|
|
Object Recognition by Neural Networks
Marák, Jaroslav ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
This thesis is focused on neural networks and their classification capability in object recognition tasks. For recognition is there used neural networks with feedforward architecture which is learned by Back Propagation algorithm. We discusses about problems which appear while a choosing topology of network or using various lerning-significant parametters while a learning process. Achieved results are presented in experiments with estimation.
|
|
|
Image processing within determination of topographic surface parameters
Boháč, Martin ; Ohlídal, Miloslav (referee) ; Šťastný, Jiří (advisor)
This work deal with determination topohraphic parameters of a randomly rough surface by the help of method of shearing interferometry. It is a optical method for determination surface roughness. The basic idea is based of on deformation interference strips which are made by interference of the same mutually translated monochrome luminous wavefronts. The wavefront is created after transit or reflection monochrome lights from the surface of a studied sample. The wavefronts creates picture with deformed interference strips , which carries information about character of the surface. This information can be profited by algorithms of image processing from the picture . The thesis was developed in research project MSM 0021630529 Intelligent Systems in Automation.
|
|
|
Similarity Measure of Points of Interest in Image
Křehlík, Jan ; Beran, Vítězslav (referee) ; Herout, Adam (advisor)
This document deals with experimental verifying to use machine learning algorithms AdaBoost or WaldBoost to make classifier, that is able to find point in the second picture that matches original point in the first picture. This work also depicts finding points of interest in image as a first step of finding correspondence. Next there are described some descriptors of points of interest. Corresponding points could be useful for 3D modeling of shooted scene.
|
|
|
Mobile System for Text Recognition on Android
Tomešek, Jan ; Kolář, Martin (referee) ; Zemčík, Pavel (advisor)
This thesis deals with creation of a mobile library for preprocessing of images with text which represents a part of a system for text recognition. The library is realized with emphasis on generality of use, efficiency and portability. The library providing a set of algorithms primarily for image quality assessment and text detection was created in this thesis. These algorithms enable a substantial decrease in volume of transmitted data and speed up and refinement of the recognition process. An example application for the Android platform able to analyze composition of foods stated on their wrappings was created as well. Overall, the library (system) simplifies development of mobile applications with focus on text extraction and analysis. The mobile application then provides a comfortable way of food harmfulness verification. The thesis offers a reader an overview of current solutions and tools available in this field, it provides a breakdown of important image preprocessing algorithms and guides him through the construction of the library and the application for mobile devices.
|
|
|
Evaluation of Object Detection in Image
Černošek, Bedřich ; Behúň, Kamil (referee) ; Zemčík, Pavel (advisor)
The main goal of this bachelor's thesis was to propose the evaluation method of object detection. Result of this work was to create a program which performs the evaluation of object detection on suitable data sample and intuitively displays result to user. The task was to propose suitable experiments and dataset for proving correctness of evaluation. Part of this work was to find optimal parameters for face detection and optimal photo preprocessing before the face detection.
|
|
| |
|
|
Biometric fingerprint identification
Ruttkay, Michal ; Smital, Lukáš (referee) ; Vítek, Martin (advisor)
This thesis describes the anatomical characteristics of fingerprints and their applications in identifying the person. The theoretical part describes the importance of papillary lines on fingerprints, statistical analysis and pre-processing of images in particular. The practical section provides the necessary operations to compare fingerprints. The implementation was done in Matlab.
|
|
|
Digital image analysis of mitotic chromosomes
Jaroš, Luboš ; Vítek, Martin (referee) ; Škutková, Helena (advisor)
The development in modern medicine has allowed to study human genome and detect predispositions to several diseases. One of very promising techniques is the analysis of human karyotype, i.e., the number and appearance of chromosomes in the cell nucleus. The most important step in the karyotype analysis is the chromosome detection and categorization. In this work, a new algorithm for detection of chromosomes from an image of microscopic DNA sample and their categorization into seven groups was developed. The algorithm was implemented in Matlab. The accuracy of segmentation and classification was tested on a set of images from two databases with 117 and 38 images, respectively. The sensitivity of the developed segmentation reached 88% while the value of positive predictivity of segmentation reached 92%. The success rate of chromosome pairing achieves 77%.
|
|
|
Microaneurysms and hemorrhages detection in retinal images
Tobiášová, Nela ; Štohanzlová, Petra (referee) ; Kolář, Radim (advisor)
Diabetic retinopathy is a serious eye complication of diabetes mellitus and one of the major causes of blindness in the world. The microaneurysms and the haemorrhages are the pathologies of diabetic retinopathy. Their detection can halt or reverse the progression of this disease and prevent blindness. The algorithms could be helpful to ophthalmologists. This bacherol’s thesis is concerned with the detection of microaneurysms and haemorrhages in fundus images. The diabetic retinopathy, the types of lesions and the treatment methods are described in the first part of the paper. Existing methods are described as follows. The practical part of this work is aimed at the proposal and the detection of the red lesions. It consists of several steps, such as selecting the correct channel of RGB images, using local methods of contrast enhancement, edge detection, thresholding, creating a training set of the feature vector and the classification with the use of the neutral network.
|
|
| |
guest ::
