169 result(s)
Page Size: 10, 20, 50
Export: bibtex, xml, json, csv
Order by:

CNR Author operator: and / or
more
Typology operator: and / or
Language operator: and / or
Date operator: and / or
more
Rights operator: and / or
2020 Report Open Access OPEN

AIMH research activities 2020
Aloia N., Amato G., Bartalesi V., Benedetti F., Bolettieri P., Carrara F., Casarosa V., Ciampi L., Concordia C., Corbara S., Esuli A., Falchi F., Gennaro C., Lagani G., Massoli F. V., Meghini C., Messina N., Metilli D., Molinari A., Moreo A., Nardi A., Pedrotti A., Pratelli N., Rabitti F., Savino P., Sebastiani F., Thanos C., Trupiano L., Vadicamo L., Vairo C.
Annual Report of the Artificial Intelligence for Media and Humanities laboratory (AIMH) research activities in 2020.

See at: ISTI Repository Open Access | CNR ExploRA Open Access


2019 Journal article Open Access OPEN

Virtual restoration and content analysis of ancient degraded manuscripts
Tonazzini A., Savino P., Salerno E., Hanif M., Debole F.
In recent years, extensive campaigns of digitization of the documental heritage conserved in libraries and archives have been performed, with the primary goal to ensure the preservation and fruition of this important part of the human cultural and historical patrimony. Besides protecting conservation, the availability of high quality digital copies has increasingly stimulated the use of image processing techniques, to perform a number of operations on documents and manuscripts, without harming the often precious and fragile originals. Among those, virtual restoration tasks are crucial, as they facilitate the traditional work of philologists and paleographers, and constitute a first step towards an automatic analysis of the written contents. Here we report our experience in this field, referring, as a case study, to the problem of removing one of the most frequent and impairing degradations affecting ancient manuscripts, i.e., the bleed-through distortion.We show that techniques of blind source separation are versatile tools to either cancel these unwanted interferences or isolate specific features for content analysis goals. Specialized algorithms, based on recto-verso models and sparse image representation, are then shown to be able to perform a fine and selective removal of the degradation, while preserving the original appearance of the manuscript.Source: International Journal of Information Science and Technology 3 (2019): 16–25.

See at: ISTI Repository Open Access | CNR ExploRA Open Access | www.innove.org Open Access


2019 Report Open Access OPEN

AIMIR 2019 Research Activities
Amato G., Bolettieri P., Carrara F., Ciampi L., Di Benedetto M., Debole F., Falchi F., Gennaro C., Lagani G., Massoli F. V., Messina N., Rabitti F., Savino P., Vadicamo L., Vairo C.
Multimedia Information Retrieval (AIMIR) research group is part of the NeMIS laboratory of the Information Science and Technologies Institute "A. Faedo" (ISTI) of the Italian National Research Council (CNR). The AIMIR group has a long experience in topics related to: Artificial Intelligence, Multimedia Information Retrieval, Computer Vision and Similarity search on a large scale. We aim at investigating the use of Artificial Intelligence and Deep Learning, for Multimedia Information Retrieval, addressing both effectiveness and efficiency. Multimedia information retrieval techniques should be able to provide users with pertinent results, fast, on huge amount of multimedia data. Application areas of our research results range from cultural heritage to smart tourism, from security to smart cities, from mobile visual search to augmented reality. This report summarize the 2019 activities of the research group.Source: AIMIR Annual Report, 2019

