We are researching and developing image processing algorithms for dermatological applications in diagnostic and decision making systems as well as for education. This includes the creation of a personalized surface model from a set of color and depth camera images, detection and classification of dermatological findings, such as psoriasis lesions and plaques, as well as longitudinal analysis of changes.

Correspondance-less alignment of 3D non-rigid objects using parametric nonlinear deformation models. Our algorithms are able to estimate the parameters of the deformation without an established set of correspondences, thus providing a fully automatic and computationally efficient registration framework for 3D objects.

The aim of this research is to study image features and processing methods for detection of visual code regions (1D barcode, 2D datamatrix, OCR characters), with special focus on highly accurate and efficient algorithms that can be adapted for real industrial applications.

A multi-layer binary Markov random field model for extracting an unknown number of possibly touching or overlapping near-circular objects.

Methods for the automatic detection and evaluation of Glaucoma and other eye diseases from screening examinations.

The goal of our project is to develop new algorithms that exploit the imaging and connectivity capabilities of modern mobil devices (smartphones and tablets) and provide new visual information contents for the users.

We apply and evaluate machine learning methods to obtain information purely from the tomographic projections of an image. We design and implement discrete tomographic reconstruction methods that can exploit the knowledge obtained in this way.

The aim of this project is to improve the well-known methods of continuous and discrete tomography. We study image processing methods that can be applied directly on the projections to improve reconstruction quality. We improve techniques to measure the quality of the reconstructed images, and develop image reconstruction methods based on parallel processing using GPUs. We also investigate which are the most valuable projections, if the reconstruction is performed from just a few of them.

A novel method for non-rigid registration of transrectal ultrasound and magnetic resonance prostate images based on a non-linear regularized framework.

Thinning is a widely used pre-processing step in digital image processing and pattern recognition. It is an iterative layer by layer erosion until only the "skeletons" of the objects are left. We proposed some parallel thinning algorithms that are based on some sufficient conditions for topology preservation.

Thinning is a frequently used method for skeletonization by modeling the fire-front propagation. We proposed some sequential and parallel 2D thinning algorithms capable of producing topologically correct skeletons.

We consider the problem of planar shape registration on binary images. Our primary goal is to investigate novel methodologies which work without feature point extraction and established correspondences; avoid the solution of complex optimization problems; and provide an exact solution regardless of the strength of the distortion. The newly developed techniques are validated on medical images.

Computer aided surgical planning with semi-automatic segmentation, implant insertion and biomechanical analysis

The thinning is an iterative layer by layer erosion until only the "skeletons" of the objects are left. We proposed various 3D thinning algorithms capable of extracting medial lines or medial surfaces as well.

A novel solution for reassembling a broken object from its parts without established correspondences, where each part is subject to a linear deformation.

The goal of the project is to develop efficient image acquisition methods to gain visual information from incorrect, noisy, and uncertain projections.

We consider the problem of estimation the parameters of transformations aligning two binary images. The advantage of our algorithm is that it is easy to implement, less sensitive to the strength of the deformation, and robust against segmentation errors.

New approaches in Discrete Tomography are investigated. Studies are concentrating on absorbed projections, fan-beam geometry, new geometrical properties of discrete sets. Besides, new application fields (such as neutron radiography) are studied.

Skeletonization has been successfully applied in the following three medical applications: assessment of laryngotracheal stenosis, assessment of infrarenal aortic aneurysm, and unravelling the colon.

The `gas of circles' (GOC) model is a tool to describe a set of circles with an approximately fixed radius. The model is based on the higher-order active contour (HOAC) framework. The method has been succesfully applied to tree crown extraction on aerial images.

The human visual system is not treating different features sequentially. Instead, multiple cues are perceived simultaneously and then they are integrated by our visual system in order to explain the observations. Therefore different image features has to be handled in a parallel fashion. In this project, we attempt to develop such a model in a Markovian framework.

A method for computationally efficient skeletonization of three-dimensional tubular structures was proposeded. It is specifically targeting skeletonization of vascular and airway tree structures in medical images but it is general and applicable to many other skeletonization tasks.

