Computer Analysis of Images and Patterns
Journal of Mathematical Imaging and Vision
This special issue comprises thirteen recent and relevant works in different aspects of the computer analysis of images and patterns. We cover important topics like stereo image processing, image and shape representation, color texture classification, parallel systems and membrane computing, face recognition, pedestrian classification, estimation of photometry and analysis of medical images. The papers have been divided into four groups-three are based on similarities in the addressed problems
... nd the last focuses on six different applications. The first section presents stereo image processing from a variety of perspectives. "Tuning of Adaptive Weight Depth Map Generation Algorithms" by Acosta et al. shows a systematic statistical approach for the parameter setting problem in disparity map estimation within the context of depth map generation algorithms. The next paper, "P-System Implementation of Dynamic Programming Stereo" by Gimel'farb et al. discusses the capabilities of the main tools of Membrane Computing, the P-systems, and how to implement the symmetric dynamic programming stereo (SDPS) matching algorithm. The authors propose a novel method of stereo matching based on a combination of local edge detection, cost aggregation computation and belief propagation optimization techniques in "Stereo Matching Based on Dissimilar Intensity Support and Belief Propagation" by Da et al. The next section studies representation and estimation of shape from different mathematical points of view. P. Real ( ) Dpto.