PDF version of this document

The Approach

Image analysis is a general problem that can be tackled in various ways. This analysis is fundamental and essential to many processes such as industrial inspection, motion analysis [], face recognition and medical image understanding. What makes this problem intrinsically laborious is the inability to take into account single pixels independently to infer the structure they form together, cohesively. The goal of such interpretation or analysis is not only to tackle the problem correctly, but also to do so efficiently, in a way that will not be overly affected by the size of the image, i.e. not reliant on the scale of the problem.

Analysis often involves measurements of meaningful structures in an image and possibly some explanation regarding the form of these structures. In order to derive any adjuvant information about a particular meaningful structure, image segmentation must first take place. Segmentation is concerned with the identification of certain regions of interest which may be characterised as belonging to the same object. By dividing the image into such regions, understanding of the nature of its constituent components can almost instantly be gained.

This report concentrates on a top-down approach to data^2.2 analysis. The approach relies on a high-level abstraction of the visual attributes of one structure. Alternatively, and often more usefully, this abstraction can represent a collection of structures that together form another aggregate structure. The reason why such an approach is referred to as a top-down approach is that it contains some existing information that it attempts to fit to the problem posed^2.3. It makes assumptions about the problem and is in some sense taking a preliminary overview on the structures in an image as Figure illustrates.

Figure: A target image $\mathbf{T}$ is being overlaid by a high-level representation (the model $\mathbf{M}$) which seeks to find a good fit.

The rest of this chapter will describe popular methods of top-down image analysis, but will focus on active appearance models at the expense of other, less relevant methods.