This Master’s Thesis was part of the project called Multimodalanalysis at the Depart-ment of Biomedical Engineering and Informatics at the Ume˚ University Hospital inUme˚ Sweden. The aim of the project is to develop multivariate measurement anda,analysis methods of the skeletal muscle physiology. One of the methods used to scanthe muscle is functional ultrasound. In a study performed by the project group datawas aquired, where test subjects were instructed to follow a certain exercise scheme,which was measured. Since there currently is no superior method to analyze the result-ing data (in form of ultrasound video sequences) several methods are being looked at.One considered method is called Independent Vector Analysis (IVA). IVA is a statisticalmethod to find independent components in a mix of components. This Master’s Thesisis about segmenting and analyzing the ultrasound images with help of IVA, to validateif it is a suitable method for this kind of tasks.First the algorithm was tested on generated mixed data to find out how well itperformed. The results were very accurate, considering that the method only usesapproximations. Some expected variation from the true value occured though.When the algorithm was considered performing to satisfactory, it was tested on thedata gathered by the study and the result can very well reflect an approximation of truesolution, since the resulting segmented signals seem to move in a possible way. But themethod has weak sides (which have been tried to be minimized) and all error analysishas been done by human eye, which definitly is a week point. But for the time being itis more important to analyze trends in the signals, rather than analyze exact numbers.So as long as the signals behave in a realistic way the result can not be said to becompletley wrong. So the overall results of the method were deemed adequate for the application at hand. / Multimodalanalys

This master thesis investigates the difficulties of constructing a depth map using one low resolution grayscale camera mounted in the front of a car. The goal is to produce a depth map in real-time to assist other algorithms in the safety system of a car. This has been shown to be difficult using the evaluated combination of camera position and choice of algorithms. The main problem is to estimate an accurate optical flow. Another problem is to handle moving objects. The conclusion is that the implementations, mainly triangulation of corresponding points tracked using a Lucas Kanade tracker, provide information of too poor quality to be useful for the safety system of a car. / I detta examensarbete undersöks svårigheterna kring att skapa en djupbild från att endast använda en lågupplöst gråskalekamera monterad framtill i en bil. Målet är att producera en djupbild i realtid som kan nyttjas i andra delar av bilens säkerhetssystem. Detta har visat sig vara svårt att lösa med den undersökta kombinationen av kameraplacering och val av algoritmer. Det huvudsakliga problemet är att räkna ut ett noggrant optiskt flöde. Andra problem härrör från objekt som rör på sig. Slutsatsen är att implementationerna, mestadels triangulering av korresponderande punktpar som följts med hjälp av en Lucas Kanade-följare, ger resultat av för dålig kvalitet för att vara till nytta för bilens säkerhetssystem.

Medical imaging is an important tool for diagnosis and treatment planning today. However as the demand for efficiency increases at the same time as the data volumes grow immensely, the need for computer assisted analysis, such as image segmentation, to help and guide the practitioner increases. Medical image segmentation could be used for various different tasks, the localization and delineation of pathologies such as cancer tumors is just one example. Numerous problems with noise and image artifacts in the generated images make the segmentation a difficult task, and the developer is forced to choose between speed and performance. In clinical practise, however, this is impossible as both speed and performance are crucial. One solution to this problem might be to involve the user more in the segmentation, using interactivite algorithms where the user might influence the segmentation for an improved result. This thesis has concentrated on finding a fast and interactive segmentation method for liver tumor segmentation. Various different methods were explored, and a few were chosen for implementation and further development. Two methods appeared to be the most promising, Bayesian Region Growing (BRG) and Level Set. An interactive Level Set algorithm emerged as the best alternative for the interactivity of the algorithm, and could be used in combination with both BRG and Level Set. A new data term based on a probability model instead of image edges was also explored for the Level Set-method, and proved to be more promising than the original one. The probability based Level Set and the BRG method both provided good quality results, but the fastest of the two was the BRG-method, which could segment a tumor present in 25 CT image slices in less than 10 seconds when implemented in Matlab and mex-C++ code on an ACPI x64-based PC with two 2.4 GHz Intel(R) Core(TM) 2CPU and 8 GB RAM memory. The interactive Level Set could be succesfully used as an interactive addition to the automatic method, but its usefulness was somewhat reduced by its slow processing time ( 1.5 s/slice) and the relative complexity of the needed user interactions.

