There has been extensive research investigating the differences between global and local feature discrimination. The role that global and local feature discrimination has in sustained attention tasks however has been relatively neglected. In the current research, participants were required to perform a sustained attention task requiring them to engage in either global or local shape stimuli discrimination. Reaction times to local feature discrimination revealed a quadratic trend with time-on-task, with performance levels showing a decline before returning to initial levels towards the end of the task. This trend was not found in the global shape discrimination condition. Functional near-infrared spectroscopy (fNIRS) was employed to assess hemispheric cerebral oxygenation during the tasks. It was found in both conditions that there was greater oxygenation in the right hemisphere compared to the left hemisphere. It was also found that right hemisphere oxygenation increased with time-on-task. Left hemisphere oxygenation decreased during the global task, while it increased during the local task with time on task. Total cerebral oxygenation, collapsed over both hemispheres, increased more over time in the local discrimination task than the global discrimination task. The performance data and the fNIRS results suggest an increased utilization of bilateral cognitive resources with time-on-task in the local discrimination condition, but not in the global discrimination condition. Results and implications are discussed.

Object tracking is an active research topic in computer vision and has appli- cation in several areas, such as event detection and robotics. Vehicle tracking is used in Intelligent Transport System (ITS) and surveillance systems. Its re- liability is critical to the overall performance of these systems. Feature-based methods that are used to represent distinctive content in visual frames are one approach to vehicle tracking. Existing feature-based tracking systems can only track vehicles under ideal conditions. They have difficulties when used under a variety of conditions, for example, during both the day and night. They are highly dependent on stable local features that can be tracked for a long time period. These local features are easily lost because of their local property and image noise caused by factors such as, headlight reflections and sun glare. This thesis presents a new approach, addressing the reliability issues mentioned above, tracking whole feature groups composed of feature points extracted with the Scale Invariant Feature Transform (SIFT) algorithm. A feature group in- cludes several features that share a similar property over a time period and can be tracked to the next frame by tracking individual feature points inside it. It is lost only when all of the features in it are lost in the next frame. We cre- ate these feature groups by clustering individual feature points using distance, velocity and acceleration information between two consecutive frames. These feature groups are then hierarchically clustered by their inter-group distance, velocity and acceleration information. Experimental results show that the pro- posed vehicle tracking system can track vehicles with the average accuracy of over 95%, even when the vehicles have complex motions in noisy scenes. It gen- erally works well even in difficult environments, such as for rainy days, windy days, and at night. We are surprised to find that our tracking system locates and tracks motor bikes and pedestrians. This could open up wider opportunities and further investigation and experiments are required to confirm the tracking performance for these objects. Further work is also required to track more com- plex motions, such as rotation and articulated objects with different motions on different parts.

Robotic automation has been a part of state-of-the-art manufacturing for many decades. Robotic manipulators are used for such tasks as welding, painting, pick and place tasks etc. Robotic manipulators are quite flexible and adaptable to new tasks, but a typical robot-based production cell requires extensive specification of the robot motion and construction of tools and fixtures for material handling. This incurs a large effort both in time and monetary expenses. The task of a vision system in this setting is to simplify the control and guidance of the robot and to reduce the need for supporting material handling machinery. This dissertation examines performance and properties of the current state-of-the-art local features within the setting of object pose estimation. This is done through an extensive set of experiments replicating various potential problems to which a vision system in a robotic cell could be subjected. The dissertation presents new local features which are shown to increase the performance of object pose estimation. A new local descriptor details how to use log-polar sampled image patches for truly rotational invariant matching. This representation is also extended to use a scale-space interest point detector which in turn makes it very competitive in our experiments. A number of variations of already available descriptors are constructed resulting in new and competitive features, among them a scale-space based Patch-duplet. In this dissertation a successful vision-based object pose estimation system is extended for multi-cue integration, yielding increased robustness and accuracy. Robustness is increased through algorithmic multi-cue integration, combining the individual strengths of multiple local features. Increased accuracy is achieved by utilizing manipulator movement and applying temporal multi-cue integration. This is implemented using a real flexible robotic manipulator arm. Besides work done on local features for ordinary image data a number of local features for range data has also been developed. This dissertation describes the theory behind and the application of the scene tensor to the problem of object pose estimation. The scene tensor is a fourth order tensor representation using projective geometry. It is shown how to use the scene tensor as a detector as well as how to apply it to the task of object pose estimation. The object pose estimation system is extended to work with 3D data. A novel way of handling sampling of range data when constructing a detector is discussed. A volume rasterization method is presented and the classic Harris detector is adapted to it. Finally, a novel region detector, called Maximally Robust Range Regions, is presented. All developed detectors are compared in a detector repeatability test.

