|
|
Application for Person Authentication Based on Retinal Images
Moncz, Oliver ; Goldmann, Tomáš (referee) ; Kavetskyi, Andrii (advisor)
This paper deals with the problem of authentication of people based on retinal images. The main objective is to design an algorithm that can match the retinal image with the registered images in the database. The outcome is the determination of identity with a degree of certainty. Another goal was to create an application that allows to perform the mentioned operations through a simple graphical user interface. Lastly, the results are evaluated and compared with existing solutions. The proposed authentication system achieved an average accuracy of 72.46 % for the DRIVE and STARE datasets, and 78.9 % for the FIRE dataset.
|
|
|
Using unlabeled data for retinal segmentation
Shemshur, Andrii ; Jakubíček, Roman (referee) ; Vičar, Tomáš (advisor)
Tato bakalářská práce se zabývá vývojem a hodnocením pokročilých metod pro segmentaci lékařských snímků v kontextu omezených trénovacích dat. Studie zkoumá techniky učení pod dohledem využívající konvoluční neuronové sítě (CNN), přenosové učení s předtrénovanými modely a strategie učení s částečným dohledem. Jako základní model byl použit model konvoluční neuronové sítě (CNN) s dohledem založený na architektuře U-Net, který dosáhl koeficientu Dice 77,6% a průniku nad sjednocením (IoU) 63,4%. Použití přenosového učení pomocí kodéru ResNet34 předtrénovaného na síti ImageNet vedlo k výraznému zlepšení výkonu s koeficientem Dice 81,9%, IoU 69,3% a přesností 96,7%. Kromě toho byly ke zvýšení výkonu modelu použity strategie učení s částečným dohledem, včetně pseudoznačení a předtrénování denoizace. Přístup pseudoznačení přinesl koeficient Dice 81,7% a IoU 69,1%, čímž prokázal účinnost využití neoznačených dat. Přístup před tréninkem denoizace prokázal robustní výkonnost a dosáhl koeficientu Dice 80,3% a IoU 67,0%, a to i v přítomnosti zašuměných a neoznačených dat. Tyto výsledky podtrhují potenciál transferového učení a poloprovozních metod pro zvýšení přesnosti segmentace při analýze lékařských snímků. Poskytují solidní základ pro budoucí výzkum v této oblasti.
|
|
|
Classification of the vascular tree in fundus images
Tebenkova, Iuliia ; Kolář, Radim (referee) ; Odstrčilík, Jan (advisor)
Retinal image analysis plays a very important role, as human gets around 90% of environment information with the help of eyes. Automation of process of retinal image analysis promotes to improve the efficiency of retinal medical examinations. The following thesis is dedicated to automatic classification methods of retinal vascular system images obtained from a digital fundus camera. Vessel classification method using classifier on the base of neural networks, which is trained and then tested on the retinal vessel segments, is investigated and implemented. In this thesis anatomical retinal survey, properties of image data from digital fundus camera and retinal image classification methods are briefly described. The last chapter is devoted to the evaluation of efficiency of retinal vessel classification with automatic methods.
|
|
|
Localisation of the optic disc from fundus photographs
Pěchotová, Barbora ; Kolář, Radim (referee) ; Odstrčilík, Jan (advisor)
This thesis deals with the analysis of retinal images from digital fundus camera, especially with structure of optic disc (OD). The theoretic part describes main features of the human visual system and princip of eyeground examination. The paper discussed sevaral available methods that have been used for localization of the optic disc. In the second part is proposed the automatic detector of the optic disc based on the principle of genetic algorithm by using the method of segmentation of vessels by matched filtering. Original images are tested with this automatic method.
|
|
|
Localisation of Optic Disc from Fundus Photographs
Pěchotová, Barbora ; Kolář, Radim (referee) ; Odstrčilík, Jan (advisor)
This thesis deals with the analysis of retinal images from digital fundus camera, especially with structure of optic disc (OD). The theoretic part describes main features of the human visual system and princip of eyeground examination. The paper discussed available methods that have been used for localization of optic disc. For further work is selected proposal of the geometrical model of vessel structure by using the method of segmentation of vessels by matched filtering. Original images are tested with semi-automatic method for optic disc localization.
