My master's thesis deals with the different techniques used in fingerprints processing for identifying fingerprints. Using the software tool Visual C++ and functions of OpenCV library I programmed a separate application, that is able to select from a database of fingerprints the most consistent with a comparative fingerprint images, even when they are mutually shifted in the direction of axes X and Y. The next step in my program is to gather the edges of the fingerprint image. Those obtained using Canny edge detector. Furthermore, getting the contours of the image edges. To determine, whether the contours are the same, just compare some characteristic points of contours. Next I use a histogram function to determine the number of points for approximation of contours and evaluating compliance fingerprints. Since the processing of the input fingerprint image (or rather the approximation of the contour points) remains in the picture as black (background) and red (the approximation of the contour points), this means, that zero and the last element of the histogram represent the number of black and red points. Comparison is in percentage and is obtained by subtracting the approximated points of contours image from the original fingerprint image of approximated contour points of matched fingerprints. It determined, what percentage of red points have disappeared, so as to match two fingerprint images. If on the resulting figure is not left neither a red point, that corresponds to 100% of the fingerprints Compliance.

No / Robust techniques are developed to put the H and Ca K line full-disk images taken at the Meudon Observatory into a standardised form of a `virtual solar image'. The techniques include limb fitting, removal of geometrical distortion, centre position and size standardisation and intensity normalisation. The limb fitting starts with an initial estimate of the solar centre using raw 12-bit image data and then applies a Canny edge-detection routine. Candidate edge points for the limb are selected using a histogram based method and the chosen points fitted to a quadratic function by minimising the algebraic distance using SVD. The five parameters of the ellipse fitting the limb are extracted from the quadratic function. These parameters are used to define an affine transformation that transforms the image shape into a circle. Transformed images are generated using the nearest neighbour, bilinear or bicubic interpolation. Intensity renormalisation is also required because of a limb darkening and other non-radial intensity variations. It is achieved by fitting a background function in polar coordinates to a set of sample points having the median intensities and by standardising the average brightness. Representative examples of intermediate and final processed results are presented in addition to the algorithms developed. The research was done for the European Grid of Solar Observations (EGSO) project.

Ces travaux de recherche visaient à développer un système d'aide à la mobilité pour les personnes ayant une déficience visuelle (VIP ‘Visually Impaired Person’) appelé ‘Smart Environment Explorer Stick (SEES)’. Le but particulier de cette recherche était de développer de nouveaux algorithmes pour améliorer la précision de la détection de la présence de poteaux de la canne SEE-stick en utilisant la méthode de calcul de distance et la recherche de paires de lignes verticales basées sur l'optimisation de la technique de détection de contour de Canny. Désormais, l'algorithme de détection des poteaux est appelé l’algorithme YuRHoS. Le SEES développé comme système de support d'aide à la mobilité VIP a été intégré avec succès à plusieurs dispositifs tels que le serveur distant dénommé iSEE, le serveur local embarqué dénommé SEE-phone et la canne intelligente dénommée SEE-stick. Les performances de SEE-stick ont été améliorées grâce à l'algorithme YuRHoS qui permet de discriminer avec précision les objets (obstacles) en forme de poteau parmi les objets détectés. La comparaison des résultats de détection des poteaux avec ceux des autres algorithmes a conclu que l'algorithme YuRHoS était plus efficace et précis. Le lieu et la couleur des poteaux de test d’évaluation étaient deux des facteurs les plus importants qui influaient sur la capacité du SEE-stick à détecter leur présence. Le niveau de précision de SEE-stick est optimal lorsque le test d’évaluation est effectué à l'extérieur et que les poteaux sont de couleur argentée. Les statistiques montrent que la performance de l'algorithme YuRHoS à l'intérieur était 0,085 fois moins bonne qu'à l'extérieur. De plus, la détection de la présence de poteaux de couleur argentée est 11 fois meilleure que celle de poteaux de couleur noir. / This research aimed to develop a technology system of mobility aid for Visually Impaired Person (VIP) called Smart Environment Explorer Stick (SEES).Particular purpose of this research was developing new algorithm in improving accuracy of SEE-stick for pole detection using distance calculation method and vertical line pair search based on Canny edge detection optimization and Hough transform. Henceforth, the pole detection algorithm was named as YuRHoS algorithm.The developed SEES as supporting system of VIP mobility aid had been successfully integrated several devices such as global remote server (iSEE), embedded local server (SEE-phone) and smart stick (SEE-stick). Performance of SEE-stick could be improved through YuRHoS algorithm, which was able to fix the accuracy of SEE-stick in detecting pole. Test comparison of pole detection results among others algorithm concluded that YuRHoS algorithm had better accuracy in pole detection.Two most significant factors affecting SEE-stick ability in detecting pole was test location and pole color. Level of accuracy of SEE-stick would be optimum once the test location was performed outdoor and pole color was silver. Statistics result shown that YuRHoS algorithm performance indoor was 0.085 times worse than outdoor. Meanwhile, silver-pole-color as object detection could increase YuRHoS algorithm performance as much as 11 times better compare to black-pole-color.

