The traffic sign identification and recognition system (TSIRS) is an essential component for autonomous vehicles to succeed. The TSIRS helps to collect and provide helpful information for autonomous driving systems. The information may include limits on speed, directions for driving, signs to stop or lower the speed, and many more essential things for safe driving. Recently, incidents have been reported regarding autonomous vehicle crashes due to traffic sign identification and recognition system failures. The TSIRS fails to recognize the traffic signs in challenging conditions such as skewed signboards, scratches on traffic symbols, discontinuous or damaged traffic symbols, etc. These challenging conditions are presented for various reasons, such as accidents, storms, artificial damage, etc. Such traffic signs contain an ample amount of noise, because of which traffic sign identification and recognition become a challenging task for automated TSIRS systems. The proposed method in this paper addresses these challenges. The sign edge is a helpful feature for the recognition of traffic signs. A novel traffic sign edge detection algorithm is introduced based on bilateral filtering with adaptive thresholding and varying aperture size that effectively detects the edges from such noisy images. The proposed edge detection algorithm and transfer learning is used to train the Convolutional Neural Network (CNN) models and recognize the traffic signs. The performance of the proposed method is evaluated and compared with existing edge detection methods. The results show that the proposed algorithm achieves optimal Mean Square Error (MSE) and Root Mean Square Error (RMSE) error rates and has a better Signal to Noise Ratio (SNR) and Peak Signal to Noise Ratio (PSNR) ratio than the traditional edge detection algorithms. Furthermore, the precision rate, recall rate, and F1 scores are evaluated for the CNN models. With the German Traffic Sign Benchmark database (GTSRB), the proposed algorithm and Inception V3 CNN model gives promising results when it receives the edge-detected images for training and testing.
In the execution of edge detection algorithms and clustering algorithms to segment image containing ore and soil, ore images with very similar textural features cannot be segmented effectively when the two algorithms are used alone. This paper proposes a novel image segmentation method based on the fusion of a confidence edge detection algorithm and a mean shift algorithm, which integrates image color, texture and spatial features. On the basis of the initial segmentation results obtained by the mean shift segmentation algorithm, the edge information of the image is extracted by using the edge detection algorithm based on the confidence degree, and the edge detection results are applied to the initial segmentation region results to optimize and merge the ore or pile belonging to the same region. The experimental results show that this method can successfully overcome the shortcomings of the respective algorithm and has a better segmentation results for the ore, which effectively solves the problem of over segmentation.
PL
W procesie algorytmu wykrywania krawędzi ufności i algorytmu grupowania do segmentacji obrazu zawierającego rudę i glebę, obraz rudy o bardzo podobnych cechach tekstury nie może być skutecznie segmentowany, gdy oba algorytmy są używane osobno. W pracy zaproponowano nowatorską metodę segmentacji obrazu opartą na połączeniu algorytmu wykrywania krawędzi ufności i algorytmu zmiany średniej, który integruje kolor, teksturę i cechy przestrzenne obrazu. Na podstawie wstępnych wyników segmentacji uzyskanych przez algorytm segmentacji zmiany średniej informacja o krawędziach oryginalnego obrazu jest wyodrębniana za pomocą algorytmu wykrywania krawędzi opartego na stopniu ufności, a otrzymane wyniki są stosowane do początkowych wyników segmentacji obszaru w celu optymalizacji i scalenia rudy lub gleby należących do tego samego obszaru. Wyniki eksperymentalne pokazują, że metoda ta może skutecznie przezwyciężyć wady odpowiedniego algorytmu i daje lepsze wyniki segmentacji dla rudy, co dobrze rozwiązuje problem nadmiernej segmentacji.
