Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 13

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
The paper explores possibility of improving Support Vector Machine-based classification performance by introducing an input data dimensionality reduction step. Feature extraction by means of two different kernel methods are considered: kernel Principal Component Analysis (kPCA) and Supervised kernel Principal Component Analysis. It is hypothesized that input domain transformation, aimed at emphasizing between-class differences, would facilitate classification problem. Experiments, performed on three different datasets show that one can benefit from the proposed approach, as it provides lower variability in classification performance at similar, high recognition rates.
2
Content available remote Maze Exploration Algorithm for Small Mobile Platforms
EN
An algorithm that enables efficient maze exploration is presented in the paper. The algorithm involves two phases: first the whole maze is explored in an ordered way and then, the shortest possible way out is determined. The algorithm has been derived in a way that combines main advantages of the two known labirynth-exploration algorithms: “Wall follower” and “Trémaux’s algorithm”. The algorithm has been tested using an autonomous vehicle, controlled by Arduino UNO, with two DC engines, ultrasonic sensors and gyroscope. It has been shown that the proposed approach provides a few crucial advantages with respect to already known solutions.
EN
The presented paper is concerned with feature space derivation through feature selection. The selection is performed on results of kernel Principal Component Analysis (kPCA) of input data samples. Several criteria that drive feature selection process are introduced and their performance is assessed and compared against the reference approach, which is a combination of kPCA and most expressive feature reordering based on the Fisher linear discriminant criterion. It has been shown that some of the proposed modifications result in generating feature spaces with noticeably better (at the level of approximately 4%) class discrimination properties.
4
EN
Robotic vehicles autonomy is the subject of many researches performed by various institutions. There is growing tendency to conduct research in the field of autonomous mobile units, unmanned vehicles and robots moving without operator supervision. Authors of this article presents the concept of the modularized subsystem implementing the functionality of the mobile platform navigation in an unknown environment. The concept has been implemented in the real mobile robot.
5
Content available remote The Concept of Subsystem for Mobile Robot Management From the Operator Console
EN
The authors in this publication present the concept of control system for the console operator to manage mobile robot platform and its components. Implemented functionality provides the ability to set the robot in different modes. The operator can control the state of robot using simple commands, and is also able to carry more complex tasks. The applied layered software architecture and Object-oriented programming (OOP) allows for easy management of software, maintenance, reusability and its further development.
6
Content available remote On a Real-Time First-Order Difference Evaluation
EN
A subject of this paper is a real-time evaluation of the first-order backward difference of a measured signal with noise. It is wellknown that a difference evaluation strongly increase a noise amplitude so the signal prefiltering is necessary. A very short review of practically used filters is given. To perform the pre-filtering operation one mainly apply linear filters characterized by their impulse response shape: finite impulse response (FIR) and infinite impulse response (IIR). To enhance the filtering operation one successfully apply so called adaptive filters and these described by non-linear characteristics. In this paper a simple non-linear filtering algorithm is proposed and examined in the first-order backward difference of a measured signal.
7
Content available remote Enhancement of Temperature Feature as a Result of Double Source Image Fusion
EN
In this paper a double source images fusion algorithm is presented. Its task is to enhance temperature feature of objects located on the scene. Presented solution is design to be executed in real-time environment. It consists of three stages: in the first part the differences between acquired double source images are examined in order to determine their intersection. Then for each analysed image the contours of all objects located on the scene are determined. This operation is essential to solve disparition issue. In the last stage, based on determined contours and their match coefficient the images are fused. The enhanced temperature feature is displayed on one image acquired from day-light camera.
8
Content available remote Distributed Mobile Platform Management System
EN
This papers presents a concept of distributed computer control system of mobile platform. This system was designed and implemented at the Institute of Applied Computer Science (formerly the Computer Engineering Department) as a grant: "Autonomous robot for surveillance and mine detection tasks". The main task was to encapsulate the functionality of the system as independent modules. They cover: the physical layer control, acquisition and fusion of data from different sensors, communication between the modules, analysis of image data. Mentioned mechanisms are also designed to perform autonomous behaviors.
EN
This paper presents a robust algorithm for simple shapes detection and tracking on image sequence. Presented solution consists of two parts. In the first part the image is analyzed and an area where object can be detected is found using correlation method. In the second part of presented algorithm a pattern which was previously found is confirmed by the original one. In order to obtain better accuracy the size and the angle of found pattern are calculated. Experimental results show that the accuracy of the pattern detection in presented algorithm is over 90% and is independent from lighting conditions and deformation of searched pattern caused by its spatial rotation.
