Wyniki wyszukiwania - BazTech

1

Diagnosis of Retinitis Pigmentosa from Retinal Images

Ravichandran Giritharan, Elangovan Poonguzhali, Nath Malaya Kumar

International Journal of Electronics and Telecommunications

|

2019

|

Vol. 65, No. 3

519--525

EN

Retinitis pigmentosa is a genetic disorder that results in nyctalopia and its progression leads to complete loss of vision. The analysis and the study of retinal images are necessary, so as to help ophthalmologist in early detection of the retinitis pigmentosa. In this paper fundus images and Optical Coherence Tomography images are comprehensively analyzed, so as to obtain the various morphological features that characterize the retinitis pigmentosa. Pigment deposits, important trait of RP is investigated. Degree of darkness and entropy are the features used for analysis of PD. The darkness and entropy of the PD is compared with the different regions of the fundus image which is used to detect the pigments in the retinal image. Also the performance of the proposed algorithm is evaluated by using various performance metrics. The performance metrics are calculated for all 120 images of RIPS dataset. The performance metrics such as sensitivity, sensibility, specificity, accuracy, F-score, equal error rate, conformity coefficient, Jaccard’s coefficient, dice coefficient, universal quality index were calculated as 0.72, 0.96, 0.97, 0.62, 0.12, 0.09, 0.59, 0.45 and 0.62, respectively.

2

Metrological Aspects of Abrasive Tool Active Surface Topography Evaluation

Lipiński D., Kacalak W.

Metrology and Measurement Systems

|

2016

|

Vol. 23, nr 4

567--577

EN

Analysis of the shape and location of abrasive grain tips as well as their changes during the grinding process, is the basis for forecasting the machining process results. This paper presents a methodology of using the watershed segmentation in identifying abrasive grains on the abrasive tool active surface. Some abrasive grain tips were selected to minimize the errors of detecting many tips on a single abrasive grain. The abrasive grains, singled out as a result of the watershed segmentation, were then analyzed to determine their geometric parameters. Moreover, the statistical parameters describing their locations on the abrasive tool active surface and the parameters characterizing intergranular spaces were determined.

3

The assessment of gas turbine vanes based on the watershed segmentation of their surface images

Błachnio J., Bogdan M.

Journal of KONES

|

2016

|

Vol. 23, No. 1

75--82

EN

The article discusses the linkage between the colour changes of the recorded surface images of vanes and their state. For this purpose, we used a selection of techniques and methods for processing and analysis of digital images (pre-processing, watershed segmentation,) for the colour representation other than RGB model [2, 4]. Difficult to interpret pixel information was replaced with an area information (extracted in the process of segmentation areas with certain characteristics and attributes – the relationship between the state of the vane and its colour). The proposed colour representation of the surface of the vanes (HSV model) allows separation of image data – the separation of luminance (brightness) from the colour (chrominance). This enables better interpretation of the obtained results of segmentation, and thus the quality of the resulting diagnostic information is increased. The obtained results allow the introduction of the tri-status classification of the vanes (operable, partially operable, and inoperable). The study adopted operated vanes (with different technical state) of gas-turbine stator vanes of an aircraft jet engine, made from ŻS – 6K alloy. The digital surface images (photos) of vanes were recorded after their dismantlement, by the use of a digital camera (fixed conditions of registration, selection of the appropriate light source, uniform illumination of the photographed surfaces, no reflections). This information is essential for the on-going, non-destructive assessment of the technical state of vanes.

4

Metoda znajdowania komórek i zliczania ognisk występowania histonu γ-H2AX w obrazach na potrzeby detekcji dwuniciowych pęknięć DNA

Wojciechowski P., Bal A.

Pomiary Automatyka Kontrola

|

2012

|

R. 58, nr 4

387-390

PL

Niniejsza praca opisuje algorytm automatyzujący proces zliczania ognisk histonu γ-H2AX w obrazach mikroskopowych. Ogniska te są miejscami, w których doszło do fosforylacji białka histonu H2AX w pozycji seryny 139 wskutek dwuniciowych pęknięć DNA. Propono-wana metoda działa dwuetapowo. Najpierw w obrazie mikroskopowym selekcjonowane są obiekty uznawane za komórki, a następnie na obszarze tych obiektów poszukiwane są miejsca wyznakowane barwnikiem fluorescencyjnym związanym z przeciwciałem anty-γ-H2AX - miejsca te wskazują na lokalizację ognisk γ-H2AX.

