Efficient and accurate set-based registration of time-separated aerial images

This paper addresses the task of time-separated aerial image registration. The ability to solve this problem accurately and reliably is important for a variety of subsequent image understanding applications. The principal challenge lies in the extent and nature of transient appearance variation that a land area can undergo, such as that caused by the change under illumination conditions, seasonal variations, or the occlusion by non-persistent objects (people, cars). Our work introduces several major novelties (i) unlike previous work on aerial image registration, we approach the problem using a set-based paradigm; (ii) we show how image space local, pair-wise constraints can be used to enforce a globally good registration using a constraints graph structure; (iii) we show how a simple holistic representation derived from raw aerial images can be used as a basic building block of the constraints graph in a manner which achieves both high registration accuracy and speed; (iv) lastly, we introduce a new and, to the best of our knowledge, the only data corpus suitable for the evaluation of set-based aerial image registration algorithms. Using this data set, we demonstrate (i) that the proposed method outperforms the state-of-the-art for pair-wise registration already, achieving greater accuracy and reliability, while at the same time reducing the computational cost of the task and (ii) that the increase in the number of available images in a set consistently reduces the average registration error, with a major difference already for a single additional image.

Localization and degmentation of 3D intervertebral discs in MR Images by data driven estimation

This paper addresses the problem of fully-automatic localization and segmentation of 3D intervertebral discs (IVDs) from MR images. Our method contains two steps, where we first localize the center of each IVD, and then segment IVDs by classifying image pixels around each disc center as foreground (disc) or background. The disc localization is done by estimating the image displacements from a set of randomly sampled 3D image patches to the disc center. The image displacements are estimated by jointly optimizing the training and test displacement values in a data-driven way, where we take into consideration both the training data and the geometric constraint on the test image. After the disc centers are localized, we segment the discs by classifying image pixels around disc centers as background or foreground. The classification is done in a similar data-driven approach as we used for localization, but in this segmentation case we are aiming to estimate the foreground/background probability of each pixel instead of the image displacements. In addition, an extra neighborhood smooth constraint is introduced to enforce the local smoothness of the label field. Our method is validated on 3D T2-weighted turbo spin echo MR images of 35 patients from two different studies. Experiments show that compared to state of the art, our method achieves better or comparable results. Specifically, we achieve for localization a mean error of 1.6-2.0 mm, and for segmentation a mean Dice metric of 85%-88% and a mean surface distance of 1.3-1.4 mm.

Moving images effective teaching with film and television in management

This book will inspire academics, teachers and trainers to use film and television in their classrooms and to shows them how it might be done. It brings together respected international scholars who recount their experiences of how they have used moving images in their classrooms (defined widely to include distance-learning) with their explanations of why they chose this method of teaching and how they put their intentions into action. The book also illustrates how particular subjects might be taught using film and television as an inspiration to demonstrate the range of opportunities that these media offer. Finally, this book considers some of the practical issues in using film and television in the classroom such as copyright, technology, and the representation of reality and drama in films. This is a 'practical, how to' book that answers the questions of those people who have considered using film and television in their classroom but until now have shied away from doing so. The opportunity to see how others have used film effectively breaks down psychological barriers and makes it seem both realistic and worthwhile.

Fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images via a learning-based method

In this paper, we address the problems of fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images. We propose a learning-based, unified random forest regression and classification framework to tackle these two problems. More specifically, in the first stage, the localization of 3D vertebral bodies is solved with random forest regression where we aggregate the votes from a set of randomly sampled image patches to get a probability map of the center of a target vertebral body in a given image. The resultant probability map is then further regularized by Hidden Markov Model (HMM) to eliminate potential ambiguity caused by the neighboring vertebral bodies. The output from the first stage allows us to define a region of interest (ROI) for the segmentation step, where we use random forest classification to estimate the likelihood of a voxel in the ROI being foreground or background. The estimated likelihood is combined with the prior probability, which is learned from a set of training data, to get the posterior probability of the voxel. The segmentation of the target vertebral body is then done by a binary thresholding of the estimated probability. We evaluated the present approach on two openly available datasets: 1) 3D T2-weighted spine MR images from 23 patients and 2) 3D spine CT images from 10 patients. Taking manual segmentation as the ground truth (each MR image contains at least 7 vertebral bodies from T11 to L5 and each CT image contains 5 vertebral bodies from L1 to L5), we evaluated the present approach with leave-one-out experiments. Specifically, for the T2-weighted MR images, we achieved for localization a mean error of 1.6 mm, and for segmentation a mean Dice metric of 88.7% and a mean surface distance of 1.5 mm, respectively. For the CT images we achieved for localization a mean error of 1.9 mm, and for segmentation a mean Dice metric of 91.0% and a mean surface distance of 0.9 mm, respectively.