In the ceramic industry, quality control is performed using visual inspection in three different product stages: green, biscuit, and the final ceramic tile. To develop a real-time computer visual inspection system, the necessary step is successful tile segmentation from its background. In this paper, a new statistical multi-line signal change detection (MLSCD) segmentation method based on signal change detection (SCD) method is presented. Through experimental results on seven different ceramic tile image sets, MLSCD performance is analyzed and compared with the SCD method. Finally, recommended parameters are proposed for optimal performance of the MLSCD method.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The strong earthquake with magnitude 6.9 occurred ofshore at the northernmost edge of the Samos Island and was strongly felt in the north Aegean islands and İzmir metropolitan city. In this study, the effective elastic thicknesses of the lithosphere and seismogenic layer thickness were correlated with each other in order to understand the nature of the earthquakes. We determined that the upper and lower depth limits of seismogenic layer are in a range of 5–15 km, meaning that only the upper crust is mostly involved in earthquakes in the study area. The fact that seismogenic layer and effective elastic thicknesses are close to each other indicates that the earthquake potential may be within the seismogenic layer. Following that, we estimate the stress feld from the geoid undulations as a proxy of gravity potential energy in order to analyze the amplitude and orientation of the stress vectors and seismogenic behavior implications. The discrete wavelet transform has been carried out to decompose the isostatic residual gravity anomalies into horizontal, vertical and diagonal detail coefcients. The results delineated edges of gravity anomalies that reveal some previously unknown features.
Real-time traffic monitoring and parking are very important aspects for a better social and economic system. Python-based Intelligent Parking Management System (IPMS) module using a USB camera and a canny edge detection method was developed. The current situation of real-time parking slot was simultaneously checked, both online and via a mobile application, with a message of Parking “Available” or “Not available” for 10 parking slots. In addition, at the time entering in parking module, gate open and at the time of exit parking module, the gate closes automatically using servomotor and sensors. Results are displayed in figures with the proposed method flow chart.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The continuous shift of shoreline boundaries due to natural or anthropogenic events has created the necessity to monitor the shoreline boundaries regularly. This study investigates the perspective of implementing artifcial intelligence techniques to model and predict the realignment in shoreline along the eastern Indian coast of Orissa (now called Odisha). The modeling consists of analyzing the satellite images and corresponding reanalysis data of the coastline. The satellite images (Landsat imagery) of the Orissa coastline were analyzed using edge detection flters, mainly Sobel and Canny. Sobel and canny flters use edge detection techniques to extract essential information from satellite images. Edge detection reduces the volume of data and flters out worthless information while securing signifcant structural features of satellite images. The image diferencing technique is used to determine the shoreline shift from GIS images (Landsat imagery). The shoreline shift dataset obtained from the GIS image is used together with the metrological dataset extracted from Modern-Era Retrospective analysis for Research and Applications, Version 2, and tide and wave parameter obtained from the European Centre for Medium-Range Weather Forecast for the period 1985–2015, as input parameter in machine learning (ML) algorithms to predict the shoreline shift. Artifcial neural network (ANN), k-nearest neighbors (KNN), and support vector machine (SVM) algorithm are used as a ML model in the present study. The ML model contains weights that are multiplied with relevant inputs/features to obtain a better prediction. The analysis shows wind speed and wave height are the most prominent features in shoreline shift prediction. The model’s performance was compared, and the observed result suggests that the ANN model outperforms the KNN and SVM model with an accuracy of 86.2%.
W artykule zaprezentowano projekt architektury oraz sprzętową implementację toru przetwarzania obrazu dedykowanego do wykrywania przewodów napowietrznych w czasie rzeczywistym. Detekcję przewodów zaimplementowano w postaci potokowej procedury sprzętowej przy użyciu algorytmów wykrywania krawędzi, a następnie ich redukcji. Projekt przetestowano w środowisku FPGA Intel Cyclone V. Przeanalizowano opóźnienia i złożoność sprzętową zsyntezowanej struktury w FPGA. Oszacowano również maksymalną szybkość przetwarzania obrazu z użyciem zaproponowanej implementacji.
EN
The paper presents the architecture design and hardware implementation of a custom image processing module dedicated for detection of high voltage lines in real time. It has been implemented in the form of a pipelined hardware procedure using edge detection and reduction algorithm. The design was tested in the Intel Cyclone V FPGA environment. Time and hardware complexity of the synthesized structure in FPGA were analyzed. The maximum image processing speed was also estimated using the proposed implementation.
8
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Image stitching refers to the process of combining multiple images of the same scene to produce a single high-resolution image, known as panorama stitching. The aim of this paper is to produce a high-quality stitched panorama image with less computation time. This is achieved by proposing four combinations of algorithms. First combination includes FAST corner detector, Brute Force K-Nearest Neighbor (KNN) and Random Sample Consensus (RANSAC). Second combination includes FAST, Brute Force (KNN) and Progressive Sample Consensus (PROSAC). Third combination includes ORB, Brute Force (KNN) and RANSAC. Fourth combination contains ORB, Brute Force (KNN) and PROSAC. Next, each combination involves a calculation of Transformation Matrix. The results demonstrated that the fourth combination produced a panoramic image with the highest performance and better quality compared to other combinations. The processing time is reduced by 67% for the third combination and by 68% for the fourth combination compared to stat-of-the-art.
