La presente tesis está enmarcada en el área de visión por computador y en ella se realizan aportaciones encaminados a resolver el problema de segmentar automáticamente objetos en imágenes de escenas adquiridas en entornos donde se está realizando actividad, es decir, aparece movimiento de los elementos que la componen, y con iluminación variable o no controlada. Para llevar a cabo los desarrollos y poder evaluar prestaciones se ha abordado la resolución de dos problemas distintos desde el punto de vista de requerimientos y condiciones de entorno. En primer lugar se aborda el problema de segmentar e identificar, los códigos de los contenedores de camiones con imágenes tomadas en la entrada de un puerto comercial que se encuentra ubicada a la intemperie. En este caso se trata de proponer técnicas de segmentación que permitan extraer objetos concretos, en nuestro caso caracteres en contenedores, procesando imágenes individuales. No sólo supone un reto el trabajar con iluminación natural, sino además el trabajar con elementos deteriorados, con contrastes muy diferentes, etc. Dentro de este contexto, en la tesis se evalúan técnicas presentes en la literatura como LAT, Watershed, algoritmo de Otsu, variación local o umbralizado para segmentar imágenes en niveles de gris. A partir de este estudio, se propone una solución que combina varias de las técnicas anteriores, en un intento de abordar con éxito la extracción de caracteres de contenedores en todas las situaciones ambientales de movimiento e iluminación. El conocimiento a priori del tipo de objetos a segmentar nos permitió diseñar filtros con capacidad discriminante entre el ruido y los caracteres. El sistema propuesto tiene el valor añadido de que no necesita el ajuste de parámetros, por parte del usuario, para adaptarse a las variaciones de iluminación ambientales y consigue un nivel alto en la segmentación e identificación de caracteres. / Rosell Ortega, JA. (2011). Study of segmentation and identification techniques applied to environments with natural illumination and moving objects [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/10863 / Palancia

In this paper, a computer vision system for counting people standing in line is presented. In this application, common techniques such as Adaptive Background Subtraction (ABS), blob tracking with Kalman filter, and occlusion resistive techniques are used to detect and track people. Additionally, a novel method using Dual Adaptive Background Subtractors (DABS) is implemented for dynamically determining the line region in a real-world crowded scene, and also as an alternative target acquisition to regular ABS. The DABS technique acts as a temporal bandpass filter for motion, helping identify people standing in line while in the presence of other moving people. This is achieved by using two ABS with different temporal adaptiveness. Unlike other computer vision papers which perform tests in highly controlled environments, the DABS technique is tested in a crowded Starbucks© at the Cal Poly student union. For any length of people standing in line, result shows that DABS has a lower mean error by one or more people when compared to ABS. Even in challenging crowded scenes where the line can reach 19 people in length, DABS achieves a Normalized RMS Error of 43%.

This diploma thesis deals with posibilities of tracking object movement using computer vision algorithms. First chapters contain review of methods used for background subtraction, there are also listed basic detection approaches and thesis also mentions algorithms which allows tracking and movement prediction. Next part of this work informs about algoritms implemented in resulting software and its graphical user interface. Evaluation and comparison of original and modified algorithms is stationed at the end of this text.

This work proposes a system that utilizes both infrared and visual imagery to create a more robust object detection and classification system. The system consists of two main parts: a moving object detector and a target classifier. The first stage detects moving objects in visible and infrared spectrum using background subtraction based on Gaussian Mixture Models. Low-level fusion is performed to combine the foreground regions in the respective domain. For the second stage, a Convolutional Neural Network (CNN), pre-trained on the ImageNet dataset is used to classify the detected targets into one of the pre-defined classes; human and vehicle. The performance of the proposed object detector is evaluated using multiple video streams recorded in different areas and under various weather conditions, which form a broad basis for testing the suggested method. The accuracy of the classifier is evaluated from experimentally generated images from the moving object detection stage supplemented with publicly available CIFAR-10 and CIFAR-100 datasets. The low-level fusion method shows to be more effective than using either domain separately in terms of detection results. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