Groupwise registration of aerial images

This paper addresses the task of time separated aerial image registration. The ability to solve this problem accurately and reliably is important for a variety of subsequent image understanding applications. The principal challenge lies in the extent and nature of transient appearance variation that a land area can undergo, such as that caused by the change in illumination conditions, seasonal variations, or the occlusion by non-persistent objects (people, cars). Our work introduces several novelties: (i) unlike all previous work on aerial image registration, we approach the problem using a set-based paradigm; (ii) we show how local, pairwise constraints can be used to enforce a globally good registration using a constraints graph structure; (iii) we show how a simple holistic representation derived from raw aerial images can be used as a basic building block of the constraints graph in a manner which achieves both high registration accuracy and speed. We demonstrate: (i) that the proposed method outperforms the state-of-the-art for pair-wise registration already, achieving greater accuracy and reliability, while at the same time reducing the computational cost of the task; and (ii) that the increase in the number of available images in a set consistently reduces the average registration error.

Multi-view subspace clustering for face images

In many real-world computer vision applications, such as multi-camera surveillance, the objects of interest are captured by visual sensors concurrently, resulting in multi-view data. These views usually provide complementary information to each other. One recent and powerful computer vision method for clustering is sparse subspace clustering (SSC); however, it was not designed for multi-view data, which break down its linear separability assumption. To integrate complementary information between views, multi-view clustering algorithms are required to improve the clustering performance. In this paper, we propose a novel multi-view subspace clustering by searching for an unified latent structure as a global affinity matrix in subspace clustering. Due to the integration of affinity matrices for each view, this global affinity matrix can best represent the relationship between clusters. This could help us achieve better performance on face clustering. We derive a provably convergent algorithm based on the alternating direction method of multipliers (ADMM) framework, which is computationally efficient, to solve the formulation. We demonstrate that this formulation outperforms other alternatives based on state-of-The-Arts on challenging multi-view face datasets.

Pose estimation of a humanoid robot using images from a mobile external camera

NOGUEIRA, Marcelo B. ; MEDEIROS, Adelardo A. D. ; ALSINA, Pablo J. Pose Estimation of a Humanoid Robot Using Images from an Mobile Extern Camera. In: IFAC WORKSHOP ON MULTIVEHICLE SYSTEMS, 2006, Salvador, BA. Anais... Salvador: MVS 2006, 2006.

Object-Space Road Extraction in Rural Areas Using Stereoscopic Aerial Images

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

DETERMINATION of THE INDIRECT ORIENTATION of ORBITAL PUSHBROOM IMAGES USING CONTROL STRAIGHT LINES

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Image processing algorithm for the identification of Martian dust devil tracks in MOC and HiRISE images

This paper presents a method for automatic identification of dust devils tracks in MOC NA and HiRISE images of Mars. The method is based on Mathematical Morphology and is able to successfully process those images despite their difference in spatial resolution or size of the scene. A dataset of 200 images from the surface of Mars representative of the diversity of those track features was considered for developing, testing and evaluating our method, confronting the outputs with reference images made manually. Analysis showed a mean accuracy of about 92%. We also give some examples on how to use the results to get information about dust devils, namelly mean width, main direction of movement and coverage per scene. (c) 2012 Elsevier Ltd. All rights reserved.