W artykule przedstawiono projekt architektury oraz implementację układową toru przetwarzania wstępnego obrazu z modułem detekcji krawędzi. Układ został zaimplementowany w FPGA Intel Cyclone. Zrealizowany moduł wykorzystuje pięć wybranych algorytmów wykrywania krawędzi, w tym Robertsa, Sobela i Prewitt.
EN
The paper presents FPGA implementation details of the hardware image processing block with the edge detection module. In the implemented video processing module we use five selected edge detection algorithms, including Roberts, Sobel and Prewitt. The structure was synthesized and packed using hardware design platform built around the Intel Cyclone V FPGA. The number of logic elements used in each implementation was compared. We also estimated the execution time and maximum possible frame rate in VGA (640x480) and FullHD (1920x1080) video stream.
W pracy przedstawiono opracowany algorytm rozpoznawania oraz lokalizacji przewodów linii wysokiego napięcia na podstawie obrazu horyzontalnego. Procedura detekcji przewodu została podzielona na trzy etapy. Pierwszy etap zawiera algorytm wykrywania krawędzi wykazujący największą czułość na krawędzie poziome, a jednocześnie brak czułości na krawędzie pionowe. Efektem jest znaczna redukcja liczby wykrytych krawędzi w porównaniu do algorytmów wykrywających krawędzie we wszystkich kierunkach ze stałą czułością. Drugi etap polega na filtracji wykrytych krawędzi oraz klasyfikacji punktów do dalszej analizy. Na podstawie lokalnych różnic między sąsiadującymi pikselami wstępnie odrzucane są miejsca zawierające przypuszczalnie szum lub tło. Finalna klasyfikacja punktów reprezentujących przewód dokonywana jest z wykorzystaniem algorytmu znajdującego najdłuższą ścieżkę. Ostatni, trzeci etap polega na lokalizacji przewodu poprzez aproksymację jego kształtu na podstawie sklasyfikowanych punktów. Kontur wykrytego przewodu uzupełniany jest zbiorem pikseli reprezentującym przerwane ścieżki, zwiększając w ten sposób dokładność finalnej aproksymacji. Procedurę wykrywania przewodu zrealizowaną przy użyciu poszczególnych algorytmów przetestowano dla zdjęć rzeczywistych przewodów elektrycznych i stwierdzono jej pełną skuteczność.
EN
The paper presents the algorithm for recognizing and locating high voltage power lines with the use of horizontal images. The algorithm has the highest sensitivity to horizontal edges and at the same time is insensitive to vertical edges. The first stage includes an edge detection algorithm. The second stage involves filtering of detected edges and classifying the points for further analysis. Using gradient method, the regions containing noise or background are discarded. The final classification of pixels representing the line is made using the longest path detection algorithm. In the third stage a shape of the wire is approximated based on the set of selected points. Then the contour of the detected wire is constructed with a set of pixels representing interrupted paths, thus increasing the accuracy of the final approximation.
Wykrywanie krawędzi jest pierwszym etapem w cyfrowym przetwarzaniu obrazów. Operacja ta polega na usunięciu informacji takich jak kolor czy też jasność, a pozostawieniu jedynie krawędzi. Efektem tej operacji jest znaczna redukcja ilości danych do dalszej analizy. Pozwala to na zastosowanie w następnych etapach przetwarzania bardziej złożonych algorytmów rozpoznawania obiektów na podstawie kształtu. W artykule zaprezentowano zastosowanie algorytmów Robertsa, Sobela, Previtt, Kirscha i Scharra. Zaproponowano też nowy, efektywny obliczeniowo algorytm dedykowany do wykrywania krawędzi poziomych. Algorytmy zostały porównane w zastosowaniu do detekcji przewodów. Kryteriami porównania były skuteczność wykrywania krawędzi przewodów wysokiego napięcia oraz szybkość działania. Algorytmy zostały zaimplementowane z wykorzystaniem biblioteki OpenCV oraz przetestowane na zestawie zdjęć przedstawiających przewody wysokiego napięcia.
