Wyniki wyszukiwania - BazTech

1

An Experimental Bench for Testing a S-CAM Front Car Camera

Gryś Sławomir

International Journal of Electronics and Telecommunications

|

2022

|

Vol. 68. No. 4

755--760

EN

The paper presents an experimental stand for testing the front car camera S-CAM with embedded image recognition systems. The camera sends CAN messages these are converted to USART messages by microprocessor based system. The messages are interpreted by MATLAB script on the basis of database of traffic signs in accordance with Polish Road Code. The testing stand is mainly aimed for educating students interested in the fields of electronics and technologies related to automotive branch, as well. The second objective is a research on efficiency of traffic sign recognition system being one of functionalities of S-CAM camera. The technical specification of testing stand, its functionality and limitations were also discussed. The bench operation was illustrated with examples of stiff images, animation and real movies.

2

System inteligentny rozpoznawania znaków drogowych

Matuszewski J.

Przegląd Elektrotechniczny

|

2016

|

R. 92, nr 1

41--44

PL

Artykuł przedstawia autorską metodę dotyczącą komputerowego rozpoznawania znaków drogowych. Problem badawczy podejmowany w artykule określony został poprzez zdefiniowanie i opisanie sposobu tworzenia wzorców znaków drogowych, wyboru miary podobieństwa obrazów cyfrowych oraz reguł decyzyjnych w systemie rozpoznania. W celu detekcji znaku wykorzystano segmentację obrazu opartą na przestrzeni barw RGB, natomiast rozpoznawanie znaku oparto o detektor narożników. Na ich podstawie wyodrębniane są cechy deskryptora SURF, takie jak: liczba cech, położenie, skala, miara, znak Laplace’a i orientacja. Deskryptor SURF jest inwariantny względem obrotu i zmian skali, wykazuje wysoką powtarzalność detekcji punktów obrazu oraz odporność na jego zakłócenia. Dopasowanie znaku do wzorca opiera się na podstawie minimalnej odległości między wektorami, np. odległości Euklidesa lub Mahalanobisa. Badania poprawności algorytmu detekcji i rozpoznawania znaków drogowych dokonano przy pomocy aplikacji w środowisku MATLAB wraz z dodatkami: Computer Vision System Toolbox oraz Image Processing Toolbox. Otrzymane wyniki badań wskazują na wysoką skuteczność wykrycia i rozpoznania znaków drogowych. Aplikacja została przetestowana na zbiorze czterdziestu rzeczywistych zdjęć testowych znaków drogowych. Poprawnej detekcji nie dokonano tylko dla dwóch zdjęć testowych.

EN

The article describes the author’s method concerning the detection and recognition of traffic signs. The article investigates the definition and description of the traffic sign patterns, identifying the similarities of the digital pictures and decision criteria in the detection systems [4]. At the beginning for the traffic signs detection the picture segmentation was made which is based on the colour spectrum RGB (red, green, blue). Next in the process of sign recognition the detector of picture corners was used. On this base the following parameters of descriptor SURF were chosen for recognition: number of picture features, its situation, scale, the sign of Laplace`a function and picture orientation. The descriptor SURF is invariant with regard to rotation and scale changes, shows the high repetitious of the picture points detection and high resistance on jamming. The adjustment of recognized traffic sign to the appropriate pattern is based on the minimal distance between vectors, that is the Euclidean or Mahalanobis distance. An investigation into the correctness of the detection algorithm were made using an application in the MATLAB environment using Computer Vision System Toolbox and Image Processing Toolbox. Results achieved prove high accuracy in detection and recognition of traffic signs [1]. The application has been tested in 40 real-life traffic sign photos. Only two photos were not identified correctly.

3

Recognition of occluded traffic signs based on two-dimensional linear discriminant analysis

Forczmański P.

Archives of Transport System Telematics

|

2013

|

Vol. 6, iss. 3

10--13

EN

Traffic signs recognition involving digital image analysis is getting more and more popular. The main problem associated with visual recognition of traffic signs is associated with difficult conditions of image acquisition. In the paper we present a solution to the problem of signs occlusion. Presented method belongs to the group of appearance-based approaches, employing template matching working in the reduced feature space obtained by Linear Discriminant Analysis. The method deals with all types of signs, regarding their shape and color in contrast to commercial systems, installed in higher-class cars, that only detect the round speed limit signs and overtaking restrictions. Finally, we present some experiments performed on a benchmark databases with different kinds of occlusion.