PL
W niniejszym artykule zaprezentowano algorytm do wykrywania i śledzenia prostych wzorców w sekwencji obrazów. Zaprezentowane rozwiązanie składa się z dwóch etapów. W pierwszej części analizowany jest obraz oraz określany jest na podstawie współczynnika korelacji obszar, w którym może znajdować się poszukiwany wzorzec. W następnym etapie prezentowanego algorytmu wzorzec, który został znaleziony, porównywany jest ze wzorcem z poprzedniego obrazu. W celu uzyskania dokładniejszych wyników obliczany jest rozmiar oraz kąt nachylenia znalezionego wzorca. Przeprowadzone badania wykazały dokładność detekcji wzorca na poziomie 90% oraz to, że poziom ten jest niezależny od warunków oświetleniowych oraz deformacji spowodowanej obrotem wzorca w przestrzeni 3D.
EN
The aim of this paper is to investigate two equivalent forms of the fractional-order backward difference (FOBD) simplified formulae evaluation. The first form is known as the Grunvald-Letnikov form the second one as the Horner form. The simplifications are forced by microprocessor systems requirements. Two simplified forms are analyzed. The investigations are illustrated by numerical examples of simplified FOBD evaluation results.
PL
Celem pracy jest analiza dokładności obliczeń uproszczonych postaci Grunvalda-Letnikowa różnicy wstecznej niecałkowitych rzędów. Podane zostały równoważne definicje obu postaci oraz ich uproszczone formy wynikające z ograniczeń narzuconych przez systemy mikroprocesorowe. Omówiono podstawowe właściwości rozważanych postaci różnicy wstecznej niecałkowitego rzędu. Rozważania ilustrują przykłady obliczeniowe pokazujące dokładność obliczeń obu uproszczonych postaci różnic.
PL
W niniejszej publikacji przedstawiona została koncepcja algorytmu umożliwiającego wykrycie poruszających się obiektów na podstawie danych pozyskiwanych z czujników podczerwonych. Algorytm przez swój charakter stanowi uzupełnienie informacji wymaganych podczas budowy mapy otoczenia robota mobilnego. Autorzy przedstawiają analizę dokładności przedstawionego algorytmu. Przeprowadzone eksperymenty pokazują, że wyniki działania przedstawionego algorytmu stanowią wiarygodną informację o prędkości robota.
EN
Concept of moving detection algorithm base on data from infrared sensors is represented in this paper. The proposed algorithm can complement information required for building environment map for mobile robot. Authors discus accuracy of proposed algorithm. Results of experiments proof that proposed algorithm gives reliable information about robots current speed.
PL
W niniejszej publikacji zostały przedstawione algorytm DCT i algorytm Tenenbauma w zastosowaniu do ogniskowania obrazu w układzie ruchomej kamery umieszczonej na przegubie robota. Możliwość kontrolowania położenia pozycji soczewki jest niezbędna w stereowizyjnych procedurach przetwarzania obrazu. Autorzy przedstawiają analizę czasu działania przedstawionych algorytmów.
EN
Algorithms DCT and Tenenbaum are presented in this paper. Those algorithms were used to find focus in image from cameras mounted on the mobile robot head. Lens position is very important to find depth map using stereovision cameras. Authors discus time of focus calculation by those algorithms.
PL
W niniejszym artykule przedstawiono wpływ reprezentacji kolorów na wyniki segmentacji metodami rozrostu obszarów oraz metodą quadtree. Pod uwagę wzięto popularnie stosowane przestrzenie kolorów, takie jak RGB, HSV, CIEL*a*b. Przedstawiono matematyczną miarę jakości segmentacji oraz skonfrontowano ją z subiektywnymi odczuciami człowieka. Przy opracowaniu wyników użyto zestawów obrazów z kolekcji COREL oraz zdjęć lotniczych użytych w serwisach Google.
EN
This paper describes influence of color space to the results of image segmentation by watershed and quadtree algorithms. The most popular color space like RGB, HSV and CIEL*a*b was analyzed. The mathematical measure of segmentation quality was presented and contrasted with the subjective human feelings. Sets of images from COREL collection and aerial images from Google were used in experiments.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.