Papers by martino pesaresi
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2008
A procedure for the calculation of a texture-derived built-up presence index (PanTex) from textur... more A procedure for the calculation of a texture-derived built-up presence index (PanTex) from textural characteristics of panchromatic satellite data is presented. The index is based on fuzzy rule-based composition of anisotropic textural co-occurrence measures derived from the satellite data by the gray-level co-occurrence matrix (GLCM). Examples are produced how the PanTex index reduces the edge effects of the nonbuilt-up linear features and improves capacity to discriminate between built-up and nonbuilt-up areas. The accuracy and robustness of the PanTex measure against seasonal changes, multisensor, multiscene, and data degradation by wavelet-based compression and histogram stretching is discussed with some examples.
In this paper an improved procedure for the automatic recognition of built-up areas, using the so... more In this paper an improved procedure for the automatic recognition of built-up areas, using the so-called PANTEX index is presented. This index is based on analysis of image textural measures extracted using anisotropic rotation-invariant GLCM statistics. These measures may overestimate the built-up areas in case of presence of scattered vegetation having the same spatial pattern of settlements. In this paper we present a methodology able to overcome this problem. This methodology is based on an additional filtering step that pre-selects the image information to be ingested by the textural analysis phase. The test presented here uses multispectral QuickBird satellite data input at the spatial resolution of 2.4 meters. In the selected test area, with the improved procedure we estimated an overall accuracy of 88.69% in the automatic recognition of built-up areas, with an overall increase of accuracy of 20.76% respect to the basic procedure.
... Block adjustment with rational polynomial camera models. Proceeding of ASCM-ASPRS Annual Conv... more ... Block adjustment with rational polynomial camera models. Proceeding of ASCM-ASPRS Annual Conventions, Washington DC, April 2002, pp. Jacobsen K., 2005. ... Soille P., 2003. Morphological Image Analysis, Springer. Suchandt S., Einedera M., Breita H., Runge H., 2006. ...
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2008
... Remote Sens., vol. 58, pp. 239–258, 2003. [5] AA Nielsen, K. Conradsen, and JJ Simpson, “Mult... more ... Remote Sens., vol. 58, pp. 239–258, 2003. [5] AA Nielsen, K. Conradsen, and JJ Simpson, “Multivariate alter-ation detection (MAD ... 4–5, 2006, un-published. [7] M. Pesaresi and I. Kanellopoulos, “Detection of urban features using morphological based segmentation and very ...
A procedure for automatic recognition of the state of built-up structures after a conflict, using... more A procedure for automatic recognition of the state of built-up structures after a conflict, using 1-meter-resolution optical satellite imagery, is presented. The procedure is based on a priori fuzzy rules formalized using two basic information derived from satellite data: structural information extracted by calculation of the derivative of the morphological profile using the panchromatic data, and presence of vegetation extracted from multi-spectral data. The procedure is validated using a per-pixel and per-region image understanding approach.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011
IEEE Transactions on Geoscience and Remote Sensing, 2002
By concentrating on the analysis of the spatial relationships between groups of pixels, mathemati... more By concentrating on the analysis of the spatial relationships between groups of pixels, mathematical morphology provides us with an image processing strategy complementary to those based on the analysis of the spectral signature of single pixels. A wide variety of morphological transformations are available for extracting structural information in spatial data. Accordingly, a stream of successful applications in geoscience and remote sensing have been reported since the mid-1980s as highlighted in a brief survey. However, recent advances in the theory of mathematical morphology still remain largely unexplored. We show in this paper that they can enhance methodologies for the processing and analysis of earth observation data for tasks as diverse as filtering, simplification, directional segmentation and crest line extraction. We also address important issues overlooked in the past and concerning the applicability of a given morphological filter to earth observation data. In particular, we point out that self-dual or even self-complementary filters are required in many applications to produce results independent of the local contrast of the searched image structures.
