Giuseppe Masi

Pansharpening by Convolutional Neural Networks

A new pansharpening method is proposed, based on convolutional neural networks. We adapt a simple and effective three-layer architecture recently proposed for super-resolution to the pansharpening problem. Moreover, to improve performance without increasing complexity, we augment the input by including several maps of nonlinear radiometric indices typical of remote sensing. Experiments on three representative datasets show the proposed method to provide very promising results, largely competitive with the current state of the art in terms of both full-reference and no-reference metrics, and also at a visual inspection.

Marker-Controlled Watershed-Based Segmentation of Multiresolution Remote Sensing Images

A new technique for the segmentation of single- and multiresolution (MR) remote sensing images is proposed. To guarantee the preservation of details at fine scales, edge-based watershed is used, with automatically generated markers that help in limiting oversegmentation. For MR images, the panchromatic and multispectral components are processed independently, extracting both the edge maps and the morphological and spectral markers that are eventually fused at the highest resolution, thus avoiding any information loss induced by pansharpening. Numerical results on object layer extraction and simple classification tasks prove the proposed techniques to provide accurate segmentation maps, which preserve fine details and, contrary to state-of-the-art products, can single out objects equally well at very different scales.

Exploration of Multitemporal COSMO-SkyMed Data via Interactive Tree-Structured MRF Segmentation

We propose a new approach for remote sensing data exploration, based on a tight human-machine interaction. The analyst uses a number of powerful and user-friendly image classification/segmentation tools to obtain a satisfactory thematic map, based only on visual assessment and expertise. All processing tools are in the framework of the tree-structured MRF model, which allows for a flexible and spatially adaptive description of the data. We test the proposed approach for the exploration of multitemporal COSMO-SkyMed data, that we appropriately registered, calibrated, and filtered, obtaining a performance that is largely superior, in both subjective and objective terms, to that of comparable noninteractive methods.

Detection of environmental hazards through the feature-based fusion of optical and SAR data: a case study in southern Italy

The use of remote-sensing images is becoming common practice in the fight against environmental crimes. However, the challenge of exploiting the complementary information provided by radar and optical data, and by more conventional sources encoded in geographic information systems, is still open. In this work, we propose a new workflow for the detection of potentially hazardous cattle-breeding facilities, exploiting both synthetic aperture radar and optical multitemporal data together with geospatial analyses in the geographic information system environment. The data fusion is performed at a feature-based level. Experiments on data available for the area of Caserta, in southern Italy, show that the proposed technique provides very high detection capability, up to 95%, with a very low false alarm rate. A fast and easy-to-use system has been realized based on this approach, which is a useful tool in the hand of agencies engaged in the protection of territory.

A marker-controlled watershed segmentation: Edge, mark and fill

The segmentation of very high resolution (VHR) images portraying complex urban scenarios is a rather challenging problem. In particular, great attention must be devoted to preserve fine man-made details, of major interest for most user applications. For this reason, edge-based segmentation methods are likely preferable to region-based methods. The latter, in fact, e.g. [1], [2], succeed in taking into account long range interactions and hence perform typically well in terms of “global” accuracy, but exhibit a lower “local” accuracy with respect to former, [3].

SAR/multispectral image fusion for the detection of environmental hazards with a GIS

In this paper we propose a GIS-based methodology, using optical and SAR remote sensing data, together with more conventional sources, for the detection of small cattle breeding areas, potentially responsible of hazardous littering. This specific environmental problem is very relevant for the Caserta area, in southern Italy, where many small buffalo breeding farms exist which are not even known to the productive activity register, and are not easily monitored and surveyed. Experiments on a test area, with available specific ground truth, prove that the proposed systems is characterized by very large detection probability and negligible false alarm rate.

Dynamic Hierarchical Segmentation of Remote Sensing Images

Recursive tree-structured segmentation is a powerful tool to deal with the non-stationary nature of images. By fitting model parameters to each region/class under analysis one can adapt the segmentation algorithm to the local image statistics, thus improving accuracy. However, a single model/segmenter cannot fit regions with wildly different nature, and one should be allowed to select in a suitable library the tool most suited to the local statistics. In this paper, we implement this dynamic segmentation/classification paradigm, using two segmenters, based on spectral and textural properties, respectively, and defining suitable rules for switching model locally. Experiments on remote-sensing mosaics show that the multiple-model dynamic algorithm largely outperforms comparable single-model segmenters.

A Watershed-Based Segmentation Technique for Multiresolution Data

A new watershed-based technique is proposed for the segmentation of multiresolution remote-sensing images. These images are composed by a high-resolution panchromatic band and a low-resolution multispectral set. To achieve a segmentation with the high resolution of the panchromatic image and the high accuracy granted by the spectral information, the two components are processed jointly, using both spectral and morphological properties. In addition, a fully automatic marker generation procedure is introduced to reduce the oversegmentation typical of watershed methods. Experiments on WorldView-2 multiresolution images demonstrate the potential of the technique.

Dynamic segmentation for image information mining

Information mining systems typically do not carry out image segmentation because a single algorithm could never perform well on the wide variety of sources and user applications encountered in practice. On the other hand, a large number of tools have been proposed in the literature that handle specific segmentation tasks very well. Dynamic segmentation is a possible solution, where the image is split recursively, in a hierarchical fashion, and different tools are used at each step to address specific segmentation tasks. In this work, the segmentation of a high-resolution test image is used as a running example and as a proof of concept of the potential of this approach.

CNN-based pansharpening of multi-resolution remote-sensing images

We propose a convolutional neural network for the pansharpening of remote-sensing imagery. A very compact architecture is designed, which enables accurate training even with small-size datasets. Prior knowledge on the remote sensing domain is taken into account by augmenting the input with several maps of radiometric indices. Extensive experiments on images from various multiresolution sensors show the proposed CNN to outperform the current state of the art in terms of both full-reference and no-reference measures.