Developing Image Processing Meta-Algorithms with Data Mining of Multiple Metrics
Figure 1
This parallel coordinates plot is a visual representation of our eleven metric values for 186 different variants of the image E4863S4I, produced by four image registration tools, AIR Warp (blue), AIR Linear (red), FSL FLIRT (green), and MINC Tracc (purple). Higher metric values are “better” (higher similarity or lower distance). Each trajectory across the plot gives the row of 11 metric values obtained by an image; altogether there are 186 such trajectories, so an entire table of metric values is plotted here. Higher scores are better, but the results of the metric/score computations in each column have been independently scaled, so the columns represent different real intervals; the spread of the vertical axis values is not as dramatic as it may appear. Notice that some disorder occurs for the edi (Entropy of Difference of Intensities) and woo (Woods) metrics, but the ordering of results is otherwise fairly consistent across metrics.
