Rodrigues I.C.,Institute for Systems and Robotics |
Rodrigues I.C.,Polytechnic Institute of Lisbon |
Sanches J.M.R.,Institute for Systems and Robotics |
Sanches J.M.R.,University of Lisbon
IEEE Transactions on Image Processing | Year: 2011
Fluorescence confocal microscopy (FCM) is now one of the most important tools in biomedicine research. In fact, it makes it possible to accurately study the dynamic processes occurring inside the cell and its nucleus by following the motion of fluorescent molecules over time. Due to the small amount of acquired radiation and the huge optical and electronics amplification, the FCM images are usually corrupted by a severe type of Poisson noise. This noise may be even more damaging when very low intensity incident radiation is used to avoid phototoxicity. In this paper, a Bayesian algorithm is proposed to remove the Poisson intensity dependent noise corrupting the FCM image sequences. The observations are organized in a 3-D tensor where each plane is one of the images acquired along the time of a cell nucleus using the fluorescence loss in photobleaching (FLIP) technique. The method removes simultaneously the noise by considering different spatial and temporal correlations. This is accomplished by using an anisotropic 3-D filter that may be separately tuned in space and in time dimensions. Tests using synthetic and real data are described and presented to illustrate the application of the algorithm. A comparison with several state-of-the-art algorithms is also presented. © 2010 IEEE.
Cabral C.,University of Lisbon |
Silveira M.,University of Lisbon |
Silveira M.,Institute for Systems and Robotics |
Figueiredo P.,University of Lisbon |
Figueiredo P.,Institute for Systems and Robotics
Pattern Recognition | Year: 2012
Decoding perceptual or cognitive states based on brain activity measured using functional magnetic resonance imaging (fMRI) can be achieved using machine learning algorithms to train classifiers of specific stimuli. However, the high dimensionality and intrinsically low signal to noise ratio (SNR) of fMRI data poses great challenges to such techniques. The problem is aggravated in the case of multiple subject experiments because of the high inter-subject variability in brain function. To address these difficulties, the majority of current approaches uses a single classifier. Since, in many cases, different stimuli activate different brain areas, it makes sense to use a set of classifiers each specialized in a different stimulus. Therefore, we propose in this paper using an ensemble of classifiers for decoding fMRI data. Each classifier in the ensemble has a favorite class or stimulus and uses an optimized feature set for that particular stimulus. The output for each individual stimulus is therefore obtained from the corresponding classifier and the final classification is achieved by simply selecting the best score. The method was applied to three empirical fMRI datasets from multiple subjects performing visual tasks with four classes of stimuli. Ensembles of GNB and k-NN base classifiers were tested. The ensemble of classifiers systematically outperformed a single classifier for the two most challenging datasets. In the remaining dataset, a ceiling effect was observed which probably precluded a clear distinction between the two classification approaches. Our results may be explained by the fact that different visual stimuli elicit specific patterns of brain activation and indicate that an ensemble of classifiers provides an advantageous alternative to commonly used single classifiers, particularly when decoding stimuli associated with specific brain areas. © 2011 Elsevier Ltd. All rights reserved.
Sousa I.,Institute for Systems and Robotics |
Sousa I.,University of Lisbon |
Sousa I.,Siemens AG |
Vilela P.,Hospital da Luz |
And 2 more authors.
NeuroImage | Year: 2014
It has recently been proposed that hypocapnic cerebrovascular reactivity (CVR) can be assessed by measuring the blood oxygenation level dependent (BOLD) response to paced deep breathing (PDB) tasks inducing mild hypocapnia and vasoconstriction. In this work, we aim to assess the test-retest reproducibility and inter-subject variability of BOLD CVR measurements obtained using a PDB task and different methods to analyse the associated BOLD signal. The respiratory protocol consisted of alternating 40s of PDB with normal free breathing; expired CO2 pressure levels (PETCO2) were continuously monitored. CVR was quantified using either a timecourse curve analysis (TCA) approach, where the magnitude of response peaks is emphasized, or general linear modelling (GLM) including optimisation of the BOLD response latencies. The GLM fit was carried out using two types of response regressors: one that was computed as the convolution of PETCO2 traces with a gamma function and another that consisted of the convolution of PDB paradigm blocks with a physiological model of the respiratory response. Haemodynamic response latencies were optimised either on a voxel basis or for the whole imaging region. We found that the GLM method based on PDB task or PETCO2 traces and voxelwise optimisation of response latencies provided the most reproducible measures of CVR. For the average grey matter CVR, the inter-subject coefficient of variation (CVinter) / intra-subject coefficient of variation (CVintra) / intra-class correlation coefficient (ICC) were 20%/8%/0.8 and 27%/8%/0.9, using the task and PETCO2 timecourses, respectively. In terms of the spatial reproducibility, the group mean (±standard deviation) of the spatial ICC (ICCspatial) was 1.04±0.23 and 1.02±0.26, for the task and PETCO2 timecourses, respectively. These results indicate generally good reproducibility of the hypocapnic CVR maps obtained using the proposed PDB task and analysis methodology. This suggests that such protocol may therefore offer a promising alternative to conventional vasoactive challenges, which avoids their discomfort and difficulty. © 2014 Elsevier Inc.
Tahri O.,Institute for Systems and Robotics |
Mezouar Y.,University Blaise Pascal |
Chaumette F.,French Institute for Research in Computer Science and Automation |
Corke P.,Queensland University of Technology
IEEE Transactions on Robotics | Year: 2010
This paper proposes a generic decoupled image-based control scheme for cameras obeying the unified projection model. The scheme is based on the spherical projection model. Invariants to rotational motion are computed from this projection and used to control the translational degrees of freedom (DOFs). Importantly, we form invariants that decrease the sensitivity of the interaction matrix to object-depth variation. Finally, the proposed results are validated with experiments using a classical perspective camera as well as a fisheye camera mounted on a 6-DOF robotic platform. © 2010 IEEE.