EN
Edge detection is the first step in digital image processing. This operation involves removing information such as colour or brightness and leaving edges. This data size reduction makes that the data amount for the further analysis is significantly smaller. This allows to use more complex algorithms for recognizing objects based on shape in the next processing stages. The work presents the application of the known edge detection algorithms as these of Roberts, Sobel, Prewitt, Kirch and Scharr. Moreover, a new, computationally effective, edge detection algorithm dedicated for horizontal edges is proposed. The algorithms have been compared for detection of electric wires. The criterions of comparison were the effectiveness of edge detection applied to high voltage wires and the speed of operation. The algorithms have been implemented using the OpenCV library and tested on a set of images of high voltage wires.
12
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule omówiono problematykę wdrożenia systemu wizyjnego do sterowania procesem spawalniczym. Zaprezentowano podstawowe sposoby usuwania zakłóceń oraz algorytmy rozpoznawania obrazu. Przetestowano dostępne na rynku czujniki.
EN
The article discusses the issues of implementation of the vision system to control the welding process. The basic methods of removing noises and image recognition algorithms were presented. Commercially available sensors were tested.
In this paper we present a novel approach for image description. The method is based on two well-known algorithms: edge detection and blob extraction. In the edge detection step we use the Canny detector. Our method provides a mathematical description of each object in the input image. On the output of the presented algorithm we obtain a histogram, which can be used in various fields of computer vision. In this paper we applied it in the content-based image retrieval system. The simulations proved the effectiveness of our method.
This paper presents a multistep approach for segmentation of micro tomography images. Various images of porous structures were studied. Proper segmentation of that images is necessary to create 3D models of these structures. The introduced algorithm concerns finding the proper way of image filtering before the use of Canny-Deriche edge detection to obtain the best possible segmentation.
Artykuł prezentuje alternatywne podejście do programowania równoległego poprzez wykorzystanie programowalnych kart graficznych w celu wsparcia obliczeń, oraz połączenie tego podejścia z klasycznym zrównolegleniem opartym o wielordzeniowe procesory. Przeprowadzone testy przedstawiają zysk czasu jaki można uzyskać dzięki odpowiedniemu połączeniu OpenMP z technologią CUDA w obliczeniach związanych z wykrywaniem krawędzi na obrazie rastrowym przy użyciu algorytmu Cannego. Badania przeprowadzone zostały na sprzęcie różnej jakości. Napisane algorytmy są zgodne z CC 1,0 (zdolność obliczeniowa karty graficznej).
EN
This paper presents an alternative approach to parallel programming by using programmable graphics card to support calculations and combines this approach with a classical parallelization based on multi-core processors. The tests show the gain time that can be achieved through a combination of OpenMP with CUDA technology in the calculation of the edge detection on the raster image using the Canny’s algorithm. Tests were carried out on the equipment of varying quality. The algorithms are compatible with CC 1.0 (compute capability graphics card).
16
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Scale-dependency is one of the well-known features of ecological boundaries. Unfortunately, there are relatively few case studies analysing boundaries of different scales. Moreover, properties of moving split window (MSW) technique, a method potentially suitable for examining boundaries at several spatial scales, are not fully understood. In this study, we used artificial data sets to test the capacities and limitations of the MSW method. We also applied field data from the Mecsek Mts (Hungary) (611 m a.s.l.) in order to reveal possible boundaries at different scales and to contribute to the knowledge on vegetation pattern of mountain areas. We found that one should apply several window-widths when using MSW, since this is the only way to detect and differentiate between boundaries of different scales. Our study revealed the vegetation pattern of Mt Tubes: there is a series of continuously intergrading mesic communities on the northern slope, while the southern slope is occupied by a mosaic of different xeric communities. In this pattern, boundaries of two different scales have been identified. We conclude that MSW could effectively be used in similar mountain regions to analyse herb layer vegetation patterns and boundaries.
17
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule przedstawiono autorskie stanowisko rejestrujące położenie przewodów napowietrznej linii elektroenergetycznej. Dodatkowo zaprezentowano wyniki pomiarów zarejestrowanych w długim przedziale czasu wraz z możliwościami potencjalnego zastosowania ich dla wspomagania procesu eksploatacji sieci elektroenergetycznej .
