|
|
Fingerprint Preprocessing Algorithm With Focuse On Future Liveness Detection
Kašpar, Jakub
The aim of this paper is to introduce the problem of fingerprint liveness detection and propose a method for preprocessing of fingerprints, which will be used for liveness detection. In the first part of this paper the basic problematic of false fingerprints is described. Further in the paper the principle of fingerprint preprocessing is introduced. The last part of this paper presents problematic fingerprint types and how well is the algorithm able deal to with it. The algorithm was tested on the LivDet database.
|
|
|
Detection of source code plagiarism
Bláhová, Barbora ; Harabiš, Vratislav (referee) ; Kašpar, Jakub (advisor)
The purpose of this thesis is to introduce the matters of plagiarism of source codes and to suggest methods, that will be used to create the original plagiarism detector. Theoretical part of the thesis states the definitions and the most common methods of plagiarism, both in general and in source codes. Furthermore it presents already existing types of the detectors and states principals, according to which the detection can be performed. Practical part deals with implementation of detecton and it´s testing.
|
|
|
Fingerprint identification algorithm
Skoupilová, Alena ; Vítek, Martin (referee) ; Kašpar, Jakub (advisor)
The main goal of this thesis is to introduce the meaning of identification and specific meaning of personal identification and mainly fingerprint identification. There is introduction to the meaning of biometrics as a key field for fingerprint recognition. Thesis also introduces principals and types of automatic processing and automatic detection of fingerprints. An existing and succesful methods are being described in the thesis and an automatic fingerprint identification alghoritm is being realized and described. Reability of the alghoritm is tested within experimental database.
|
|
|
Detection of pathologies in ECG signals
Němec, Radek ; Mézl, Martin (referee) ; Kašpar, Jakub (advisor)
This thesis attempts with methods for detection of pathologies in ECG signals. The first part is focused on the basic theory of heart anatomy, electrocardiography and individual heart pathologies. The second part describes the preprocessing of ECG signals, detection of QRS complexes and detection of individual pathologies. In the third part the detector is being tested on CSE database and my own measured data.
|
|
|
Detection of QRS complex in resting ECG and exercise ECG
Ondráček, Pavel ; Odstrčilík, Jan (referee) ; Kašpar, Jakub (advisor)
This thesis deals with methods for detection of QRS complex in ECG signal. There is described electrical activity of cardiac muscle and the ECG signal itself. There are also described some of the methods for detection of QRS complex and detailed procedure of realization of the two methods in matlab. Also, the efficiency of each method is discussed and compared at the end of the thesis.
|
|
|
Detection of similarity in program codes
Maťašová, Kristýna ; Vítek, Martin (referee) ; Kašpar, Jakub (advisor)
The Bachelor introduces the concept of plagiarism and possible kinds of plagiarism. It focuses on the problem of detecting the similarity of source codes, especially with graphical interfaces in the MATLAB environment. It also describes already existing detectors. The practical part of thesis is focused on finding appropriate flags for detection of similarity in source codes and introduces the metric of detected flags. It also describes the internal logic of created detector of similarity and discusses the results of its testing.
|
|
|
Detection of pathologies in ECG signals
Němec, Radek ; Mézl, Martin (referee) ; Kašpar, Jakub (advisor)
This thesis attempts with methods for detection of pathologies in ECG signals. The first part is focused on the basic theory of heart anatomy, electrocardiography and individual heart pathologies. The second part describes the preprocessing of ECG signals, detection of QRS complexes and detection of individual pathologies. In the third part the detector is being tested on CSE database and my own measured data.
|
|
|
Basic fingerprint liveness detection
Horák, Tomáš ; Smital, Lukáš (referee) ; Kašpar, Jakub (advisor)
This work focuses on recognizing fingerprints liveness based purely on software-methods evaluating symptoms from just one fingerprint image. At first in this work was described the issue of biometry as such, comparing the advantages and disadvantages of such systems. Next part deal with more detailed process of fingerprint biometry including papillary lines and overall fingerprints as such. In the next phase, the problems and utilization of both software and hardware methods are discussed, including principles of individual approaches. This part is followed by a selection of used fingertip symptoms. This is followed by the practical part and the LivDet 2011 database, which was used for finger recognition. In the practical part is also described the used neural network capturing minor differences in fingerprints according to 13 symptoms.
|
|
|
Plagiarism Detection in Program Codes Using Mapping Technique
Kašpar, Jakub
The aim of this paper is to introduce the problem of plagiarism and propose a method for plagiarism detection in program codes. In the first part of this paper the basic definition of plagiarism is described. Further in the paper the principle of preprocessing and localization process for signs of plagiarism is introduced. The last part of this paper presents an algorithm for comparison of the detected signs to get the best results possible. The detector was tested on student projects from the BTBIO study program.
|
|
|
Basic fingerprint liveness detection
Horák, Tomáš ; Smital, Lukáš (referee) ; Kašpar, Jakub (advisor)
Biometric data is unique, safe and often used to protect information. Even in some cases, detectors can be fooled, and therefore liveness is needed to check. This work focuses on the basic recognition of the liveliness of the fingerprint. Only one finger image is used. For perfect recognition, you need to compare the prints, so have more than one print, and that's why this detection is very demanding. Pre-processing was done with a binary segmentation mask. Methods using gray scale ratios, mean, standard deviations or histogram equalization were also used. All the methods were done on the LivDet 2011 database, in the Matlab environment and the goal was to recognize a live fingerprints from a false ones.
|
guest ::
