Wyniki wyszukiwania - BazTech

1

Koncepcja analizy stanów emocjonalnych użytkowników w kontekście systemów zabezpieczeń transportowych

Bobkowska K, Janowski A., Przyborski M., Szulwic J.

Logistyka

|

2014

|

nr 6

2062-2067

PL

Autorzy, przywołując własne i światowe badania nad rozpoznawaniem emocji ludzkich z obrazu twarzy, wskazują na możliwość zastosowania algorytmów komputerowych i ich implementacji w komputerach osobistych (i innych urządzeniach personalnych wyposażonych w dostatecznie silny procesor obliczeniowy). Zastosowanie takiego rozwiązania może poprawić bezpieczeństwo użytkowania urządzeń, maszyn i pojazdów, których operatorzy muszą gwarantować niezawodność systemu. Wskazana koncepcja nadaje się do zastosowania w pojazdach, w których występuje bezpośrednia (kontaktowa) obsługa przez człowieka, ale także w rozwiązaniach wykorzystujących zdalne sterowanie (drony, nadzór ruchu). Badania realizowane przez autorów wskazują na możliwość oceny stanu emocjonalnego oraz wykrycia problemów psychologicznych na podstawie teledetekcyjnej oceny twarzy człowieka. Dotychczasowe rozwiązania opierają się na analizie obrazów z szybkich kamer, jednak postęp badań oraz rozwój technologiczny wskazuje, że rozwiązanie takie może być skutecznym elementem zabezpieczeń transportowych.

EN

The authors, citing their own and global researches over the recognition of human emotions from the face image indicate that there is a possibility to use computer algorithms and their implementation in personal computers (and other devices equipped with personnel sufficiently powerful processor). Such a solution can improve the safety of use of the equipment, machinery and vehicles for which the operators must guarantee the system reliability. The indicated concept is suitable for use in vehicles where there is a direct (contact) service by human, but also in solutions using a remote control (drones, traffic control) Research conducted by the authors indicate the possibility of evaluation of the emotional condition and possibility to detect the psychological problems on the basis of remote sensing assessment of the human face. Existing solutions are based on the analysis of images from high-speed cameras, but the progress of research and technological development indicate that such a solution can be an effective element of transport security.

2

Genetic algorithms for classifiers' training sets optimisation applied to human face recognition

Kawulok M.

Journal of Medical Informatics & Technologies

|

2007

|

Vol. 11

135--143

EN

Human face recognition is a multi-stage process within which many classification problems must be solved. This is performed by learning machines which elaborate classification rules based on a given training set. Therefore, one of the most important issues is selection of a training set which would properly represent the data that will be further classified. This paper presents an approach which utilizes genetic algorithms for selecting classifiers' training sets. This approach was implemented for the Support Vector Machines which is applied in two areas of automatic human face recognition: face verification and feature vectors comparison. Effectiveness of the presented concept was confirmed with appropriate experiments which results are described in this paper.

3

Integration of kernel and DCT coefficient fuzzy matching techniques for human face recognition

Bhattacharjee D., Basu D. K., Nasipuri M., Kundu M.

Foundations of Computing and Decision Sciences

|

2002

|

Vol. 27, No. 2

61-76

EN

In this paper we describe results of human face recognition using integration of two different fuzzy matching methods. The first method is based on the kernel matching concept. In this method the system is trained with different images of the same class. During training the presence of (55)-derived kernels are considered. Based on the presence or absence of kernels in different images of the same class one reference matrix is computed and using a fuzzy membership function on deviation of individual images from the values of reference matrix two tolerance matrices are created. In the second method using local shift invariant Discrete Cosine Transform (DCT) coefficients of images of a class, another set of reference matrix and two tolerance matrices are created in a same way as in the first method. For testing the image in a class, the values of the matching of input image with reference matrices for two different methods are calculated and classified based on their respective tolerance matrices. The final result is obtained by integrating the results from both the methods. The result of recognition is very high, where error rate is only 2.625% with false acceptance rate of 1.86%.