Magnetic images of the disintegration process of tablets in the human stomach by ac biosusceptometry

Oral administration of solid dosage forms is usually preferred in drug therapy. Conventional imaging methods are essential tools to investigate the in vivo performance of these formulations. The non-invasive technique of ac biosusceptometry has been introduced as an alternative in studies focusing on gastrointestinal motility and, more recently, to evaluate the behaviour of magnetic tablets in vivo. The aim of this work was to employ a multisensor ac biosusceptometer system to obtain magnetic images of disintegration of tablets in vitro and in the human stomach. The results showed that the transition between the magnetic marker and the magnetic tracer characterized the onset of disintegration (t(50)) and occurred in a short time interval (1.1 +/- 0.4 min). The multisensor ac biosusceptometer was reliable to monitor and analyse the in vivo performance of magnetic tablets showing accuracy to quantify disintegration through the magnetic images and to characterize the profile of this process.

Application of wavelets to the evaluation of phantom images for mammography quality control

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Pixel intensity and fractal analyses: detecting osteoporosis in perimenopausal and postmenopausal women by using digital panoramic images

Objective. Pixel intensity values (PI) and fractal dimensions (FD) were compared in selected mandibular regions on digital panoramic images of normal, osteopenic, and osteoporotic perimenopausal and postmenopausal women to evaluate their relative efficacies in detecting osteoporotic-associated bone density changes.Study design. Standardized mandibular angle, body, and canine/premolar (C/PM) regions on 54 charge-coupied device (CCD) digital panoramic images of normal and potentially osteoporotic postmenopausal women were analyzed for PI and FD. Lumbar spine and femoral neck dual-energy x-ray absorptiometry QXA) on each patient served as the reference standard examination. Pearson correlation coefficients and analysis of variance (ANOVA) were performed.Results. There was significant correlation among PI measurements (P < 0.01), and no significant correlation between FD. C/PM had significantly lower PI than control C/PM (P = 0.049).Conclusions. Osteoporotic changes in mandibular C/PM cancellous bone were detected in our study population on CCD digital panoramic images by using a robust image analysis paradigm. Future automated application of such image analysis could enable widespread, cost effective screening for osteoporosis in dental settings.

Evaluation of the RB-RB/LB-LB mnemonic rule for recording optimally projected intraoral images of dental implants: an in vitro study

Objective: The aim of this study was to evaluate a simple mnemonic rule (the RB-RB/LB-LB rule) for recording intra-oral radiographs with optimal projection for the control of dental implants.Methods: 30 third-year dental students received a short lesson in the RB-RB/LB-LB mnemonic rule. The rule is as follows: if right blur then raise beam (RB-RB), i.e. if implant threads are blurred at the right side of the implant, the X-ray beam direction must be raised towards the ceiling to obtain sharp threads on both implant sides; if left blur then lower beam (LB-LB), i.e. if implant threads are blurred at the left side of the implant, the X-ray beam direction must be lowered towards the floor to obtain sharp threads on both implant sides. Intra-oral radiographs of four screw-type implants placed with different inclination in a Frasaco upper or lower jaw dental model (Frasaco GmbH, Tettnang, Germany) were recorded. The students were unaware of the inclination of the implants and were instructed to re-expose each implant, implementing the mnemonic rule, until an image of the implant with acceptable quality (subjectively judged by the instructor) was obtained. Subsequently, each radiograph was blindly assessed with respect to sharpness of the implant threads and assigned to one of four quality categories: (1) perfect, (2) not perfect, but clinically acceptable, (3) not acceptable and (4) hopeless.Results: For all implants, from one non-perfect exposure to the following, a higher score was obtained in 64% of the cases, 28% received the same score and 8% obtained a lower score. Only a small variation was observed among exposures of implants with different inclination. on average, two exposures per implant (range: one to eight exposures) were needed to obtain a clinically acceptable image.Conclusion: The RB-RB/LB-LB mnemonic rule for recording intra-oral radiographs of dental implants with a correct projection was easy to implement by inexperienced examiners. Dentomaxillofacial Radiology (2012) 41, 298-304. doi: 10.1259/dmfr/20861598