IRIS Univ. Trentohttps://iris.unitn.itIl sistema di repository digitale IRIS acquisisce, archivia, indicizza, conserva e rende accessibili prodotti digitali della ricerca.Thu, 02 Dec 2021 04:34:59 GMT2021-12-02T04:34:59Z10181Roof planes detection via a second-order variational modelhttp://hdl.handle.net/11572/200472.14Titolo: Roof planes detection via a second-order variational model
Abstract: The paper describes a unified automatic procedure for the detection of roof planes in gridded height data. The procedure exploits the Blake-Zisserman (BZ) model for segmentation in both 2D and 1D, and aims to detect, to model and to label roof planes. The BZ model relies on the minimization of a functional that depends on first- and second-order derivatives, free discontinuities and free gradient discontinuities. During the minimization, the relative strength of each competitor is controlled by a set of weight parameters. By finding the minimum of the approximated BZ functional, one obtains: (1) an approximation of the data that is smoothed solely within regions of homogeneous gradient, and (2) an explicit detection of the discontinuities and gradient discontinuities of the approximation. Firstly, input data is segmented using the 2D BZ. The maps of data and gradient discontinuities are used to isolate building candidates and planar patches (i.e. regions with homogeneous gradient) that correspond to roof planes. Connected regions that can not be considered as buildings are filtered according to both patch dimension and distribution of the directions of the normals to the boundary. The 1D BZ model is applied to the curvilinear coordinates of boundary points of building candidates in order to reduce the effect of data granularity when the normals are evaluated. In particular, corners are preserved and can be detected by means of gradient discontinuity. Lastly, a total least squares model is applied to estimate the parameters of the plane that best fits the points of each planar patch (orthogonal regression with planar model). Refinement of planar patches is performed by assigning those points that are close to the boundaries to the planar patch for which a given proximity measure assumes the smallest value. The proximity measure is defined to account for the variance of a fitting plane and a weighted distance of a point from the plane. The effectiveness of the proposed procedure is demonstrated by means of its application to urban digital surface models characterized by different spatial resolutions. Results are presented and discussed along with some promising developments.
Mon, 01 Jan 2018 00:00:00 GMThttp://hdl.handle.net/11572/200472.142018-01-01T00:00:00ZThe Blake-Zisserman model for digital surface models segmentationhttp://hdl.handle.net/11572/66286Titolo: The Blake-Zisserman model for digital surface models segmentation
Tue, 01 Jan 2013 00:00:00 GMThttp://hdl.handle.net/11572/662862013-01-01T00:00:00ZSequential Spectral Change Vector Analysis for Iteratively Discovering and Detecting Multiple Changes in Hyperspectral Imageshttp://hdl.handle.net/11572/101286Titolo: Sequential Spectral Change Vector Analysis for Iteratively Discovering and Detecting Multiple Changes in Hyperspectral Images
Thu, 01 Jan 2015 00:00:00 GMThttp://hdl.handle.net/11572/1012862015-01-01T00:00:00ZPiecewise linear approximation of vector-valued images and curves via second-order variational modelhttp://hdl.handle.net/11572/180695Titolo: Piecewise linear approximation of vector-valued images and curves via second-order variational model
Abstract: Variational models are known to work well for addressing image restoration/regularization problems. However, most of the methods proposed in literature are defined for scalar inputs and are used on multiband images (such as RGB or multispectral imagery) by the composition of a simple band-wise processing. This involves suboptimal results and may introduce artifacts. Only in a few cases variational models are extended to the case of vector-valued inputs. However, the known implementations are restricted to 1st-order models, while 2nd-order models are never considered. Thus, typical problems of 1st-order models such as the staircasing effect cannot be overtaken. This paper considers a 2nd-order functional model to function approximation with free discontinuities given by Blake-Zisserman (BZ) and proposes an efficient minimization algorithm in the case of vector-valued inputs. In the BZ model, the Hessian of the solution is penalized outside a set of finite length, therefore the solution is forced to be piecewise linear. Moreover, the model allows the formation of free discontinuities and free gradient discontinuities. The proposed algorithm is applied to difficult color image restoration/regularization problems and to piecewise linear approximation of curves in space.