|
|
|
Advanced retinal vessel segmentation methods in colour fundus images
Svoboda, Ondřej ; Jan, Jiří (referee) ; Odstrčilík, Jan (advisor)
Segmentation of vasculature tree is an important step of the process of image processing. There are many methods of automatic blood vessel segmentation. These methods are based on matched filters, pattern recognition or image classification. Use of automatic retinal image processing greatly simplifies and accelerates retinal images diagnosis. The aim of the automatic image segmentation algorithms is thresholding. This work primarily deals with retinal image thresholding. We discuss a few works using local and global image thresholding and supervised image classification to segmentation of blood tree from retinal images. Subsequently is to set of results from two different methods used image classification and discuss effectiveness of the vessel segmentation. Use image classification instead of global thresholding changed statistics of first method on healthy part of HRF. Sensitivity and accuracy decreased to 62,32 %, respectively 94,99 %. Specificity increased to 95,75 %. Second method achieved sensitivity 69.24 %, specificity 98.86% and 95.29 % accuracy. Combining the results of both methods achieved sensitivity up to72.48%, specificity to 98.59% and the accuracy to 95.75%. This confirmed the assumption that the classifier will achieve better results. At the same time, was shown that extend the feature vector combining the results from both methods have increased sensitivity, specificity and accuracy.
|
|
|
Analysis of fundus images aimed to localize pathological areas
Hartlová, Marie ; Kolář, Radim (referee) ; Odstrčilík, Jan (advisor)
Diabetic retinopathy is a serious eye complication of diabetes mellitus and one of the major causes of blindness in the world. This thesis deals with detection of neovascularizations, which is the first manifestation of diabetic retinopathy in the retina. In summary, in this thesis describe the properties image data from digital fundus camera, image segmentation methods, methods for automatic blood vessels segmentation and detection of neovaskularizations. This information are used to create own method to detect neovaskularization.
|
|
|
Processing of high-resolution retinal images
Vraňáková, Sofia ; Chmelík, Jiří (referee) ; Valterová, Eva (advisor)
Diplomová práca je zameraná na spracovávanie obrazov sietnice s vysokým rozlíšením. Cieľom práce je zlepšiť výslednú kvalitu výsledných snímkov sietnice získaných zo sekvencie snímkov nižšej kvality. Jednotlivé snímky sú najskôr spracované pomocou bilaterálnej filtrácie a zlepšenia kontrastu. v ďalšom kroku sú odstránené rozmazané snímky a snímky zobrazujúce iné časti sietnice. Posun medzi jednotlivými snímkami v sekvencii sa odhaduje pomocou fázovej korelácie, a tieto obrazy sú potom fúzované do výsledného snímku s vysokým rozlíšením pomocou priemerovania a využitia superrozlišovacej techniky, presnejšie regularizácie pomocou bilaterálneho celkového rozptylu. Výsledné mediánové hodnoty skóre kvality získaných obrazov sú PIQUE 0.2600, NIQE 0.0701, a BRISQUE 0.3936 pre techniku priemerovania, a PIQUE 0.1063, NIQE 0.0507, and BRISQUE 0.1570 pre superrozlišovaciu techniku.
|
|
|
Blood vessel segmentation in retinal images using deep learning approaches
Serečunová, Stanislava ; Vičar, Tomáš (referee) ; Kolář, Radim (advisor)
This diploma thesis deals with the application of deep neural networks with focus on image segmentation. The theoretical part contains a description of deep neural networks and a summary of widely used convolutional architectures for segmentation of objects from the image. Practical part of the work was devoted to testing of an existing network architectures. For this purpose, an open-source software library Tensorflow, implemented in Python programming language, was used. A frequent problem incorporating the use of convolutional neural networks is the requirement on large amount of input data. In order to overcome this obstacle a new data set, consisting of a combination of five freely available databases was created. The selected U-net network architecture was tested by first modification of the newly created data set. Based on the test results, the chosen network architecture has been modified. By these means a new network has been created achieving better performance in comparison to the original network. The modified architecture is then trained on a newly created data set, that contains images of different types taken with various fundus cameras. As a result, the trained network is more robust and allows segmentation of retina blood vessels from images with different parameters. The modified architecture was tested on the STARE, CHASE, and HRF databases. Results were compared with published segmentation methods from literature, which are based on convolutional neural networks, as well as classical segmentation methods. The created network shows a high success rate of retina blood vessels segmentation comparable to state-of-the-art methods.
|
|
|
Blood vessel segmentation in fundus images using classification methods
Šťastný, Pavel ; Štohanzlová, Petra (referee) ; Odstrčilík, Jan (advisor)
Segmentation blood vessel the local images of retinal is very important for evaluation and for diagnostics eye’s disease, especially diabetic retinopathy and glaucoma. This bachelor’s thesis is deal with segmentation blood vessel by classification methods. I used simple neural network as a classifier. First of all I taught her by delta rule and then I used matched filtering on the prepare image. At the end I compared all information with gold standard. Average va-lues from score for healthy images were sensitivity 0,7717, specificity 0,9571 and accuracy score 0,9225.
|
guest ::