In this thesis, an investigation was performed to find ways of differencing between firesand vehicles at waste stations in hope of removing vehicles as a source of error duringearly fire detection. The existing system makes use of a heat camera, which rotates in 48different angles (also known as zones) in a fixed position. If the heat is above a certainvalue within a zone the system sounds the fire alarm.The rotation of the camera results in an unwanted displacement between two successiveframes within the same zone. By use of image registration, this displacement wasremoved. After the registration of an image, segmentation was performed where coldobjects are eliminated as an error source. Lastly, an analysis was performed upon thewarm objects.At the end, it was proven that the image registration had been a successful improvementof the existing system. It was also shown that vehicles can, to some extent, beeliminated as an error source. / I denna uppsats görs en undersökning av sätt att urskilja mellan bränder och fordon vid avfallsbunkrar, i hopp om att ta bortfordon som felkälla under tidig branddetektion. Dagens system använder sig av en värmekamera som roterar i 48 vinklar(även kallade zoner) från en fix position och larmar då det blir för varmt i någon zon.Roteringen av kameran medför en icke önskvärd förskjutning mellan två efterföljande bilder inom samma zon. Processenbildregistrering används för att eliminera denna förskjutning. Efter registreringen utförs en segmentering där kalla objekt tasbort som felkälla. När detta är utfört görs en analys av de varma objekten med en mängd mätningar.I slutet bevisas att registreringen har fungerat mycket väl, likaså att det går till viss del att eliminera fordon som felkällaunder tidig brandetektion.

Two novel contributions to Content Based Image Retrieval are presented and discussed. The first is a search engine for font recognition. The intended usage is the search in very large font databases. The input to the search engine is an image of a text line, and the output is the name of the font used when printing the text. After pre-processing and segmentation of the input image, a local approach is used, where features are calculated for individual characters. The method is based on eigenimages calculated from edge filtered character images, which enables compact feature vectors that can be computed rapidly. A system for visualizing the entire font database is also proposed. Applying geometry preserving linear- and non-linear manifold learning methods, the structure of the high-dimensional feature space is mapped to a two-dimensional representation, which can be reorganized into a grid-based display. The performance of the search engine and the visualization tool is illustrated with a large database containing more than 2700 fonts. The second contribution is the inclusion of color-based emotion-related properties in image retrieval. The color emotion metric used is derived from psychophysical experiments and uses three scales: activity, weight and heat. It was originally designed for single-color combinations and later extended to include pairs of colors. A modified approach for statistical analysis of color emotions in images, involving transformations of ordinary RGB-histograms, is used for image classification and retrieval. The methods are very fast in feature extraction, and descriptor vectors are very short. This is essential in our application where the intended use is the search in huge image databases containing millions or billions of images. The proposed method is evaluated in psychophysical experiments, using both category scaling and interval scaling. The results show that people in general perceive color emotions for multi-colored images in similar ways, and that observer judgments correlate with derived values. Both the font search engine and the emotion based retrieval system are implemented in publicly available search engines. User statistics gathered during a period of 20 respectively 14 months are presented and discussed.

The trichromatic principle of representing color has for a long time been dominating in color imaging. The reason is the trichromatic nature of human color vision, but as the characteristics of typical color imaging devices are different from those of human eyes, there is a need to go beyond the trichromatic approach. The interest for multi-channel imaging, i.e. increasing the number of color channels, has made it an active research topic with a substantial potential of application. To achieve consistent color imaging, one needs to map the imaging-device data to the device-independent colorimetric representations CIEXYZ or CIELAB, the key concept of color management. As the color coordinates depend not only on the reflective spectrum of the object but also on the spectral properties of the illuminant, the colorimetric representation suffers from metamerism, i.e. objects of the same color under a specific illumination may appear different when they are illuminated by other light sources. Furthermore, when the sensitivities of the imaging device differ from the CIE color matching functions, two spectra that appear different for human observers may result in identical device response. On contrary, in multispectral imaging, color is represented by the object’s physical characteristics namely the spectrum which is illuminant independent. With multispectral imaging, different spectra are readily distinguishable, no matter they are metameric or not. The spectrum can then be transformed to any color space and be rendered under any illumination. The focus of the thesis is high quality image-acquisition in colorimetric and multispectral formats. The image acquisition system used is an experimental system with great flexibility in illumination and image acquisition setup. Besides the conventional trichromatic RGB filters, the system also provides the possibility of acquiring multi-channel images, using 7 narrowband filters. A thorough calibration and characterization of all the components involved in the image acquisition system is carried out. The spectral sensitivity of the CCD camera, which can not be derived by direct measurements, is estimated using least squares regression, optimizing the camera response to measured spectral reflectance of carefully selected color samples. To derive mappings to colorimetric and multispectral representations, two conceptually different approaches are used. In the model-based approach, the physical model describing the image acquisition process is inverted, to reconstruct spectral reflectance from the recorded device response. In the empirical approach, the characteristics of the individual components are ignored, and the functions are derived by relating the device response for a set of test colors to the corresponding colorimetric and spectral measurements, using linear and polynomial least squares regression. The results indicate that for trichromatic imaging, accurate colorimetric mappings can be derived by the empirical approach, using polynomial regression to CIEXYZ and CIELAB. Because of the media-dependency, the characterization functions should be derived for each combination of media and colorants. However, accurate spectral data reconstruction requires for multi-channel imaging, using the model-based approach. Moreover, the model-based approach is general, since it is based on the spectral characteristics of the image acquisition system, rather than the characteristics of a set of color samples. / Report code: LiU-TEK-LIC- 2006:70