See at: ISTI Repository Open Access | CNR ExploRA Open Access


2019 Journal article Open Access OPEN

Bleed-through cancellation in non-rigidly misaligned recto-verso archival manuscripts based on local registration
Savino P., Tonazzini A., Bedini L.
Ancient manuscripts written on both pages of the sheet are frequently affected by ink bleeding from the reverse side. This phenomenon produces a significant degradation of both the foreground text and the general appearance of the manuscript. Effective digital image restoration techniques may require the use of the content of both document sides, thus needing their perfect alignment. Although often available, recto and verso are usually not aligned, either for rigid misalignments occurring during acquisition, or for non-rigid deformations of the sheet. In this paper, we propose a novel method to restore color recto-verso manuscript images in a piecewise manner, without the need of a preliminary, global registration of the two sides. We assume that at the local level any deformation can be approximated by a displacement and subdivide the two images into small patches of same size. For each pair of patches at the same location, their relative shift is estimated by cross-correlation, thus allowing their straightforward alignment. A bleed-through removal algorithm is then applied to the registered patches. By spanning the entire images, this procedure returns free-of-interferences versions of the images in their original acquisition layout. The experiments show that the restoration results so obtained are better than those obtained with the classical approach that first registers the whole recto-verso pair and then performs restoration. Further advantages are a much lower computational cost, the possibility to manage non-global and non-rigid deformations, and the unaltered geometry and color appearance of the two restored images.Source: International journal on document analysis and recognition (Print) 22 (2019): 163–176. doi:10.1007/s10032-019-00323-2
DOI: 10.1007/s10032-019-00323-2

See at: ISTI Repository Open Access | International Journal on Document Analysis and Recognition (IJDAR) Restricted | International Journal on Document Analysis and Recognition (IJDAR) Restricted | International Journal on Document Analysis and Recognition (IJDAR) Restricted | link.springer.com Restricted | International Journal on Document Analysis and Recognition (IJDAR) Restricted | International Journal on Document Analysis and Recognition (IJDAR) Restricted | CNR ExploRA Restricted


2019 Journal article Open Access OPEN

A data model and a cataloguing, storage and retrieval system for ancient document archives
Savino P., Tonazzini A., Debole F.
Digitalization of ancient manuscripts is becoming a common practice in many archives and libraries, mainly for preservation purposes. This opens many new opportunities for the diffusion of these precious cultural assets, since several scholars and researchers, as well as the general public, may access and use them for research purposes, for study, and for general information. This is made possible if the documents, their descriptions, and the result of all processing activities performed on them are acquired at a good quality and can be easily accessed by using simple and powerful retrieval mechanisms. Acquired manuscripts suffer of degradations that may require different types of elaborations on the digital images, to improve their visual quality and legibility, or to discover hidden text that is not visible. Natural Language Processing requires the creation of transcriptions of the text contained in the manuscript, as well as encoding of the document structure and creation of user annotations. This paper presents a document management system and a metadata schema that make possible the storage and content-based retrieval of original documents, elaborations performed to improve their readability, textual transcriptions, and linguistic annotations. The archive will offer the possibility of describing, storing and accessing all the available manuscript versions, document transcriptions and annotations, and to search and retrieve documents based on all this information.Source: International Journal of Information Science and Technology 3 (2019): 6–15.

See at: innove.org Open Access | ISTI Repository Open Access | CNR ExploRA Open Access


2018 Journal article Open Access OPEN

Sparse representation based inpainting for the restoration of document images affected by bleed-through
Hanif M., Tonazzini A., Savino P., Salerno E.
Bleed-through is a commonly encountered degradation in ancient printed documents and manuscripts, which severely impair their readability. Digital image restoration techniques can be effective to remove or significantly reduce this degradation. In bleed-through document image restoration the main issue is to identify the bleed-through pixels and replace them with appropriate values, in accordance to their surroundings. In this paper, we propose a two stage method, where a pair of properly registered images of the document recto and verso is first used to locate the bleed-through pixels in each side, and then a sparse representation based image inpainting technique is used to fill-in the bleed-through areas according to the neighbourhood, in such a way to preserve the original appearance of the document. The advantages of the proposed inpainting technique over state-of-the-art methods are illustrated by the improvement in the visual results.Source: Proceedings (MDPI) 2 (2018). doi:10.3390/proceedings2020093
DOI: 10.3390/proceedings2020093

See at: Proceedings Open Access | Proceedings Open Access | Proceedings Open Access | ISTI Repository Open Access | CNR ExploRA Open Access | Proceedings Open Access | Proceedings Open Access


