Maciej Hrebień

Segmentation of Breast Cancer Fine Needle Biopsy Cytological Images

Segmentation of Breast Cancer Fine Needle Biopsy Cytological Images This paper describes three cytological image segmentation methods. The analysis includes the watershed algorithm, active contouring and a cellular automata GrowCut method. One can also find here a description of image pre-processing, Hough transform based pre-segmentation and an automatic nuclei localization mechanism used in our approach. Preliminary experimental results collected on a benchmark database present the quality of the methods in the analyzed issue. The discussion of common errors and possible future problems summarizes the work and points out regions that need further research.

Computational Intelligence Techniques in Image Segmentation for Cytopathology

A variety of computational intelligence approaches to nuclei segmentation in the microscope images of fine needle biopsy material is presented in this chapter. The segmentation is one of the most important steps of the automatic medical diagnosis based on the analysis of the microscopic images, and is crucial to making a correct diagnostic decision. Due to complex nature of biological images, standard segmentation methods are not effective enough. In this chapter we present and discuss some modified versions of watershed algorithm, active contours, cellular automata, GrowCut technique, as well as new approaches like fuzzy sets of I and II type, and the sonar-like method.

Hough Transform, (1+1) Search Strategy and Watershed Algorithm in Segmentation of Cytological Images

This paper describes an early stage of cytological image recognition and presents a proposition of a hybrid segmentation method. The analysis includes the Hough transform in conjunction with the watershed algorithm. One can also find here a short description of image pre-processing and an automatic nuclei localization mechanism used in our approach. Preliminary experimental results collected on a handprepared benchmark database are also presented with discussion of common errors and possible future problems.

Human identification based on fingerprint local features

To meet todays demands for constructing even better biometric systems of human identification, a few fingerprint matching techniques are presented in this paper. One can also find here a short description of fingerprint image pre-processing and the minutiae extraction scheme used in our research. Because there is still a need to find the best matching algorithm, preliminary research was conducted to compare quality differences and answer times between the analysed methods for a prepared on-line system.

Hough Transform in Music Tunes Recognition Systems

This paper presents a method of music tunes recognition based on adopted Hough transform. One can also find here experimental results showing the effectiveness of the presented solution. Perspectives of further work and quality improvements are also stated as a base for subsequent research.

Hough and fuzzy Hough transform in music tunes recognition systems

This paper presents a method of music tunes recognition based on the Hough transform as well as its fuzzy version. One can also find here experimental results showing the effectiveness of the presented solutions. Perspectives of further work and quality improvements are also stated as a base for subsequent research.

Memory usage reduction in Hough transform based music tunes recognition systems

This work presents a proposition of the Hough transforms accumulator on-line compression method in the context of music tunes identification system. One can also find here a short discussion on the methods properties as well as experimental results obtained during the research.

Database Supported Fine Needle Biopsy Material Diagnosis Routine

This paper describes cytological image segmentation and diagnosis method. The analysis includes an expert database supported Hough transform for irregular structures, image pre-processing and pre-segmentation, nuclei features extraction and final diagnosis stage. One can also find here experimental results collected on a hand-prepared benchmark database that show the quality of the proposed method for typical and non-typical cases.

Automatic Fingerprint Identification Based on Minutiae Points

In recent years security systems have played an important role in our community. Payment operations without cash, restricted access to specific areas, secrecy of information stored in databases are only a small part of our daily living that requires special treatment. Beside traditional locks, keys or ID cards, there is an increased interest in biometric technologies, that is, human identification based on one’s individual features [2, 19].

Comparison of Minutiae Matching Techniques

This paper presents comparison of three minutiae matching techniques, i.e. Hough transform, global star method and orientation correlation. Short description of the pre-processing stage based on filtering, thinning and minutiae extraction is presented. The investigations are performed with a high quality fingerprint scanner.