Integrating these approaches, the overall region query strategy consists of computing individual queries on color set, region location, area and spatial extent, as specified by the user. The process is summarized in Figure 9.
The single region distance is given by the weighted sum of the color set (Eq. 10), location (Eq. 11), area (Eq. 13) and spatial extent (Eq. 14) distances. The user may also assign a relative weighting to each of these attributes of each region. For example, the user may weight the size parameter more heavily than color and location in the query. The overall single region query distance is given by
Figure 9: Overall strategy for a single region query with parameters of color set, location, area and spatial extent.
We now outline the strategy for processing the joint color, absolute location and size queries for a single region. For example, given the single region query: find the region that best matches , the query is processed by first computing the individual queries for color, location, size and spatial extent. The intersection of the region match lists is then computed to obtain the set of common images; the best match minimizes the total distance. Here, , and are the search thresholds for location, area and spatial extent, respectively, which are set by the system or by the user.