In this thesis we study the problem of multi-session dense rgb-d slam for 3D reconstruc- tion. Multi-session reconstruction can allow users to capture parts of an object that could not easily be captured in one session, due for instance to poor accessibility or user mistakes. We first present a thorough overview of single-session dense rgb-d slam and describe the multi-session problem as a loosening of the incremental camera movement and static scene assumptions commonly held in the single-session case. We then implement and evaluate sev- eral variations on a system for doing two-session reconstruction as an extension to a single- session dense rgb-d slam system. The extension from one to several sessions is divided into registering separate sessions into a single reference frame, re-optimizing the camera trajectories, and fusing together the data to generate a final 3D model. Registration is done by matching reconstructed models from the separate sessions using one of two adaptations on a 3D object detection pipeline. The registration pipelines are evaluated with many different sub-steps on a challenging dataset and it is found that robust registration can be achieved using the proposed methods on scenes without degenerate shape symmetry. In particular we find that using plane matches between two sessions as constraints for as much as possible of the registration pipeline improves results. Several different strategies for re-optimizing camera trajectories using data from both ses- sions are implemented and evaluated. The re-optimization strategies are based on re- tracking the camera poses from all sessions together, and then optionally optimizing over the full problem as represented on a pose-graph. The camera tracking is done by incrementally building and tracking against a tsdf volume, from which a final 3D mesh model is extracted. The whole system is qualitatively evaluated against a realistic dataset for multi-session re- construction. It is concluded that the overall approach is successful in reconstructing objects from several sessions, but that other fine grained registration methods would be required in order to achieve multi-session reconstructions that are indistinguishable from singe-session results in terms of reconstruction quality.

Pedestrian detection is an important field with applications in active safety systems for cars as well as autonomous driving. Since autonomous driving and active safety are becoming technically feasible now the interest for these applications has dramatically increased.The aim of this thesis is to investigate convolutional neural networks (CNN) for pedestrian detection. The reason for this is that CNN have recently beensuccessfully applied to several different computer vision problems. The main applications of pedestrian detection are in real time systems. For this reason,this thesis investigates strategies for reducing the computational complexity offorward propagation for CNN.The approach used in this thesis for extracting pedestrians is to use a CNN tofind a probability map of where pedestrians are located. From this probabilitymap bounding boxes for pedestrians are generated. A method for handling scale invariance for the objects of interest has also been developed in this thesis. Experiments show that using this method givessignificantly better results for the problem of pedestrian detection.The accuracy which this thesis has managed to achieve is similar to the accuracy for some other works which use CNN.

Because of the increase in the number of security cameras, there is more video footage available than a human could efficiently process. In combination with the fact that computers are getting more efficient, it is getting more and more interesting to solve the problem of detecting and recognizing people automatically. Therefore a method is proposed for estimating a 3D-path of a person of interest in multiple, non overlapping, monocular cameras. This project is a collaboration between two master theses. This thesis will focus on recognizing a person of interest from several possible candidates, as well as estimating the 3D-position of a person and providing a graphical user interface for the system. The recognition of the person of interest includes keeping track of said person frame by frame, and identifying said person in video sequences where the person of interest has not been seen before. The final product is able to both detect and recognize people in video, as well as estimating their 3D-position relative to the camera. The product is modular and any part can be improved or changed completely, without changing the rest of the product. This results in a highly versatile product which can be tailored for any given situation.

In most cases today when a specific person's whereabouts is monitored through video surveillance it is done manually and his or her location when not seen is based on assumptions on how fast he or she can move. Since humans are good at recognizing people this can be done accurately, given good video data, but the time needed to go through all data is extensive and therefore expensive. Because of the rapid technical development computers are getting cheaper to use and therefore more interesting to use for tedious work. This thesis is a part of a larger project that aims to see to what extent it is possible to estimate a person of interest's time dependent 3D position, when seen in surveillance videos. The surveillance videos are recorded with non overlapping monocular cameras. Furthermore the project aims to see if the person of interest's movement, when position data is unavailable, could be predicted. The outcome of the project is a software capable of following a person of interest's movement with an error estimate visualized as an area indicating where the person of interest might be at a specific time. This thesis main focus is to implement and evaluate a people detector meant to be used in the project, reduce noise in position measurement, predict the position when the person of interest's location is unknown, and to evaluate the complete project. The project combines known methods in computer vision and signal processing and the outcome is a software that can be used on a normal PC running on a Windows operating system. The software implemented in the thesis use a Hough transform based people detector and a Kalman filter for one step ahead prediction. The detector is evaluated with known methods such as Miss-rate vs. False Positives per Window or Image (FPPW and FPPI respectively) and Recall vs. 1-Precision. The results indicate that it is possible to estimate a person of interest's 3D position with single monocular cameras. It is also possible to follow the movement, to some extent, were position data are unavailable. However the software needs more work in order to be robust enough to handle the diversity that may appear in different environments and to handle large scale sensor networks.