Sun, 01 Jan 2017 00:00:00 GMThttp://hdl.handle.net/11572/1806952017-01-01T00:00:00ZEdge-crease detection and surface reconstruction from point clouds using a second-order variational modelhttp://hdl.handle.net/11572/97876Titolo: Edge-crease detection and surface reconstruction from point clouds using a second-order variational model
Wed, 01 Jan 2014 00:00:00 GMThttp://hdl.handle.net/11572/978762014-01-01T00:00:00ZRayleigh-Rice mixture parameter estimation via EM algorithm for change detection in multispectral imageshttp://hdl.handle.net/11572/111254Titolo: Rayleigh-Rice mixture parameter estimation via EM algorithm for change detection in multispectral images
Abstract: The problem of estimating the parameters of a Rayleigh-Rice mixture density is often encountered in image analysis (e.g., remote sensing and medical image processing). In this paper we address this general problem in the framework of change detection (CD) in multitemporal and multispectral images. One widely used approach to change detection in multispectral images is based on Change Vector Analysis (CVA). Here, the distribution of the magnitude of the difference image can be theoretically modeled by a Rayleigh-Rice mixture density. However, given the complexity of this model, in applications a Gaussian-mixture approximation is often considered, which may affect the change detection results. In this paper we present a novel technique for parameter estimation of the Rayleigh-Rice density that is based on a specific definition of the Expectation-Maximization (EM) algorithm. The proposed technique, which is characterized by good theoretical properties, iteratively updates the parameters and does not depend on specific optimization routines. Several numerical experiments on synthetic data demonstrate the effectiveness of the method which is general and can be applied to any image processing problem involving the Rayleigh-Rice mixture density. In the change detection context, the Rayleigh-Rice model (which is theoretically derived) outperforms other empirical models. Experiments on real multitemporal and multispectral remote sensing images confirm the validity of the model by returning significantly higher change detection accuracies than those obtained by using state-of-the-art approaches.
Thu, 01 Jan 2015 00:00:00 GMThttp://hdl.handle.net/11572/1112542015-01-01T00:00:00ZBig Data from Space for Precision Agriculture Applicationshttp://hdl.handle.net/11572/254862Titolo: Big Data from Space for Precision Agriculture Applications
Wed, 01 Jan 2020 00:00:00 GMThttp://hdl.handle.net/11572/2548622020-01-01T00:00:00ZA parallel approach for image segmentation by numerical minimization of a second-order functionalhttp://hdl.handle.net/11572/154409Titolo: A parallel approach for image segmentation by numerical minimization of a second-order functional
Fri, 01 Jan 2016 00:00:00 GMThttp://hdl.handle.net/11572/1544092016-01-01T00:00:00ZA Theoretical Framework for Change Detection Based on a Compound Multiclass Statistical Model of the Difference Imagehttp://hdl.handle.net/11572/195216Titolo: A Theoretical Framework for Change Detection Based on a Compound Multiclass Statistical Model of the Difference Image
Abstract: The change detection (CD) problem is very important in the remote sensing domain. The advent of a new generation of multispectral (MS) sensors has given rise to new challenges in the development of automatic CD techniques. In particular, typical approaches to CD are not able to well model and properly exploit the increased radiometric resolution characterizing new data as this results in a higher sensitivity to the number of natural classes that can be statistically modeled in the images. In this paper, we introduce a theoretical framework for the description of the statistical distribution of the difference image as a compound model where each class is determined by temporally correlated class transitions in the bitemporal images. The potential of the proposed framework is demonstrated on the very common problem of binary CD based on setting a threshold on the magnitude of the difference image. Here, under some simplifying assumptions, a multiclass distribution of the magnitude feature is derived and an unsupervised method based on the expectation–maximization algorithm and Bayes decision is proposed. Its effectiveness is demonstrated on a large variety of data sets from different MS sensors. In particular, experimental tests confirm that: 1) the fitting of the magnitude distribution significantly improves if compared with already existing models and 2) the overall CD error is close to the optimal value.
Mon, 01 Jan 2018 00:00:00 GMThttp://hdl.handle.net/11572/1952162018-01-01T00:00:00ZSerial and parallel approaches for image segmentation by numerical minimization of a second-order functionalhttp://hdl.handle.net/11572/195212Titolo: Serial and parallel approaches for image segmentation by numerical minimization of a second-order functional
Abstract: Because of its attractive features, second order segmentation has shown to be a promising tool in remote sensing. A known drawback about its implementation is computational complexity, above all for large set of data. Recently in Zanetti et al. [1], an efficient version of the block-coordinate descent algorithm (BCDA) has been proposed for the minimization of a second order elliptic approximation of the Blake–Zissermann functional. Although the parallelization of linear algebra operations is expected to increase the performance of BCDA when addressing the segmentation of large-size gridded data (e.g., full-scene images or Digital Surface Models (DSMs)), numerical evidence shows that this is not sufficient to get significant reduction of computational time. Therefore a novel approach is proposed which exploits a decomposition technique of the image domain into tiles. The solution can be computed by applying BCDA on each tile in parallel way and combining the partial results corresponding to the different blocks of variables through a proper interconnection rule. We prove that this parallel method (OPARBCDA) generates a sequence of iterates which converges to a critical point of the functional on the level set devised by the starting point. Furthermore, we show that the parallel method can be efficiently implemented even in a commodity multicore CPU. Numerical results are provided to evaluate the efficiency of the parallel scheme on large images in terms of computational cost and its effectiveness with respect to the behavior on the tile junctions.
Mon, 01 Jan 2018 00:00:00 GMThttp://hdl.handle.net/11572/1952122018-01-01T00:00:00Z