Abstract: A device for establishing one or several recording regions from an image region is provided. The device includes a neighborhood determining unit for determining whether two objects of the image region are neighboring. Additionally, the device includes a graph set determining unit for determining a graph set such that the graph set includes one or several complete graphs, each element of the complete graph being neighboring to every other element of this complete graph. Furthermore, the device includes a selection set determining unit configured to add a complete graph to the selection set when one of the elements of an object set is included by precisely this one of the complete graphs of the graph set. In addition, the device includes a recording region determining unit for determining the one or several recording regions based on the complete graphs of the selection set.

Abstract: A device for determining objects in a color recording has an identifier, a feature extractor and a classifier. The identifier is implemented to identify the connected regions whose size or shape correspond to a predetermined condition from a plurality of connected regions existing in a binary image derived from a color recording based on a size or a shape of these connected regions. The feature extractor is implemented, for each of the identified connected regions, to extract a feature set from the color recording. The classifier is implemented to classify the identified connected regions into at least two disjunct groups based on the extracted feature sets for the identified connected regions.

Abstract: A method of segmenting biological cells in a picture so that the biological cells represent a foreground of the picture includes a step of applying a first fast marching algorithm to the picture or to a pre-processed version of same in order to obtain a first fast marching image. In addition, the method includes a step of segmenting the first fast marching image or a further-processed version of same into a plurality of homogeneous regions. Furthermore, the method includes a step of mapping each of the homogeneous regions to one node of a graph, respectively. In addition, the method includes a step of classifying each homogeneous region either as background or foreground on the basis of the graph. Moreover, the method includes a step of applying a second fast marching algorithm within the homogeneous regions classified as foreground so as to segment the foreground into individual biological cells.

Abstract: A device for determining objects in a color recording has an identifier, a feature extractor and a classifier. The identifier is implemented to identify the connected regions whose size or shape correspond to a predetermined condition from a plurality of connected regions existing in a binary image derived from a color recording based on a size or a shape of these connected regions. The feature extractor is implemented, for each of the identified connected regions, to extract a feature set from the color recording. The classifier is implemented to classify the identified connected regions into at least two disjunct groups based on the extracted feature sets for the identified connected regions.

Abstract: Automated localization of a valid area of a blood smear and, thus, localization requiring less effort and being more objective is enabled in that a picture of the blood smear pixels are classified at least into first pixels, which represent blood cells, and second pixels, which do not represent the blood cells, and the valid area is then found on the basis of a local frequency of pixel clusters of at least Amin first pixels, Amin being a minimum threshold for a number of first pixels of a pixel cluster, and a local average size of the pixel clusters for laterally distributed areas of the blood smear.

Abstract: Automated localization of a valid area of a blood smear and, thus, localization requiring less effort and being more objective is enabled in that a picture of the blood smear pixels are classified at least into first pixels, which represent blood cells, and second pixels, which do not represent the blood cells, and the valid area is then found on the basis of a local frequency of pixel clusters of at least Amin first pixels, Amin being a minimum threshold for a number of first pixels of a pixel cluster, and a local average size of the pixel clusters for laterally distributed areas of the blood smear.

Abstract: A method of segmenting biological cells in a picture so that the biological cells represent a foreground of the picture includes a step of applying a first fast marching algorithm to the picture or to a pre-processed version of same in order to obtain a first fast marching image. In addition, the method includes a step of segmenting the first fast marching image or a further-processed version of same into a plurality of homogeneous regions. Furthermore, the method includes a step of mapping each of the homogeneous regions to one node of a graph, respectively. In addition, the method includes a step of classifying each homogeneous region either as background or foreground on the basis of the graph. Moreover, the method includes a step of applying a second fast marching algorithm within the homogeneous regions classified as foreground so as to segment the foreground into individual biological cells.

Abstract: Apparatus for segmenting microcalcifications in a mammographic image having a bandpass filter for bandpass filtering the mammographic image obtaining the filtered mammographic image, a marker and a processor for individual processing. The marker marks image points in the filtered mammographic image as potential regions of microcalcifications when a they exceed or fall below a predetermined threshold. The processor for individual processing processes one of the regions of adjacent marked image points for changing an extension of the one region for illustrating a segmentation of microcalcifications.