A common set of perceptual observables for grouping, figure-ground discrimination, and texture classification

We present a complete set of perceptual observables that provides a unified image description for grouping, figure-ground separation, and texture analysis. Although much progess has been made recently in treating contours and texture simultaneously for image segmentation and grouping, current approaches rely on different models for contours, regions, and texture such as one-dimensional intensity discontinuities for contours and filter bank responses for texture. This results in expensive computation that arbitrates between these disparate representations at each pixel. In our approach, salient image content such as contours, regions, and texture are represented in a common, low-level framework of image observables. We model the image as a partition of surfaces bounded by intensity discontinuities and derive perceptual measures as relations between neighboring surfaces. This enables us to extend the traditional Gestalt measures based on local edge geometry and contrast to region-based measures that jointly exploit large-scale image topology, photometry, and geometry. These measures provide a natural basis for grouping on multidimensional similarity criteria and texture is directly derived as relational properties on local region neighborhoods. The viability of our model is demonstrated by applying the common observables to texture recognition, figure-ground separation, and generic image segmentation. The texture classification algorithm approaches or exceeds the accuracy of filter bank approaches on both periodic and nonperiodic textures that have significant 3D structure. The measures are invariant to image rotation and slowly varying against large changes in illumination, viewpoint, and scale. The same perceptual measures are successfully applied in a difficult figure-ground separation problem in aerial images. Regions are first filtered, then grouped, using an efficient search algorithm based on perceptual salience to delineate objects of interest. Results for both are shown on large sets of complex, real-world images exhibiting difficult conditions.