Eva Savina Malinverni

Making Visible the Invisible. Augmented Reality Visualization for 3D Reconstructions of Archaeological Sites

In archaeology, findings are completely freed from the earth which had been covering for centuries but in most cases they must be covered again in order to protect them. In this paper we present a augmented reality AR experience for the visualization of 3D models in-situ, giving the possibility to see conceived findings. The experience was carried out for archaeological purposes in Chan Chan, the Americas greatest pre-Columbian town. From 2001, our mission is operating at Chan Chan carrying on a wide action of documentation, conservation and exploitation. We propose an interesting workflow: image acquisition, 3D photogrammetric reconstruction, 3D simplification, AR visualization. Also a knowledge base applied to archaeological sites is here presented.

Pixel, object and hybrid classification comparisons

The choice of the best classification approach for thematic map generation relies on many factors, such as image resolution and minimum mapping unit. The generalized GIS-ready products derived from the results of pixel-based approaches and the availability of higher-resolution imagery have directed research towards object-based classification approaches. In this paper we present the superior performance of a hybrid methodology that combines the results of automatic segmentation with the land cover information derived from a pixel classification by means of the Winner Takes All (WTA) algorithm. Land use and land cover results obtained through this hybrid classification approach are compared with those of a One Against All (OAA) object-oriented classification approach.

A Hybrid Approach to Land Cover Classification from Multi Spectral Images

This work is part of a wider project whose general objective is to develop a methodology for the automatic classification, based on CORINE land-cover (CLC) classes, of high resolution multispectral IKONOS images. The specific objective of this paper is to describe a new methodology for producing really exploitable results from automatic classification algorithms. Input data are basically constituted by multispectral images, integrated with textural and contextual measures. The output is constituted by an image with each pixel assigned to one out of 15 classes at the second level of the CLC legend or let unclassified (somehow a better solution than a classification error), plus a stability map that helps users to separate the regions classified with high accuracy from those whose classification result should be verified before being used.

InSAR decorrelation to assess and prevent volcanic risk

Abstract SAR can be invaluable describing pre-eruption surface deformation and improving the understanding of volcanic processes. This work studies correlation of pairs of SAR images focusing on the influence of surface, climate conditions and acquisition band. Chosen L-band and C-band images (ENVISAT, ERS and ALOS) cover most of the Yellowstone caldera (USA) over a span of 4 years, sampling all the seasons. Interferograms and correlation maps are generated and studied in relation to snow depth and temperature. To isolate temporal decorrelation pairs of images with the shortest baseline are chosen. Results show good performance during winter, bad attitude towards wet snow and good coherence during summer with L-band performing better over vegetation.

Road pavement crack automatic detection by MMS images

The research topic was to test different feature extraction methods to localize road pavement cracks useful to construct a spatial database for the pavement distress monitoring. Several images were acquired by means of a line scan camera that assembled in a Mobile Mapping System (MMS) allows tracking directly the position of the images by a GPS-INS system. Following an automatic digital image processing was performed by means of several algorithms based on different approaches (edge detection and fuzzy set theory). The detected cracks were described with some parameters in relation to some shape characteristics (dimension, typology, direction), which are necessary to recognize the gravity of the road pavement conditions. The edge detection techniques tested in this research allowed identifying fatigue cracking or alligator cracking and also thin linear cracks in images with strong radiometric jumps by applying filters, gradient functions and morphological operators. The snake approach was one of them, in particular the type called Gradient Vector Flow (GVF). Another approach was based on the fuzzy theory. The advantage of this method is that the pixels, necessary to identify the cracks in road pavement, are darker than their surroundings in an image. The last stage was the pavement distress spatial database collection. The Mobile Mapping System (MMS) has allowed localizing the raster data and consequently the vector features of the detected cracks, associating into the table their attributes too. The proposed approaches allow to automatically localize and classify the kind of road pavement crack.

Smart maintenance of riverbanks using a standard data layer and Augmented Reality

Linear buffer strips (BS) along watercourses are commonly adopted to reduce run-off, accumulation of bank-top sediments and the leaking of pesticides into fresh-waters, which strongly increase water pollution. However, the monitoring of their conditions is a difficult task because they are scattered over wide rural areas. This work demonstrates the benefits of using a standard data layer and Augmented Reality (AR) in watershed control and outlines the guideline of a novel approach for the health-check of linear BS. We designed a mobile environmental monitoring system for smart maintenance of riverbanks by embedding the AR technology within a Geographical Information System (GIS). From the technological point of view, the systems architecture consists of a cloud-based service for data sharing, using a standard data layer, and of a mobile device provided with a GPS based AR engine for augmented data visualization. The proposed solution aims to ease the overall inspection process by reducing the time required to run a survey. Indeed, ordinary operational survey conditions are usually performed basing the fieldwork on just classical digitized maps. Our application proposes to enrich inspections by superimposing information on the device screen with the same point of view of the camera, providing an intuitive visualization of buffer strip location. This way, the inspection officer can quickly and dynamically access relevant information overlaying geographic features, comments and other contents in real time. The solution has been tested in fieldwork to prove at what extent this cutting-edge technology contributes to an effective monitoring over large territorial settings. The aim is to encourage officers, land managers and practitioners toward more effective monitoring and management practices. HighlightsDefinition of a standard data layer for describing riverbank maintenance.Combination of GIS and Augmented Reality for on-site mobile visualization.Methodology to provide authorities with a smart monitoring tool.Health-check of linear vegetated buffer strips protecting riverbanks.Effective decision-making chain and containment of risk.

