Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Viola-Jones

Sortuj według:

Ogranicz wyniki do:

Rozpoznawanie obiektów w obrazie wideo z kamery do komputera

Cherednyk Oleksandr, Miłosz Elżbieta

Journal of Computer Sciences Institute

2018

Vol. 8

215--219

Celem artykułu jest określenie skuteczności wykrywania obiektu w obrazie wideo za pomocą kamery internetowej. W trakcie pracy zostały zbadane i opisane podstawowe metody rozpoznawania obiektów na zdjęciu, a mianowicie wykorzystanie sztucznych sieci neuronowych i metod Viola-Jones. Do przeprowadzenia eksperymentów, na podstawie metody Viola-Jones, realizowano aplikację do rozpoznawania obiektów na wideo. Za pomocą tej aplikacji, przeprowadzono badanie w celu określenia skuteczności metody Viola-Jones do wykrywania obiektów na wideo.

The goal is to determine the effectiveness of object detection in a video using the camera for the computer. In the course of work studied and described the main methods of recognition of objects in the image, namely the use of artificial neural networks and techniques of Viola-Jones. For the study, based on the method of Viola-Jones, implemented the application for object recognition in video, as this method is effective for solving this problem. With this application, a study was conducted to determine the effectiveness of the method of viola-Jones to detect objects in the video.

Research on human silhouette detection methods for a non-cooperative biometric system

Kacperski D., Włodarczyk M., Grabowski K.

International Journal of Microelectronics and Computer Science

2015

Vol. 6, nr 1

10--13

Together with the development of effective and efficient people identification algorithms, biometric authentication systems become increasingly popular and widespread, leading to a significant growth in the number of institutions interested in implementing and using such systems. Although, several research works focused their efforts on these type of solutions, none of the commonly available systems provide a non-cooperative approach to object identification. For this reason, they are not suitable for use in some specific situations, such as people entering the stadium. Therefore, we decided to go up against these limitations and develop biometric identification system for less constrained scenarios. In this paper, we present an evaluation of different algorithms suitable for human silhouette detection in such environment. We focus on investigating their effectiveness and performance under unconstrained conditions, such as different lighting.