The authors propose a two-step algorithm for almost unsupervised detection of linear structures, in particular, main axes road networks, as seen synthetic aperture radar (SAR) images. first step is local and used to extract features from the speckle image, which are treated road-segment candidates. present two line detectors well method fusing information these detectors. In second global step, they identify real roads among segment candidates by defining Markov random field (MRF) on set...

We propose a classification method suitable for high-resolution synthetic aperture radar (SAR) images over urban areas. When processing SAR images, there is strong need statistical models of scattering to take into account multiplicative noise and high dynamics. For instance, the process needs be based on use statistics. Our main contribution choice an accurate model areas its in Markovian algorithm. Clutter becomes non-Gaussian when resolution or area man-made. Many have been proposed fit...

Because of the importance security issues in management medical information, we suggest use watermarking techniques to complete existing measures for protecting images. We discuss necessary requirements such a system be accepted by staff and its complementary role with respect systems. present different scenarios, one devoted authentication tracing images, second integrity control patient's record.

In this letter, we are interested in the annotation of large satellite images, using semantic concepts defined by user. This task combines a step supervised classification patches image and integration spatial information between these patches. Given training set images for each concept, learning is based on latent Dirichlet allocation (LDA) model. hierarchical model represents item collection as random mixture topics, where topic characterized distribution over words. The LDA-based...

In this paper, we propose a fast and efficient multitemporal despeckling method. The key idea of the proposed approach is use ratio image, provided by between an image temporal mean stack. This easier to denoise than single thanks its improved stationarity. Besides, temporally stable thin structures are well preserved mean. can be divided into three steps: 1) estimation “superimage” averaging possibly spatial denoising; 2) denoising noisy interest “superimage”; 3) computation denoised...

Astronomical images obtained with adaptive optics systems can be enhanced by using image restoration techniques. However, this usually requires an accurate knowledge of the system point-spread function (PSF) which is variable in time. We present a method to estimate PSF related each image, data from control computer, namely, wave-front sensor measurements and commands deformable mirror, accumulated synchronization acquisition. This no extra observing time has been successfully tested on...

We propose a method to locate three vanishing points on an image, corresponding orthogonal directions of the scene. This is based two cascaded Hough transforms. show that, even in case synthetic images high quality, naive approach may fail, essentially because errors due limitation image size. take into account these as well detection inaccuracy segments, and provide efficient, real complex scenes.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML"...

In multipass spaceborne synthetic aperture radar (SAR) interferometry, the two acquisitions often present low correlation levels and very noisy phase measurements that are incompatible with automatic unwrapping. Instead of dealing many residues due to erroneous-wrapped differences, authors propose use local frequency as measured by a spectral analysis algorithm presented in previous paper, E. Trouve et al. (1996). For this purpose, conventional unwrapping algorithms, one other global, which...

The control of the integrity and authentication medical images is becoming ever more important within Medical Information Systems (MIS). intra- interhospital exchange images, such as in PACS (Picture Archiving Communication Systems), ease copying, manipulation distribution have brought forth security aspects. In this paper we focus on role watermarking for MIS address problem images. We discuss alternative schemes to extract verification signatures compare their tamper detection performance.

The authors propose a method aiming to characterize the spatial organization of main cartographic elements synthetic aperture radar (SAR) image and thus giving an almost automatic interpretation scene. Their approach is divided into three steps which build whole gradually. first step consists applying low-level detectors taking speckle statistics account extracting some raw information from detector responses are then fused in second using Dempster-Shafer theory, allowing modeling knowledge...

A three-dimensional (3D) reconstruction of the vessel lumen from two angiographic views, based on a series cross-sections, is proposed. Assuming uniform mixing contrast medium and background subtraction, cross-section each reconstructed through binary representation. priori information about both slice to be relationships between adjacent slices are incorporated lessen ambiguities reconstruction. Taking into account knowledge normal geometry, an initial solution created using elliptic...

The retrieval of 3-D surface models the Earth is a major issue remote sensing. Some nice results have already been obtained at medium resolution with optical and radar imaging sensors. For instance, missions such as Shuttle Radar Topography Mission (SRTM) or SPOT HRS provided accurate digital terrain models. computation model (DSM) over urban areas new challenging issue. Since recent improvements in image resolution, synthetic aperture (SAR) interferometry, which had proved its efficiency...

A new algorithm to estimate the two-dimensional local frequencies of phase interferometric data is described. With a complex sine-wave model, demonstration given that conventional multiple-signal classification (MUSIC) can be used in spite multiplicative noise perturbations. faster dedicated processing interferograms developed and measure confidence proposed. We studied numerical performances using synthetic fringes. As result frequency estimation, knowledge fringe width orientation applied...

Exact predictions of the performances Hough detection straight lines in two-dimensional images are presented for rectangular and circular retinas, Cartesian normal parameterization, case noisy signals. Detection circles, under same assumptions, is discussed. The limits adaptive quantization to reduce intrinsic noise presented, it shown that a signal processing approach especially convenient measure detector based on transform.

We propose a method for the reconstruction of spectral reflectance function every pixel fine art painting, from series acquisitions made through commercially available chromatic filters. rigorously determine minimal number necessary filters, as well their choice in an set. give exact formulae functions, most adequate way to display painting on monitor known colorimetric parameters.

Tropospheric inhomogeneities can form a major error source in differential synthetic aperture radar interferometry measurements, which are used slow-deformation monitoring. Indeed, variations of atmospheric conditions between two acquisitions produce the signal path images and, thus, additional fringes on interferograms. These effects have strong influence interferograms and must be compensated to obtain reliable deformation measurements. This paper presents methodological approach reduce at...

The authors propose the combination of photometric and stereometric information to solve stereo vision problem in case a man-made environment. A method introduce geometrical models process order improve accuracy depth measurement extend map points where no measurements have been made is presented. This based on parameterization object surfaces relies systematic comparison result with (or gray-level) image. proposed approach improves its density by introducing hypothesis surfaces. Two kinds...