The aim of this work is to build a computational model able to automatically identify, after training, dynamic speckle pattern regions with similar properties. The process is carried out using a set of descriptors applied to the intensity variations with time in every pixel of a speckle image sequence. An image obtained by projecting a self-organized map is converted into regions of similar activity that can be easily distinguished. We propose a general procedure that could be applied to numerous situations. As examples we show different situations: (a) an activity test in a simplified situation; (b) a non-biological example and (c) biological active specimens. The results obtained are encouraging; they significantly improve upon those obtained using a single descriptor and will eventually permit automatic quantitative assessment.

Solving the over segmentation problem in applications of Watershed Transform

The process of formation, growth and distribution of vessels within the developing central nervous system is difficult to analyze due to the complexity of the paths and branches within the system. The study of images of this area poses particular problems because the high levels of noise, blurring and poor contrast often prevent the objects of interest detected correctly. The design of algorithms for digital image processing suitable for this type of imagery remains a constant challenge. The aim of this work is to develop a computer tool to assist the specialist in processing these images. This paper proposes the use of morphological grayscale reconstruction and other morphological operators in order to segment the images properly. The results show that the algorithms allow a suitable segmentation of the objects of interest. Moreover, the interface developed for processing enables easy and simple analysis of them by the specialists.

Neurite growth in vitro, neuritogenesis, is a widespread methodology in the field of developmental neurobiology. Morphological analysis of growing neurites is usually a difficult task because of their thinness and their low contrast that makes it difficult to observe clearly their shape, number, length and spatial orientation. This paper presents the use of the granulometric size distribution function to obtain, automatically, information about the shape, size and spatial orientation of growing axons in tissue cultures. The results presented show the efficiency of the granulometric size distribution for this application. The automatic detection of growing axons and the precise characterization of a relevant parameter, indicative of the axonal growth spatial orientation, was obtained by this morphological tool. The developed algorithms facilitates the analysis of these images by automatically quantifying the angle of deviation of the direction of growth, which is important given the large number of images that need to be processed for this type of study.