Showing papers on "Image segmentation published in 1976"

Saturation, Heu, And Normalized Color: Calculation, Digitization Effects, and Use.

Abstract: : Three transformations used in the analysis of tricolor natural scenes are analyzed. All have nonremovable singularities, near which they are highly unstable. Given digital input, the distribution of their transformed balues is highly nonuniform, characterized by spurious modes and gaps. These effects are quantified and illustrated. In addition, a significantly faster algorithm for hue is derived. Image segmentation techniques of edge detection, region growing cluster, and region splitting are affected arbitrarily badly by such problems. Some stratagems are illustrated that help minimize the bad behavior. Linear transformations are presented as a generally favorable alternative to these three nonlinear ones. (Author)

On the problem of embedding picture elements in regions

Abstract: A few new algorithms for region growing in pictures were developed. These algorithms are a step toward finding a satisfactory solution to the image segmentation problem, and in-depth understanding of the problems of nonsemantic image segmentation. The algorithms utilize a sequential decision approach for region boundaries detection. The sequential decisions are supported by a stochastic algorithm that maintains local statistics of the region near the boundaries as the region grows. A few illustrations of the algorithm's performance are included.