Integrating elevation data and multispectral high-resolution images for an improved hybrid Land Use/Land Cover mapping

ABSTRACT The combination of elevation data together with multispectral high-resolution images is a new methodology for obtaining land use/land cover classification. It represents a step forward for both the accuracy and automation of LULC applications and allows users to setup thematic assignments through rules based on feature attributes and human expert interpretation of land usage. The synergy between different types of information means that LiDAR can give new hints at both the segmentation and hybrid classification steps, leading to a joint use of multispectral, spatial and elevation data. The output is a thematic map characterized by a custom-designed legend that is able to discriminate between land cover classes with similar spectral characteristics (level 3 of the CLC legend). Experimental results from a hilly farmland area with some urban structures (Musone river basin, Ancona, Italy) are used to highlight how the proposed methodology enhances land cover classification in heterogeneous environments.

Finalizing a Low-Cost Photogrammetric Workflow: From Panoramic Photos to Heritage 3D Documentation and Visualization:

The research aims to optimize a workflow of architecture documentation: starting from panoramic photos and tackling available instruments and technologies the aim is to propose an integrated, quick and low-cost solution of Virtual Architecture. The broader research background shows how to use spherical panoramic images for the architectural metric survey. The input data (oriented panoramic photos), the level of reliability and Image-based Modeling methods constitute an integrated and flexible 3D reconstruction approach: from the professional survey of cultural heritage to its communication in virtual museum. The proposed work results from the integration and implementation of different techniques (Multi-Image Spherical Photogrammetry, Structure from Motion, Dense Image Matching and Image-based Modeling) with the aim to achieve high metric accuracy and photorealistic performance. Different documentation chances are possible within the proposed workflow: from the virtual navigation of spherical panoramas to...

A Survey of Augmented, Virtual, and Mixed Reality for Cultural Heritage

A multimedia approach to the diffusion, communication, and exploitation of Cultural Heritage (CH) is a well-established trend worldwide. Several studies demonstrate that the use of new and combined media enhances how culture is experienced. The benefit is in terms of both number of people who can have access to knowledge and the quality of the diffusion of the knowledge itself. In this regard, CH uses augmented-, virtual-, and mixed-reality technologies for different purposes, including education, exhibition enhancement, exploration, reconstruction, and virtual museums. These technologies enable user-centred presentation and make cultural heritage digitally accessible, especially when physical access is constrained. A number of surveys of these emerging technologies have been conducted; however, they are either not domain specific or lack a holistic perspective in that they do not cover all the aspects of the technology. A review of these technologies from a cultural heritage perspective is therefore warranted. Accordingly, our article surveys the state-of-the-art in augmented-, virtual-, and mixed-reality systems as a whole and from a cultural heritage perspective. In addition, we identify specific application areas in digital cultural heritage and make suggestions as to which technology is most appropriate in each case. Finally, the article predicts future research directions for augmented and virtual reality, with a particular focus on interaction interfaces and explores the implications for the cultural heritage domain.

Stability maps for really exploitable automatic classification results

The paper describes a new methodology for producing really exploitable results from automatic classification algorithms. The output of these algorithms is usually constituted by an image with each region assigned to one out of n classes. If the end user, on the basis of results obtained from a control set provided with a ground truth, simply knows that classification over the whole dataset can be considered correct at, for example, 85% (s)he cannot know where correct and erroneously classified regions are really located in the whole dataset. Obviously, the result obtained can be exploited to effectively compute global indexes over the dataset, but it cannot be used as a thematic map. Thus, in addition to the assignment of a class to each region we propose an approach that provides a stability map, a binary image that separates regions (S) classified with high accuracy from those (U) whose classification result should be verified before being used. Two further benefits derive from the construction of the stability map: the control set can be used to set up a good threshold for binarizing the stability map (that is, a threshold by which all regions S are effectively correctly classified); unreliable regions U can help the end user to identify principal causes of (types of regions leading to) misclassification and corresponding (fuzzy, neural, rule based, etc.) approaches to overcome them.