Mariusz Kubanek

Method of Speech Recognition and Speaker Identification using Audio-Visual of Polish Speech and Hidden Markov Models

Mainstream automatic speech recognition has focused almost exclusively on the acoustic signal. The performance of these systems degrades considerably in the real word in the presence of noise. It was needed novel approaches that use other orthogonal sources of information to the acoustic input that not only considerably improve the performance in severely degraded conditions, but also are independent to the type of noise and reverberation. Visual speech is one such source not perturbed by the acoustic environment and noise. In this paper, it was presented own approach to lip-tracking and fusion of signals audio and video for audio-visual speech and speaker recognition system. It was presented video analysis of visual speech for extraction visual features from a talking person in color video sequences. It was developed a method for automatically localization of face, eyes, region of mouth, corners and contour of mouth. It was proposed synchronous and two asynchronous of methods of fusion of signals audio and video. Finally, the paper will show results of lip-tracking depending on various factors (lighting, beard), results of speech and speaker recognition in noisy environments.

Users Verification Based on Palm-Prints and Hand Geometry with Hidden Markov Models

This paper presents a biometric system implementing identity verification on the basis of the geometry of a hand and palm-prints. Hidden Markov Models were used here as teaching and verifying tools. The paper includes suggestions of our own algorithms as well as hand features extraction methods and their specific coding for Hidden Markov Models. The obtained results show that human hand carries a lot of useful information and appropriately modified can become an interesting proposal among biometric systems, particularly carrying out the verification process with a limited number of users.

Czestochowa-Faces and Biometrics of Asymmetrical Face

Traditionally, in face authentication/identification methods the presumption concerning face/head symmetry is used. For novel applications that concern creating techniques by means of which it is possible to reproduce the extraordinary complexity of skin, muscle, eye and hair movements, which convey emotion, gesture, psychological state or psycho-sociological traits, we begin to create new research direction called Biometrics of Asymmetrical Face.

A Hybrid Method of User Identification with Use Independent Speech and Facial Asymmetry

Speaker identification is the process of identifying an unknown speaker from a set of known speakers. In a speaker identification or verification, the prime interest is not in recognizing the words but determining who is speaking the words. In systems of speaker identification, a test of signal from an unknown speaker is compared to all known speaker signals in the set. The signal that has the maximum probability is identified as the unknown speaker. In security systems based on speaker identification, faultless identification has huge meaning for safety. In aim of increasing safety, in this work it was proposed own approach to user identification, based on independent speech and facial asymmetry. Extraction of the audio features of persons speech is done using mechanism of cepstral speech analysis. The part of the work that deals with face recognition was based on the technique of automatic authentication of a person with assumption that the use of automatically extracted, structural characteristics of the face asymmetry (in particular within the eyes and mouth regions as the most informative parts of the face) leads to improvement of the biometrical authentication systems. Finally, the paper will show results of user identification.

Feature Extraction of Palm Vein Patterns Based on Two-Dimensional Density Function

The pattern of blood vessels of a hand to build a biometric system will be presented in the article. Being a unique feature this pattern is impossible to be forged. Acquisition of a given biometrics will be described. The method to improve contrast of the input image based on three stages: the image histogram equalization, a smoothing operation of the filter and image normalization will be presented. The feature extraction method based on the two-dimensional density function will be shown. The location of blood vessels method based on the nearest neighbour matching method will be discussed. This paper contains a comparison of available in the scientific literature common methods used in the process of creating the biometric system based on the distribution of blood vessels in the hand.

Using Facial Asymmetry Properties and Hidden Markov Models for Biometric Authentication in Security Systems

This work concerns the use of biometric features, resulting from the look of a face, for the authentication purposes. For this we propose several different methods of selection and feature analysis during face recognition. The description contains mainly the possibility of the analysis and in later stages also identity verification based on asymmetric facial features. The new authentication method has been introduced on the basis of designated characteristic points of face. The method includes propositions of our own algorithms of face detection, as well as face features extraction methods and their specific coding in the form of observation vectors and recognition using Hidden Markov Models.

System of Acoustic Assistance in Spatial Orientation for the Blind

The technological development provides better and more mobile devices to aid disabled people. All work conducted to improve the life of disabled people constitute an important part of science. The paper describes the issues related to anatomy and physiology of sight and hearing organs, imagination abilities of the blind and assisting devices for those people. The authors of this work have developed a prototype of an electronic device which navigates a blind person by means of sound signals. Sounds are meant to provide the blind with a simplified map of the object depth in their path. What makes the work innovative is the Kinect sensor applied to scan the space in front of the user as well as the set of algorithms designed to learn and generate acoustic space, which also take into account the tilt of the head. The carried out experiments indicate the correct interpretation of the sound signals being modelled. The tests conducted on the people prove the developed concept to be highly efficient.

Verification of Identity Based on Palm Vein and Palm-Print

This paper presents a biometric system implementing identity verification based on palm vein and palm-print. The paper includes suggestions of our own algorithms as well as palm feature extraction methods and their specific coding for verification. The obtained results show that the human palm carries a lot of useful information and, appropriately modified, can become an interesting proposal among biometric systems, also carrying out the verification process with a large number of users.

Single camera based location estimation with dissimilarity measurement

In this paper we present a method that uses a single camera as a passive sensor for location estimation. The implementation is focused on the image processing techniques and dissimilarity measurement. Experimental results have indicated that good state estimates could be obtained, considering very little use of visual information. Consequently, popular cameras can be perceived as a simple and interesting solution for visual odometry and present very promising possibility in autonomous navigation.

GPU accelerated image processing for lip segmentation

This paper presents the problem of lip segmentation in parallel environment using computational capabilities of GPUs and CUDA. The presented implementation of lip segmentation is based on image processing methods using the most popular transformations such as morphological operations and convolution filters. The obtained experimental results for the parallel implementation on GPU indicate significant speedup in comparison to its sequential counterpart. Consequently, the use of popular graphics cards provides a very promising possibility of quick lips segmentation.