Object tracking is an active research topic in computer vision and has appli- cation in several areas, such as event detection and robotics. Vehicle tracking is used in Intelligent Transport System (ITS) and surveillance systems. Its re- liability is critical to the overall performance of these systems. Feature-based methods that are used to represent distinctive content in visual frames are one approach to vehicle tracking. Existing feature-based tracking systems can only track vehicles under ideal conditions. They have difficulties when used under a variety of conditions, for example, during both the day and night. They are highly dependent on stable local features that can be tracked for a long time period. These local features are easily lost because of their local property and image noise caused by factors such as, headlight reflections and sun glare. This thesis presents a new approach, addressing the reliability issues mentioned above, tracking whole feature groups composed of feature points extracted with the Scale Invariant Feature Transform (SIFT) algorithm. A feature group in- cludes several features that share a similar property over a time period and can be tracked to the next frame by tracking individual feature points inside it. It is lost only when all of the features in it are lost in the next frame. We cre- ate these feature groups by clustering individual feature points using distance, velocity and acceleration information between two consecutive frames. These feature groups are then hierarchically clustered by their inter-group distance, velocity and acceleration information. Experimental results show that the pro- posed vehicle tracking system can track vehicles with the average accuracy of over 95%, even when the vehicles have complex motions in noisy scenes. It gen- erally works well even in difficult environments, such as for rainy days, windy days, and at night. We are surprised to find that our tracking system locates and tracks motor bikes and pedestrians. This could open up wider opportunities and further investigation and experiments are required to confirm the tracking performance for these objects. Further work is also required to track more com- plex motions, such as rotation and articulated objects with different motions on different parts.

How does the focus of attention influence the encoding of information? Research has shown that size and allocation of the attentional window has an influence on what information is attended to or missed. The size-scale of features also effects processing of visual information. Previous research involving hierarchical stimuli suggests precedence for global features. In the present experiment, I investigated the influence of attentional window size on accuracy of encoding hierarchical stimuli at the global and local level. Here I introduce a new method for manipulating the size of the attentional window and for collecting unconstrained responses. At the start of each trial, observers tracked a dashed-line rectangular box, which either broadened or narrowed in size after onset. This sequence was immediately followed by a brief presentation of two hierarchical letters presented simultaneously on the left and right sides of the screen. The box preceding the hierarchical letters either broadened to a size large enough to include both letters at the global level, or narrowed to a size small enough to include a maximum of two letters at the local level at either side of the screen. Observers reported all letters they were able to identify. Results from two experiments indicate an overall precedence of global letters. However, a narrow attentional window reduced global precedence, as would be expected with more focused attention. The narrow windows also produced more same-side identifications of both global and local letters. The second experiment also showed that reducing the processing time decreased the global advantage.

<p>This thesis presents methods useful in a bin picking application, such as detection and representation of local features, pose estimation and multi-cue integration.</p><p>The scene tensor is a representation of multiple line or edge segments and was first introduced by Nordberg in [30]. A method for estimating scene tensors from gray-scale images is presented. The method is based on orientation tensors, where the scene tensor can be estimated by correlations of the elements in the orientation tensor with a number of 1<em>D</em> filters. Mechanisms for analyzing the scene tensor are described and an algorithm for detecting interest points and estimating feature parameters is presented. It is shown that the algorithm works on a wide spectrum of images with good result.</p><p>Representations that are invariant with respect to a set of transformations are useful in many applications, such as pose estimation, tracking and wide baseline stereo. The scene tensor itself is not invariant and three different methods for implementing an invariant representation based on the scene tensor is presented. One is based on a non-linear transformation of the scene tensor and is invariant to perspective transformations. Two versions of a tensor doublet is presented, which is based on a geometry of two interest points and is invariant to translation, rotation and scaling. The tensor doublet is used in a framework for view centered pose estimation of 3<em>D</em> objects. It is shown that the pose estimation algorithm has good performance even though the object is occluded and has a different scale compared to the training situation.</p><p>An industrial implementation of a bin picking application have to cope with several different types of objects. All pose estimation algorithms use some kind of model and there is yet no model that can cope with all kinds of situations and objects. This thesis presents a method for integrating cues from several pose estimation algorithms for increasing the system stability. It is also shown that the same framework can also be used for increasing the accuracy of the system by using cues from several views of the object. An extensive test with several different objects, lighting conditions and backgrounds shows that multi-cue integration makes the system more robust and increases the accuracy.</p><p>Finally, a system for bin picking is presented, built from the previous parts of this thesis. An eye in hand setup is used with a standard industrial robot arm. It is shown that the system works for real bin-picking situations with a positioning error below 1 mm and an orientation error below 1^o degree for most of the different situations.</p> / Report code: LiU-TEK-LIC-2005:15.