We present investigation and designing of a system that interacts with big audience, sitting in a dimmed theater environment. The goal is to automatically detect audiences and some of their actions. Test results indicate that because of low light condition we can’t rely on RGB camera footage in a dimmed environment. We use Microsoft Kinect Sensor to collect data from environment. Kinect is designed to be used with Microsoft Xbox 360 for gaming purposes. It has both RGB and Infrared depth camera. Change in amount of visible light doesn’t affect data from depth camera. Kinect is not a strong camera so it has limitations that we should deal with. Viewing angles of both cameras and depth range of Infrared camera are limited. Viewing angles of depth camera are 43° vertical and 57° horizontal. Most accurate range of depth camera is 1 meter to 4 meters from camera. Non-infrared reflective surfaces cause gaps in depth data. We evaluate possibility of using Kinect camera in a large environment with big audience. “Dome 3D theater” in Norrkoping Visualization Center C, is selected as environment to investigate and test the system. We ran some tests to find the best place and best height for camera to have most coverage. Our system works with optimized image processing algorithms that use 3D depth data instead of regular RGB or Grayscale image. We use “libfreenect”, Open Kinect library to get Kinect sensor up and running. C++ and OpenGL are used as programing languages and graphics interface, respectively. Open GLUT (OpenGL Utility Toolkit) is used for system’s user interface. It was not possible to use Dome environment for every test during the programming period so we recorded some depth footage and used for later tests. While evaluating the possibility of using Kinect in Dome environment, we realized that implementing a voting system would make a good demonstration and test application. Our system counts votes after audiences raise their hands to vote for something.

A method for detecting buildings from satellite/aerial images is proposed in this study. The aim is to extract rectilinear buildings by using hypothesize first verify next manner. Hypothesis generation is accomplished by using edge detection and line generation stages. Hypothesis verification is carried out by using information obtained both from the color segmentation of HSV representation of the image and the shadow detection stages&rsquo / output. Satellite/aerial image is firstly filtered to sharpen the edges. Then, edges are extracted using Canny edge detection algorithm. These edges are the input for the Hough Transform stage which will produce line segments according to these extracted edges. Then, extracted line segments are used to generate building hypotheses. Verification of these hypotheses makes use of the outputs of the HSV color segmentation and shadow detection stages. In this study, color segmentation is processed on the HSV representation of the satellite/aerial image which is less sensitive to illumination. In order to perform the shadow detection, the basic information which is shadow areas have higher value of saturation component and lower value of value component in HSV color space is used and according to this information a mask is applied to the HSV representation of the image to produce shadow pixels. The proposed method is implemented as software written in MATLAB programming software. The approach was tested in several different areas. The results are encouraging.

This thesis adresses designing a device that detects lane departure of a vehicle via a video feed from a camera module. This device is intended to be attached onto the windshield of the vehicle. The initial part of the thesis will cover the current methods of lane departure detection through a video feed. In the following part the selection of suitable hardware, specifically the latest model of a Raspberry Pi, has been made. Afterwards a suitable container for the aforementioned hardware has been designed and created using a 3D printer. Subsequently an appropriate LDWS algorithm is chosen and designed. In the next part, the range and parameters of a testing database through which the proper functionality of the device will be tested on are chosen. The final part of the thesis contains evaluation of the success rate of detection via the acquired database.

