many different descriptors, a structured overview is required of color invariant descriptors in the context of image category recognition.
Therefore, this chapter gives an overview of the invariance properties and the distinctiveness of the different color descriptors. The analytical invariance properties of color descriptors are explored using a taxonomy based on invariance properties with respect to photometric transformations, and evaluated experimentally using two benchmarks from the image domain  and the video domain . The benchmarks are very different in nature: the image benchmark consists of consumer photographs and the video benchmark consists of key frames from broadcast news videos.
This chapter is organized as follows. In Section 16.1, the diagonal model is revisited to provide a taxonomy of photometric invariance. Then, ...