Novel methods of object recognition that form a bridge between today&rsquo / s local feature frameworks and previous decade&rsquo / s strong but deserted geometric invariance field are presented in this dissertation. The rationale behind this effort is to complement the lowered discriminative capacity of local features, by the invariant geometric descriptions. Similar to our predecessors, we first start with constrained cases and then extend the applicability of our methods to more general scenarios. Local features approach, on which our methods are established, is reviewed in three parts / namely, detectors, descriptors and the methods of object recognition that employ them. Next, a novel planar object recognition framework that lifts the requirement for exact appearance-based local feature matching is presented. This method enables matching of groups of features by utilizing both appearance information and group geometric descriptions. An under investigated area, scene logo recognition, is selected for real life application of this method. Finally, we present a novel method for three-dimensional (3D) object recognition, which utilizes well-known local features in a more efficient way without any reliance on partial or global planarity. Geometrically consistent local features, which form the crucial basis for object recognition, are identified using affine 3D geometric invariants. The utilization of 3D geometric invariants replaces the classical 2D affine transform estimation /verification step, and provides the ability to directly verify 3D geometric consistency. The accuracy and robustness of the proposed method in highly cluttered scenes with no prior segmentation or post 3D reconstruction requirements, are presented during the experiments.

This work deals with medical image analysis, using variety of statisic and numeric methods implemented in Eclipse and Rapidminer environments in Java programming language. Sets of images (slices), which are used here, are the results of magnetic resonance brain examination of several subejcts. Segments in this 3D image are analyzed and some local features are computed, based on which data sets for use in training algorythms are generated. The ability of successful identification of healthy or unhealthy tissues is then practically tested using available data.

Ce travail s’inscrit dans une tentative de liaison entre la communauté classique de la Vision par ordinateur et la communauté du traitement d’images de documents, analyse être connaissance (DAR). Plus particulièrement, nous abordons la question des détecteurs de points d’intérêts et des descripteurs locaux dans une image. Ceux-ci ayant été conçus pour des images issues du monde réel, ils ne sont pas adaptés aux problématiques issues du document dont les images présentent des caractéristiques visuelles différentes.Notre approche se base sur la résolution du problème de la confusion entre les descripteurs,ceux-ci perdant leur pouvoir discriminant. Notre principale contribution est un algorithme de réduction de la confusion potentiellement présente dans un ensemble de vecteurs caractéristiques d’une même image, ceci par une approche probabiliste en filtrant les vecteurs fortement confusifs. Une telle conception nous permet d’appliquer des algorithmes d’extractions de descripteurs sans avoir à les modifier ce qui constitue une passerelle entre ces deux mondes. / This work tries to establish a bridge between the field of classical computer vision and document analysis and recognition. Specificaly, we tackle the issue of keypoints detection and associated local features computation in the image. These are not suitable for document images since they were designed for real-world images which have different visual characteristic. Our approach is based on resolving the issue of reducing the confusion between feature vectors since they usually lose their discriminant power with document images. Our main contribution is an algorithm reducing the confusion between local features by filtering the ones which present a high confusing risk. We are tackling this by using tools from probability theory. Such a method allows us to apply features extraction algorithms without having to modify them, thus establishing a bridge between these two worlds.

Práce představuje novou adaptabilní metodu pro on-line vyhledávání videa v reálném čase pomocí vizuálních slovníků. Nová metoda se zaměřuje na nízkou výpočetní náročnost a přesnost vyhledání při on-line použití. Metoda vychází z technik využitých u statických vizuálních slovníků. Tyto běžné techniky jsou upraveny tak, aby byly schopné se adaptovat na proměnlivá data. Postupy, které toto u nové metody řeší, jsou - dynamická inverzní frekvence dokumentů, adaptabilní vizuální slovník a proměnlivý invertovaný index. Navržený postup byl vyhodnocen na úloze vyhledávání videa a prezentované výsledky ukazují, jaké vlastnosti má adaptabilní metoda ve srovnání se statickým přístupem. Nová adaptabilní metoda je založena na konceptu plovoucího okna, který definuje, jakým způsobem se vybírají data pro adaptaci a ke zpracování. Společně s konceptem je definován i matematický aparát, který umožňuje vyhodnotit, jak koncept nejlépe využít pro různé metody zpracování videa. Praktické využití adaptabilní metody je konkrétně u systémů pro zpracování videa, kde se očekává změna v charakteru vizuálních dat nebo tam, kde není předem známo, jakého charakteru vizuální data budou.