2018 Journal article Open Access OPEN

Non-local sparse image inpainting for document bleed-through removal
Hanif M., Tonazzini A., Savino P., Salerno E.
Bleed-through is a frequent, pervasive degradation in ancient manuscripts, which is caused by ink seeped from the opposite side of the sheet. Bleed-through, appearing as an extra interfering text, hinders document readability and makes it difficult to decipher the information contents. Digital image restoration techniques have been successfully employed to remove or significantly reduce this distortion. This paper proposes a two-step restoration method for documents affected by bleed-through, exploiting information from the recto and verso images. First, the bleed-through pixels are identified, based on a non-stationary, linear model of the two texts overlapped in the recto-verso pair. In the second step, a dictionary learning-based sparse image inpainting technique, with non-local patch grouping, is used to reconstruct the bleed-through-contaminated image information. An overcomplete sparse dictionary is learned from the bleed-through-free image patches, which is then used to estimate a befitting fill-in for the identified bleed-through pixels. The non-local patch similarity is employed in the sparse reconstruction of each patch, to enforce the local similarity. Thanks to the intrinsic image sparsity and non-local patch similarity, the natural texture of the background is well reproduced in the bleed-through areas, and even a possible overestimation of the bleed through pixels is effectively corrected, so that the original appearance of the document is preserved. We evaluate the performance of the proposed method on the images of a popular database of ancient documents, and the results validate the performance of the proposed method compared to the state of the art.Source: JOURNAL OF IMAGING 4 (2018). doi:10.3390/jimaging4050068
DOI: 10.3390/jimaging4050068

See at: Journal of Imaging Open Access | Journal of Imaging Open Access | Journal of Imaging Open Access | ISTI Repository Open Access | CNR ExploRA Open Access | DOAJ-Articles Open Access | Journal of Imaging Open Access


2018 Conference article Open Access OPEN

Document bleed-through removal using sparse image inpainting
Hanif M., Tonazzini A., Savino P., Salerno E., Tsagkatakis G.
Bleed-through is a pervasive degradation in ancient documents, caused by the ink of the opposite side of the sheet that has seeped through the paper fiber, and appears as an extra, interfering text. Bleed-through severely impairs document readability and makes it difficult to decipher the contents. Digital image restoration techniques have been successfully employed to remove or significantly reduce this distortion. The main theme is to identify the bleedthrough pixels and estimate an appropriate replacement for them, in accordance to their surrounding. This paper proposes a two-step image restoration method, exploiting information from the recto and verso images. First, based on a non-stationary linear model of the two texts overlapped in the recto-verso pair, the bleed-through pixels are identified. In the second step, a sparse representation based image inpainting technique, with a non-negative sparsity constraint, is used to find an appropriate replacement for the bleedthough pixels. Thanks to the power of dictionary learning and sparse image reconstruction methods, the natural texture of the background is well reproduced in the bleed-through areas, and even a their possible overestimation is effectively corrected, so that the original appearance of the document is preserved. The experiments are conducted on the images of a popular database of ancient documents, and the results validate the performance of the proposed method compared to the state of the art.Source: DAS 2018 - 13th IAPR International Workshop on Document Analysis Systems, pp. 281–286, Vienna, Austria, 24-27 April 2018
DOI: 10.1109/das.2018.21

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | doi.org Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2018 Conference article Open Access OPEN

Archiving and retrieving digital elaborations of ancient manuscripts
Savino P., Tonazzini A., Debole F., Salerno E.
Digitalization of ancient manuscripts is becoming a standard in libraries and archives. In many cases, manuscripts suffer of degradations that may require performing different types of elaborations on the digital images, in order to improve their legibility and analyze their contents. Digital archives containing digital images of manuscripts and all the elaborations performed on these images are thus of primary importance for a complete exploitation of all available information regarding the manuscripts themselves. This paper presents a metadata schema suitable for the management of such an archive. The archive will offer the possibility of describing, storing and accessing all the available manuscript versions, and to search them based on their content.Source: CiST 2018 - IEEE 5th International Congress on Information Science and Technology, pp. 172–177, Marrakech, Marocco, 21-27 October 2018
DOI: 10.1109/cist.2018.8596505

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | doi.org Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2018 Conference article Open Access OPEN