Barata C.,Institute for Systems and Robotics |
Marques J.S.,Institute for Systems and Robotics |
Celebi M.E.,Louisiana State University
International Symposium on Image and Signal Processing and Analysis, ISPA | Year: 2013
The classification of skin lesions in dermoscopy images depends on three critical steps: i) lesion segmentation, ii) feature extraction and iii) feature classification. Lesion segmentation plays an important role since segmentation errors may jeopardize the other two steps, leading to erroneous decisions. This paper studies the robustness of a skin lesion classifier based on a Bag-of-features approach in the presence of segmentation errors. We compare the performance achieved by the system using an automatic segmentation algorithm with the performance obtained using manual segmentation provided by a specialist. We observe a degradation of the system accuracy by 8% when automatic segmentation is used. We also show that these results can be improved if manually segmented images are used in training phase, keeping a fully automatic solution during the testing phase. © 2013 University of Trieste and University of Zagreb.
Tahri O.,Institute for Systems and Robotics |
Araujo H.,Institute for Systems and Robotics
IEEE International Conference on Intelligent Robots and Systems | Year: 2012
Catadioptric cameras combine conventional cameras and mirrors to create omnidirectional sensors providing 360o panoramic views of a scene. Modeling such cameras has been subject of significant research interest in the computer vision community leading to a deeper understanding of the image properties and also to different models for different types of configurations. Visual servoing applications using catadioptric cameras have essentially been using central cameras and the corresponding unified projection model. So far only in very few cases more general models have been used. In this paper we address the problem of visual servoing using the so-called the radial model. The radial model can be applied to many camera configurations and in particular to non-central catadioptric systems with mirror shapes that are symmetric around the optical axis. In this case we show that the radial model can be used with a non-central catadioptric camera to allow effective image-based visual servoing (IBVS) of a mobile robot. Using this model, which is valid for a large set of catadioptric cameras, new visual features are proposed to control the degrees of freedom of a mobile robot moving on a plane. Several simulation results are provided to validate the effectiveness of such features. © 2012 IEEE.
Seabra J.,Institute for Systems and Robotics
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference | Year: 2010
Carotid plaques are the main cause of neurological symptoms due to distal embolization or flow reduction. An objective classification of such lesions into symptomatic or asymptomatic is crucial for optimal treatment planning.
Neves C.,Institute for Systems and Robotics |
Ventura R.,Institute for Systems and Robotics
Adaptive Mobile Robotics - Proceedings of the 15th International Conference on Climbing and Walking Robots and the Support Technologies for Mobile Machines, CLAWAR 2012 | Year: 2012
Traditional methods for robotic biped locomotion employing stiff actuation display low energy efficiency and high sensitivity to disturbances. However, a recent trend based on the use of passive elements together with actuation has emerged, inspired by biological locomotion. This paper presents a survey of methodologies for semi-passive legged locomotion, focusing on foot design and ankle control solutions, granting adaptability to uneven terrains. The semi- passive, or hybrid, approach for biped locomotion allows for the (1) maximiza- tion of energy efficiency and (2) significant improvement of adaptability to external disturbances. By exploring the dynamics of Energy Storage and Return (ESAR) elements, like springs or rubber bands, it is possible to power joints by releasing stored energy instead of only using torque provided by electrical mo- tors and to create gaits that minimize energy consumption. These elements also grant complacency to the robot which increases the gait's stability, specially on irregular terrains, and robustness to external disturbances. Copyright © 2012 by World Scientific Publishing Co. Pte. Ltd.
Gaspar T.,Institute for Systems and Robotics |
Oliveira P.,Institute for Systems and Robotics
Proceedings - IEEE International Conference on Robotics and Automation | Year: 2011
In this paper, new methodologies for the estimation of the depth of a target with unknown dimensions, based on depth from focus strategies, are proposed. The image-based measurements are detailed, through the minimization of a new functional, deeply rooted on optical characteristics of the lens system, namely the point spread function. This work complements an inexpensive single pan and tilt camera-based indoor positioning and tracking system, resorting to complementary filters for depth estimation. A motivation example is provided, where the target dimensions are assumed as known. Then, an extension corresponding to an higher-order filter is presented, that tackles the problem at hand. To assess the performance of the proposed system, a series of indoor experimental tests for a range of operation of up to ten meter were carried out. A centimetric accuracy was obtained under realistic conditions. © 2011 IEEE.
Batista P.,Institute for Systems and Robotics |
Silvestre C.,Institute for Systems and Robotics |
Oliveira P.,Institute for Systems and Robotics
Proceedings of the 2010 American Control Conference, ACC 2010 | Year: 2010
This paper addresses the problem of navigation of autonomous vehicles based on the range to a single beacon. The vehicle is equipped with a standard Inertial Measurement Unit (IMU) and range measurements to a single source are available as an aiding observation, in addition to angular velocity readings. The contribution of the paper is twofold: i) necessary and sufficient conditions on the observability of the system are derived that can be used for the motion planning and control of the vehicle; ii) a linear model is developed that mimics the exact dynamics of the nonlinear range-based system, and a Kalman filter is applied to estimate the relative position of the source, as well as the linear velocity of the vehicle and the acceleration of gravity, all expressed in body-fixed coordinates. Simulation results are presented in the presence of realistic measurement noise that illustrate the performance achieved with the proposed solution. © 2010 AACC.