Context: On airport runways, monitoring is done by Precision Runway Monitor (PRM) method with the help of radar. Most of the airports are built near the forests so there is a greater chance of mam-mal intrusion onto the runways leading to massive accidents. At many airports, there are applied old traditional, mostly manual methods in detecting mammals on the runway. Accidents caused by wildlife strikes between aircraft and mammals are increasing day to day, and this is approximately 3%-10% of all reported collisions [1]. We propose a system that monitors the airport runway by detecting mammals. Objectives: The main objective of this project is to investigate and evaluate the possibility of using thermal vision methods to detect the obstacles encountered on the runways. The system should work in real time. Methods: Mammals detection can be done by using a thermal camera with a thermal sensitivity of less than 50mK and a resolution of 640 x 480 pixels. The thermal camera uses an uncooled microbolometer sensor which is lighter, consumes less power and can see through almost all weather conditions like mist, fog, snow etc. Machine Learning based algorithms like background subtraction are used in detecting the mammal, and contours are used to estimate the size and distance. Results: As a result, the mammals moving on the runway can be detected at a distance of up to 400 m. The system estimates a distance of a moving animal and its size with an accuracy of around 90%. Conclusions: A runway monitoring system is needed to prevent wildlife strikes in airports. The proposed system prevents accidents to some extent. However, further tests are required before its commercialisation. There is a need for further quantitative and qualitative validation of the models in full-scale industry trials.

Correct insertion of computer graphics into live-action broadcasts of outdoor sports requires precise knowledge of the foreground, i.e. players present in the scene. This thesis proposes a foreground detection and segmentation- framework with focus on real-time performance for 1080p resolution. A dataset consisting of four scenes; single-, multi-segment-, transcending-foreground and a light-witch scene all with dynamic backgrounds was constructed together with 26 ground-truths. Results show that the framework should run internally at 288p using GPU acceleration with geometrical nearest-neighbour-interpolation to attain real-time-capability. To maximize accuracy of the results, the framework uses two instances of OpenCV MOG2 in parallel on differently downsampled frames that are bitwise-joined to increase robustness. A set of morphological operations provides post-processing to get spatial coherence and a specific turf- consideration gives accurate contours. Thanks to additional camera- operator input, a crude distance-estimate lets foreground segments fade into background at a predetermined depth. The framework suffers from inaccurate segmentation during rapid light-switches, but recovers in a matter of seconds like the 'vanilla' MOG algorithm. For the specific scenes the framework provides excellent performance, especially considering the light-switch scene by comparison to the MOG-algorithm. For non-specific scenes of the 'BMC 2012' performance does not exceed the current state-of-the-art. / Korrekt placering av datorgrafik i video för tv-produktion kräver god känndedom om aktuell förgrund. Denna avhandling föreslår ett förgrundsdetektions- och segmenterings- ramverk med fokus på realtidsbearbetning av full-HD upplöst sport i utomhusmiljö. För utvärdering skapades ett dataset bestående av fyra scener; singel-, multisegment-, avlägsnande-förgrund och en ljusomväxlingsscen tillsammans med 26 referensförgrunder. För att erhålla realtidsbearbetning skall ramverket internt nyttja 288p upplösning med GPU acceleration och geometrisk närmaste-granne-interpolation. Resultaten visade att maximal noggranhet och ökad robusthet erhölls med två instanser av OpenCV MOG2 arbetandes parallellt på olikt nerskalade bilder för att därefter pixelvis förenas. För att erhålla sammanhängande förgrundssegment nyttjades morfologiska operationer på den binära sammansatta förgrunden vilket tillsammans med en specifik gräskantskorrektion ger precisa konturer. Tack vare givna kameraparametrar kan djupet till förgrundselementen uppskattas därmed låts de övergå till bakgrund för ett visst djupt. Ramverket lider av oprecis segmententering vid snabba ljusomväxlingar, men återhämtar sig när bakgrundsmodellen uppdaterats till de nya ljusförutsättningarna. För ovan nämnda specifika scener presterar ramverket utmärkt, speciellt med avseende på ljusomväxlingen, där prestandan är flerfaldigt bättre än den enskilda 'MOG'-metoden. För generella scener ur 'BMC 2012' datasetet presterar vår metod dock inte bättre än state-of-the-art.