EN

This paper describes an algorithm for automating the process of counting foci of histone γ-H2AX in microscope images. Foci are the places where there occurred H2AX protein phosphorylation on Ser139 site in response to DNA double-strand breaks (DSBs). In a γ-H2AX genotoxic test the number of foci per a single cell is counted. The proposed method works in two stages. The first stage is segmentation of a microscopic image (Fig. 1a) for selecting cell regions. For this purpose fusion of binarization methods is used - for the each pixel the result is obtained by comparison of the results from the Otsu method, the triangle method and the method in which the threshold is equal to the image average brightness. The pixel is classified as an object's pixel when at least two methods give such results. For improving segmentation results a morphological filter is used. The results (Fig. 1b) usually comprise regions representing single cells (Fig. 1c) and a set of agglomerated cells (Fig. 1d). A modified watershed segmentation and region classify method is used for separation of agglomerated cells into regions representing only one cell. The final result (Fig.1e) contains only regions representing single cells. In the second stage for each single cell region the foci (which are marked with a fluorescent indicator joined with anti- γ-H2AX antibody) are found (Fig. 3) with use of watershed segmentation on the pseudogradient image (which is obtained by aggregation of the results of cell image binarization with different thresholds; Fig. 2). The final results in form of foci number as a function of time after irradiation (Fig. 4) are similar to the test results of non-lethal DNA damages founded in the literature (e.g. [5]) - this allows stating that the presented method gives good results.

5

Image Processing System for Localising Macromolecules in Cryo-Electron Tomography

Gedda M., Öfverstedt L. G., Skoglund U., Svensson S.

Machine Graphics and Vision

|

2010

|

Vol. 19, No. 2

159-184

EN

A major challenge in today's molecular biology research is to understand the interaction between proteins at the molecular level. Cryo-electron tomography (ET) has come to play an important role in facilitating objective qualitative experiments on protein structures and their interaction. Various protein conformation structures can be qualitatively analysed as complete galleries of proteins are captured by ET. To facilitate fast and objective macromolecular structure analysis procedures, image processing has become a crucial tool. This paper presents an image processing system for localising individual proteins from in vitro samples imaged by ET. We have evaluated the system using simulated data as well as experimental data.

6

Tworzenie obrazu gradientowego na potrzeby segmentacji wododziałowej barwnych obrazów cyfrowych

Bal A.

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2008

|

R. 81, nr 6

706-710

PL

Przedstawiono wpływ, jaki na wynik segmentacji wododziałowej obrazów barwnych ma zastosowanie różnych metod tworzenia obrazu gradientowego. Omówiono główne metody tworzenia obrazu gradientowego oraz ich wpływ na wyniki segmentacji wododziałowej. Przedstawiono także porównanie wyników segmentacji uzyskanych w wyniku stosowania różnych metod wyznaczania obrazów gradientowych.

EN

In this paper the influence of various methods for creating of gradient image on the results of watershed image segmentation method is presented. The main methods of gradient image creating are described and its influence on the results of color images watershed segmentation is presented. The comparison of the final segmentation results obtained by the usage of different gradient image creating method is also presented.

7

Metoda segmentacji obrazów jąder komórkowych roślin w cytometrii DNA z heterogenicznym rozkładem wewnątrz jądra

Gocławski J., Anioł P.

Automatyka / Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie

|

2006

|

T. 10, z. 3

299-311

PL

W artykule przedstawiono algorytm segmentacji obrazów jąder komórkowych roślin poprzedzający procedurę pomiaru zawartości DNA. Opiera się on o iteracyjne wieloprogowanie obrazu z odcieniami jasności, uzyskanego w wyniku wstępnego odejmowania składowych koloru w przestrzeni RGB. Odejmowanie to redukuje niewybarwione artefakty tła obrazu. Zapewniono także rozdzielanie sklejonych jąder komórkowych z użyciem metody działów wodnych w odniesieniu do transformaty odległości tych obiektów. Pokazano, że wieloprogowanie ma kluczowe znaczenie dla wyodrębnienia profilów jąder z wysoce nierównomiernym rozkładem DNA.

EN

The paper presents an algorithm of plant nuclei image segmentation preceding the procedure of DNA contents measurement. It is based on iterative multilevel thresholding of grey-level differential image, obtained by preliminary subtraction of components in RGB color space. The subtraction reduces non-stained background artifacts. Splitting of merged cell nuclei has been provided using shape-based watershed method referred to the distance transform of the objects. It has been shown, that multilevel thresholding plays a key role in the proper discrimination of nuclei profiles with highly nonuniform DNA distribution.