Our automatic registration method was tailored to the needs of a model-based segmentation framework of the pelvic area in CT studies in collaboration with GE Medical Systems. Registration was used as a preprocesing task. The goal of it was the acceptable alignment of the pubic bone area (where the two organs of interest, the prostate and bladder are located) of studies from different patients.

Theoretical aspects of main problems in Discrete Tomography are studied such as existence, consistency, and reconstruction under several assumptions. Efficient reconstruction algorithms are also developed by incorporating prior knowledge into the reconstruction.

SZOTE-PACS is the Picture Archiving and Communication System of Albert Szent-Gyorgyi Medical University. It is able to collect studies from CT, MR, SPECT, US and modalities and convert them into DICOM format.

Study and development of image segmentation algorithms for different organs from CT images for radiotherapy planning purposes.

The goal of this project is to propose a method which is able to segment a color image without any human intervention. The only input is the observed image, all other parameters are estimated during the segmentation process. The algorithm finds the most likely number of classes, their associated model parameters and generates a segmentation of the image by classifying the pixels into these classes.

A joint project is in process between the University of Szeged Image Processing Group and Department of Optics and Quantum Electronics and GE Medical Systems (GEMS) Budapest, to study the problems connected with CT image corrections.

Stochastic paintbrush rendering is an algorithm to simulate the painting process to get a picture similar to a real painting. The image can be described by the parameters of the consecutive paintbrush strokes, resulting in a parameter-series. In this project, we will explore this stroke-representation of images for Content Based Image Retrieval (CBIR).

An automatic registration method utilizing mutual information and its various medical and neutron tomography applications.

Helicopter rotor blades were tested by combined neutron and X-ray radiography. Due to the large size of the rotor blades, several images had to be taken and put together to compose one big image. We developed an interactive solution based on artificial markers.

An agricultural statistical system has been elaborated, the first really significant GIS system of the Hungarian Central Statistical Office. The first version manages the data of vineyard and fruit cadastre.

We developed a unified framework for percutaneous therapies utilizing localization frames and implemented an application using CT imaging and a robot.

We developed an image processing method for MRI intensity standardization. We also introduced new, fast implementations of the fuzzy connectedness algorithm that allows segmentation at interactive speeds. We developed a new segmentation "workshop" for brain MRI segmentation using standardized MR images and the fast fuzzy connectedness algorithms.

We investigated registration methods based on interactively identified point pairs used in medical image registration. We proposed an affine search method and gave a sufficient existence condition for the unique solution. The properties of rigid-body and affine methods were examined via numerical simulations.

Scanned cadastral maps have been vectorized, and map objects have been recognized using graph algorithms and neural network.

The main contribution of this project is the implementation of a Markovian image segmentation model on a new chip, called CNN (Cellular Neural Networks), which is capable to do the task in real time.

This project is about image segmentation algorithms using a Markovian approach. The main contribution is a new hierarchical MRF model and a Multi-Temperature Annealing (MTA) algorithm proposed for the energy minimization of the model. The convergence of the MTA algorithm has been proved towards a global optimum in the most general case, where each clique may have its own local temperature schedule.

The well-known thin plate spline interpolation method has been adopted and improved to produce a digital elevation model (DEM) covering the whole territory of Hungary.

MicroSEGAMS is an AMIGA-based system to perform and evaluate isotope-diagnostic studies. It was created using the experiences with SUPER-SEGAMS.

Supplement to SEGAMS-80 with the creation and evaluation of SPECT studies. It became possible to create user programs for the system in FORTRAN.

The improvement of the SEGAMS system that made possible to automate the image acquisition and evaluation. The system could be generated in different languages as well.

SEGAMS: A tree-structured hierarchical data processing system. The software system of a gamma camera connected to a minicomputer that can be used to acquire and evaluate the most common types of nuclear medicine studies used in clinical routine.

Detecting changes in images acquired by a scintigraph about the distribution of radioactive isotopes is a difficult task. Image processing attempts were made to improve the detection of changes.