No description available.

Raman spectroscopy is an effective optical analysis of the biochemically specific characterization of tissues without contrast agents or exogenous dyes. Applications of Raman spectroscopy include analysis and biomarker investigation, disease diagnosis and surgical guidance. One major challenge in Raman spectroscopy is removing inherent fluorescence background present in samples to acquire Raman signatures. In some tissues, like liver, kidney and darkly pigment skin, the auto-fluorescence background is strong enough to overwhelm the Raman peaks in conventional Near-Infrared (NIR) Raman systems. Recent publications have shown that using Raman systems with excitation sources with wavelengths beyond 830 nm and short-wave infrared (SWIR) InGaAs Array detectors resulted in dramatically reduced auto-fluorescence. The unique characteristics of Raman signals collected from SWIR systems versus NIR Raman systems requires inspection of the suitability of spectral pre-processing techniques. This thesis focused on the development of spectral processing techniques at three different steps; 1) detector background & noise reduction; 2) Auto-fluorescence background subtraction; 3) detection of outlier measurements to assist statistical classification. Detector background and noise reduction was compared between two different techniques, and a direct subtraction method resulted in better performance to reduce fixed pattern noise unique to InGaAs arrays. For the aim 2, three different algorithms for fluorescence background removal were developed, and a modified polynomial fitting method was found to be most appropriate for the low signal-to-noise (SNR) spectra. Finally, local outlier factor(LOF), a multivariate statistical outlier metric, was implemented in a two-stage fashion, and shown to be effective at identifying raw measurement errors and Raman spectra outliers. The overall outcome of this thesis was the evaluation of spectral processing techniques for SWIR Raman spectroscopy systems, and the development of specific techniques to optimize data quality and best prepare spectra for statistical analysis. / Bioengineering

This report shows that a reliable motion detection is needed to make an accurate prediction of future activity. Several experiments are carried out to obtain information about the object ́s behaviour and the best settings for the motion detection. A moving object is captured using two cameras, for two image sequences, and motion detection is applied to the stereoscopic data. Background subtraction algorithm followed by image segmentation algorithm, morphology algorithm, and blob analy- sis are performed on the images to find the coordinates for the centroid of the moving object. Two models are created to make a statistical inter- pretation of the data: one model for the height over the width and one statistical model for the distance between the cameras and the moving object over the width. The mean and standard deviation values are calculated to make a reliable interpretation of the captured images and the moving object. The Kalman filter is used for the prediction of future activity. The filters of the statistical models are trained with the first coordinates of the detected balls, and the next coordinates are predicted.

Background subtraction is the algorithmic process that segments out the region of interest often known as foreground from the background. Extensive literature and numerous algorithms exist in this domain, but most research have focused on videos captured by static cameras. The proliferation of portable platforms equipped with cameras has resulted in a large amount of video data being generated from moving cameras. This motivates the need for foundational algorithms for foreground/background segmentation in videos from moving cameras. In this dissertation, I propose three new types of background subtraction algorithms for moving cameras based on appearance, motion, and a combination of them. Comprehensive evaluation of the proposed approaches on publicly available test sequences show superiority of our system over state-of-the-art algorithms. The first method is an appearance-based global modeling of foreground and background. Features are extracted by sliding a fixed size window over the entire image without any spatial constraint to accommodate arbitrary camera movements. Supervised learning method is then used to build foreground and background models. This method is suitable for limited scene scenarios such as Pan-Tilt-Zoom surveillance cameras. The second method relies on motion. It comprises of an innovative background motion approximation mechanism followed by spatial regulation through a Mega-Pixel denoising process. This work does not need to maintain any costly appearance models and is therefore appropriate for resource constraint ego-vision systems. The proposed segmentation combined with skin cues is validated by a novel application on authenticating hand-gestured signature captured by wearable cameras. The third method combines both motion and appearance. Foreground probabilities are jointly estimated by motion and appearance. After the mega-pixel denoising process, the probability estimates and gradient image are combined by Graph-Cut to produce the segmentation mask. This method is universal as it can handle all types of moving cameras.