A new segmentation method based on the morphological characteristic of connected components in im... more A new segmentation method based on the morphological characteristic of connected components in images is proposed. The formalisation of the morphological characteristic is based on a composition of the residuals of morphological opening and closing transforms by reconstruction. In case of multi-scale segmentation, this concept is generalised through the derivative of the morphological profile. Multi-scale segmentation is particularly well suited for complex image scenes such as aerial or fine-resolution satellite images, where very thin, enveloped and/or nested regions have to be retained. The proposed method performs well in the presence of both low radiometric contrast and relative low spatial resolution, which may produce textural and border effects and ambiguity in the object/background distinction. Examples of the proposed segmentation approach applied on satellite images are given.
In this paper we provide an efficient parallel algorithm for reconstruction from markers, and mul... more In this paper we provide an efficient parallel algorithm for reconstruction from markers, and multi-scale analysis through differential morphological profiles, which are top-hat scale spaces based on openings and closings by reconstruction. The new algorithms provide speed gain in two ways: (i) through parallelism, and (ii) through more efficient re-use of previously computed data. The best version of the algorithm provided a 17× speed-up on 24 cores, over computation of the same algorithm on a single core. Compared to the basic method of repeated reconstructions by a sequential algorithm, a speed gain of 25.1 times was obtained.
This paper explores a new method for quantitative estimation of building density and settlement e... more This paper explores a new method for quantitative estimation of building density and settlement edges based on textural measurements. The evaluation employs an ordered multi-scale, linear regression scheme in which scale corresponds to spatial resolution and is represented by a kernel. The kernel size in the experiment described, ranges from 30 to 1000 meters for a test area of approximately
Recent experience in using meter and sub-meter satellite images in support to crisis management i... more Recent experience in using meter and sub-meter satellite images in support to crisis management is discussed in this paper. Technical challenges and conceptual gaps in automatic information extraction are described from a perspective of the Joint Research Center's experience in crisis management. Some methodological propositions are given to cope with these challenges, which were revealed in experiments realized in various crisis scenarios.
IEEE Transactions on Geoscience and Remote Sensing, 2011
A new unsupervised change detection method for modeling nonlinear temporal dependences based on l... more A new unsupervised change detection method for modeling nonlinear temporal dependences based on local information is proposed. A theoretical analysis is presented, demonstrating how to derive optimal parameters for automating the method. It is then validated on both simulated data and very high resolution remote sensing imagery. The results show a clear improvement in change detection using the proposed method compared to other state-of-the-art change detection techniques.
In order to describe, to extract image information content, segmentation is a well-known approach... more In order to describe, to extract image information content, segmentation is a well-known approach to represent the information in terms of objects. Image segmentation is a common image processing technique aiming at disintegrating an image into a partition of its support. Hierarchical of fuzzy segmentation are extension of segmentation definition, in order to provide a covering of the image support with overlapping segments. In this paper, we propose a novel approach for breaking up an image into multi-scale overlapping objects. The image is decomposed by granulometry or differential morphological pyramid, resulting in a discrete scale-space representation. Then, the scale-space transform is segmented by a region based method. Projecting the obtained scalespace partition into space constitutes the disintegrated image representation, which enables a multi-scale object based image description.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012
Differential area profiles (DAPs) are point-based multiscale descriptors used in pattern analysis... more Differential area profiles (DAPs) are point-based multiscale descriptors used in pattern analysis and image segmentation. They are defined through sets of size-based connected morphological filters that constitute a joint area opening top-hat and area closing bottom-hat scale-space of the input image. The work presented in this paper explores the properties of this image decomposition through sets of area zones. An area zone defines a single plane of the DAP vector field and contains all the peak components of the input image, whose size is between the zone's attribute extrema. Area zones can be computed efficiently from hierarchical image representation structures, in a way similar to regular attribute filters. Operations on the DAP vector field can then be computed without the need for exporting it first, and an example with the leveling-like convex/concave segmentation scheme is given. This is referred to as the one-pass method and it is demonstrated on the Max-Tree structure. Its computational performance is tested and compared against conventional means for computing differential profiles, relying on iterative application of area openings and closings. Applications making use of the area zone decomposition are demonstrated in problems related to remote sensing and medical image analysis.