This diploma thesis deals with the issue of the recognition of road signs in the video sequence. Such systems increase the traffic safety and are implemented by major car factories in the manufactured cars (Opel, BMW). First, the motivation for the utilisation of these systems is presented, followed by the survey of the current state of the art methods. Finally, a specific road-sign detection method is chosen and described in detail. The method uses advanced techniques of signal processing. Segmentation method in color space is used for sign detection and subsequent classification is accomplished by linear classification with optional use of PCA method. In addition, the method contains the prediction of road sign positions based on Kalman filtering. Implemented system yields relatively accurate results and overall analysis and discussion is enclosed.

The aim of this thesis is to examine the possibility of a court detection program for sports videos that can identify the court even when some important elements are not visible. The study will also analyze what external factors may impact the program's accuracy in detecting all relevant elements.  These questions are answered through a combination of computer vision techniques and algorithms. The study utilizes Design Science Research (DSR) as its research methodology to develop an artifact. A dataset of padel sports videos are evaluated to measure the artifacts accuracy. The artifact utilizes multiple computer vision techniques from the OpenCV library to detect relevant lines and edges and project them onto the frame using a predetermined court model as reference.  The findings indicated that the developed artifact demonstrated a relatively consistent level of accuracy in court detection across multiple courts, whenever a detection was made. However, the frequency of successful detections exhibited some inconsistency. The research also found that external factors did not significantly influence the accuracy of court detection, yet they posed challenges to the program's overall consistency.

Image segmentation is a process to identify regions of interest from digital images. Image segmentation plays an important role in medical image processing which enables a variety of clinical applications. It is also a tool to facilitate the detection of abnormalities such as cancerous lesions in the brain. Although numerous efforts in recent years have advanced this technique, no single approach solves the problem of segmentation for the large variety of image modalities existing today. Consequently, brain MRI segmentation remains a challenging task. The purpose of this thesis is to demonstrate brain MRI segmentation for delineation of tumors, ventricles and other anatomical structures using Insight Segmentation and Registration Toolkit (ITK) routines as the foundation. ITK is an open-source software system to support the Visible Human Project. Visible Human Project is the creation of complete, anatomically detailed, three-dimensional representations of the normal male and female human bodies. Currently under active development, ITK employs leading-edge segmentation and registration algorithms in two, three, and more dimensions. A goal of this thesis is to implement those algorithms to facilitate brain segmentation for a brain cancer research scientist.

Perception, hence the usage of visual sensors is indispensable in mobile and autonomous robotics. Visual sensors such as cameras, rigidly mounted on a robot frame are the most common usage scenario. In this case, the motion of the camera due to the motion of the moving platform as well as the resulting shocks or vibrations causes a number of distortions on video frame sequences. Two most important ones are the frame-to-frame changes of the line-of-sight (LOS) and the presence of motion blur in individual frames. The latter of these two, namely motion blur plays a particularly dominant role in determining the performance of many vision algorithms used in mobile robotics. It is caused by the relative motion between the vision sensor and the scene during the exposure time of the frame. Motion blur is clearly an undesirable phenomenon in computer vision not only because it degrades the quality of images but also causes other feature extraction procedures to degrade or fail. Although there are many studies on feature based tracking, navigation, object recognition algorithms in the computer vision and robotics literature, there is no comprehensive work on the effects of motion blur on different image features and their extraction. In this thesis, a survey of existing models of motion blur and approaches to motion deblurring is presented. We review recent literature on motion blur and deblurring and we focus our attention on motion blur induced degradation of a number of popular feature detectors. We investigate and characterize this degradation using video sequences captured by the vision system of a mobile legged robot platform. Harris Corner detector, Canny Edge detector and Scale Invariant Feature Transform (SIFT) are chosen as the popular feature detectors that are most commonly used for mobile robotics applications. The performance degradation of these feature detectors due to motion blur are categorized to analyze the effect of legged locomotion on feature performance for perception. These analysis results are obtained as a first step towards the stabilization and restoration of video sequences captured by our experimental legged robotic platform and towards the development of motion blur robust vision system.