EN
The article presents the position of a original instrument for recording the position of wires on overhead power lines. In addition, presents the results of the measurements recorded over a long period of time, along with the capabilities of their potential application to support electricity grid operation process (support network dispatcher, to prevent failures of power lines).
In this paper the problem of accurate edge detection in images of heat-emitting specimens of metals is discussed. The images are provided by the computerized system for high temperature measurements of surface properties of metals and alloys. Subpixel edge detection is applied in the system considered in order to improve the accuracy of surface tension determination. A reconstructive method for subpixel edge detection is introduced. The method uses a Gaussian function in order to reconstruct the gradient function in the neighborhood of a coarse edge and to determine its subpixel position. Results of applying the proposed method in the measurement system considered are presented and compared with those obtained using different methods for subpixel edge detection.
19
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Texture identification and matching a sample fabric within a known collection of produced fabrics is a time-consuming and difficult process as a human activity. In this study, a computational method for textile texture identification is introduced using an image analysis technique. For this purpose, images of fabrics were captured by a digital flat scanner. Texture features were extracted using the Edge frequency and Gray Level Co-occurrence Matrix (GLCM) methods. In this way, a library of texture features was collected. To match a new texture with library samples, the closest texture feature based on Euclidian distance was identified as the fabric texture. Experimental results for 33 different textures showed the successful identification of textures with both methods. However, the edge frequency method is more feasible and acceptable due to its computational simplicity and lower processing time. In addition, it was shown that the edge frequency method is extremely insensitive to the colour and scanning direction of the fabric.
PL
Identyfikacja struktury powierzchni tkanin jest zajęciem pracochłonnym. Opracowano metodę identyfikacji struktury wykorzystując metodę automatycznej analizy obrazu. W celu identyfikacji wykonuje się zdjęcia skaningowe powierzchni materiałów, które następnie ulegają obróbce z wykorzystaniem metody wykrywania krawędzi i metody macierzy zależności poziomów szarości. W ten sposób opracowane informacje o poszczególnych strukturach zostają wprowadzone do komputerowej bazy danych, gdzie są porównywane z odpowiednimi wzorcami. Wprowadzenie 33 obrazów różnych struktur wykazało prawidłowe działanie układu. Niemniej należy zaznaczyć, ze metoda z wykrywaniem krawędzi jest bardziej użyteczna ze względu na prostszy sposób przetwarzania i krótszy czas obróbki. W dodatku stwierdzono, ze metoda ta jest niewrażliwa na kolor materiału i kierunek detekcji.
Reconstruction of three dimensional models of objects from images has been a long lasting research topic in photogrammetry and computer vision. The demand for 3D models is continuously increasing in such fields as cultural heritage, computer graphics, robotics and many others. The number and types of features of a 3D model are highly dependent on the use of the models, and can be very variable in terms of accuracy and time for their creation. In last years, both computer vision and photogrammetric communities have approached the reconstruction problems by using different methods to solve the same tasks, such as camera calibration, orientation, object reconstruction and modelling. The terminology which is used for addressing the particular task in both disciplines is sometimes diverse. On the other hand, the integration of methods and algorithms coming from them can be used to improve both. The image based modelling of an object has been defined as a complete process that starts with image acquisition and ends with an interactive 3D virtual model. The photogrammetric approach to create 3D models involves the followings steps: image pre-processing, camera calibration, orientation of images network, image scanning for point detection, surface measurement and point triangulation, blunder detection and statistical filtering, mesh generation and texturing, visualization and analysis. Currently there is no single software package available that allows for each of those steps to be executed within the same environment. For high accuracy of 3D objects reconstruction operators are required as a preliminary step in the surface measurement process, to find the features that serve as suitable points when matching across multiple images. Operators are the algorithms which detect the features of interest in an image, such as corners, edges or regions. This paper reports on the first phase of research on the generation of high accuracy 3D model measurement and modelling, focusing upon the application of different operators for accurate feature point extraction. The implementation of those operators is discussed and performance of differen operators is analysed. The optimal operator for high accuracy close range object reconstruction is then highlighted. This research has facilitated a development of the feature extraction and image measurement process that will be central to the development of an automatic procedure for high accuracy point cloud generation in multi image networks where robust orientation and 3D point determination will facilitate surface measurement and modelling within a single software system.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.