Very High Resolution (VHR) satellite images are products which are very useful for assessing huma... more Very High Resolution (VHR) satellite images are products which are very useful for assessing humanitarian situations following a disaster or a crisis. When disaster or conflict hits, generally several square kilometers on Earth surface are involved. Having a high resolution and an extended surface, the scenes, from which the information is extracted, become huge in the number of pixels. Built-up structures are the first visible objects from these images which give information about the situation. In this paper, an interactive image mining paradigm is presented for detecting structures of interest in gigapixel images. The proposed approach is tested for refugee tents detection in submetric panchromatic image.
IEEE Geoscience and Remote Sensing Letters, 2012
Roofless buildings are encountered in case of conflict and disasters as well as construction site... more Roofless buildings are encountered in case of conflict and disasters as well as construction sites. A methodology for characterizing and counting roofless buildings in Very High Resolution optical images is presented. Using morphological transform to extract specific image components, the proposed method spatially aggregates them in a fuzzy logic framework. The result is a map of roofless building membership. This membership map is then used with partial ground truth to estimate the total number of roofless buildings. Finally, the methods are validated with a WorldView panchromatic image, where the membership map and the counting estimate show good accuracies.
Pattern Recognition Letters, 2010
We propose a new procedure for quantitative evaluation of object detection algorithms. The proced... more We propose a new procedure for quantitative evaluation of object detection algorithms. The procedure consists of a matching stage for finding correspondences between reference and output objects, an accuracy score that is sensitive to object shapes as well as boundary and fragmentation errors, and a ranking step for final ordering of the algorithms using multiple performance indicators. The procedure is illustrated on a building detection task where the resulting rankings are consistent with the visual inspection of the detection maps.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011
The so-called PANTEX methodology for the automatic recognition of built-up areas is based on anal... more The so-called PANTEX methodology for the automatic recognition of built-up areas is based on analysis of image textural measures extracted using anisotropic rotation-invariant gray-level co-occurrence matrix (GLCM) statistics . These measures may overestimate the built-up areas in case of presence of scattered trees having the same spatial pattern of settlements. This overestimation is especially remarkable in case of bright soil background as in desert areas. In this paper we compare two options able to reduce this problem. One method is based on the subtraction of the vegetated areas from the built-up areas detected using the PANTEX index. The other method is based on the introduction of a morphological filtering step that pre-selects the image information to be ingested by the textural analysis phase. The test presented here uses multispectral Quick Bird satellite data input at the spatial resolution of 2.4 meters. In the selected test area, the application of the standard PANTEX procedure achieves the overall accuracy of 67.92%. The improvement of the procedure using the vegetation index achieves the accuracy of 70.37%, while the improvement based on morphological filtering achieves the accuracy of 88.69%, with an increase respect to the standard procedure of 2.44% and 20.76%, respectively.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011
ABSTRACT The work presented here tests an automatic procedure able to recognize the presence of b... more ABSTRACT The work presented here tests an automatic procedure able to recognize the presence of built-up areas in the satellite images with the output nominal scale of 1:50,000. The input data is a set of 54 Ikonos and Quick Bird scenes considered as representative of the variety of human settlement patterns in large cities at global level. The methodology for automatic image information extraction is based on calculation of anisotropic rotation-invariant textural grey-level co-occurrence measures, also called PANTEX methodology. The total area analyzed covers 35,000 $\hbox{km}^{2}$ . The data under test shows high variety in latitude, season, sun elevation and sun azimuth at the time of image data collection. The output of the automatic image information retrieval is evaluated by comparison with a collection of reference information visually interpreted from the same satellite data input. Two complementary evaluation strategies are presented here: i) interactive selection of one threshold level in the textural measurement and then unsupervised application of the same threshold level to all the datasets under test, and ii) per-scene optimization of the threshold based on the available reference samples. This work briefly summarizes the nature of the errors and implications for global settlement classification.
Uploads
Papers by martino pesaresi