A first step towards NLP from digitized manuscripts: virtual restoration
Debole F., Hanif M., Salerno E., Savino P., Tonazzini A.
Digitization of the documental heritage conserved in libraries and archives is a common practice, in order to ensure the preservation and fruition of this extended part of the human cultural and historical patrimony. For the most precious, fragile and difficult to read and decipher manuscripts, specialized though portable digitization equipment, such as high resolution multispectral/hyperspectral cameras, is nowadays available. Digitization made it possible the increasingly extensive use of digital image processing techniques, to perform a number of virtual restoration tasks, which constitute a first, often necessary step prior subsequent automatic analysis of the writing contents, with the ultimate goal to perform automatic transcription and/or natural language processing tasks. Here we report our experience in this field, referring, as a case study, to the problem of removing one of the most frequent and impairing degradation affecting many ancient manuscripts, i.e., the bleed-through distortion. In this case, virtual restoration gives also the immediate benefit to facilitate the work of philologists and paleographers interested in examining and transcribing the manuscript in a traditional way.Source: CiST 2018 - IEEE 5th International Congress on Information Science and Technology, pp. 188–193, Marrakech, Marocco, 21-27 October 2018
DOI: 10.1109/cist.2018.8596494

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2017 Journal article Open Access OPEN

Removal of achromatic reflections from a single color image
Bedini L., Savino S., Tonazzini A.
In this paper we consider the problem of removing achromatic reflections from a picture of a scene taken through a semi-transparent medium, assuming that the reflection pattern is due to a light source or another object located in front of the object of interest. While other works assume the availability of multiple observations, we consider the more challenging problem of having as data a single color image. We suppose a data model where the virtual reflected image combines additively with the real transmitted image of the object, through unknown coefficients. This highly underdetermined problem is handled by means of a blind estimation technique that exploits the strict dependence of the gradients of the three color channels of the ideal image, and their independence from the gradient of the grayscale reflected image. The model parameters are estimated through independent component analysis, and then the component images are estimated through a regularization technique. The whole algorithm is very fast, and its performance is quantitatively evaluated on numerically generated images, and qualitatively tested on real images.Source: Pattern recognition and image analysis 27 (2017): 675–685. doi:10.1134/S1054661817040034
DOI: 10.1134/s1054661817040034

See at: ISTI Repository Open Access | Pattern Recognition and Image Analysis Restricted | Pattern Recognition and Image Analysis Restricted | Pattern Recognition and Image Analysis Restricted | Pattern Recognition and Image Analysis Restricted | CNR ExploRA Restricted


2017 Conference article Open Access OPEN

Sparse representation based inpainting for the restoration of document images affected by bleed-through
Hanif M., Tonazzini A., Savino P., Salerno E.
Bleed-through is a commonly encountered degradation in ancient printed documents and manuscripts, which severely impair their readability. Digital image restoration techniques can be effective to remove or significantly reduce this degradation. In bleed-through document image restoration the main issue is to identify the bleed-through pixels and replace them with appropriate values, in accordance to their surroundings. In this paper, we propose a two stage method, where a pair of properly registered images of the document recto and verso is first used to locate the bleed-through pixels in each side, and then a sparse representation based image inpainting technique is used to fill-in the bleed-through areas according to the neighbourhood, in such a way to preserve the original appearance of the document. The advantages of the proposed inpainting technique over state-of-the-art methods are illustrated by the improvement in the visual results.Source: IWCIM 2017, Kos, Greece, 01 September 2017

See at: CNR ExploRA Open Access | www.eurasip.org Open Access


2016 Journal article Open Access OPEN

Digital restoration of ancient color manuscripts from geometrically misaligned recto-verso pairs
Savino P., Tonazzini A.
We propose a fast automatic procedure for registration and restoration of images of recto-verso pairsof color manuscripts affected by bleed-through distortion. The registration algorithm assumes a rigidprojective deformation of a side with respect to the other. The coefficients of the geometric transformationare computed from a large number of pairs of matching points, automatically detected by exploiting theestimates of local shifts between pairs of small patches. We validate the efficiency of the registrationalgorithm through the performance of a restoration method based on a model that relates each coupleof corresponding pixels in the two images, and thus requiring a very accurate alignment of the twosides. The experiments show that this combined procedure of registration plus restoration can providean excellent removal of the bleed-through pattern, while leaving unaltered the salient features of theoriginal manuscript.Source: Journal of cultural heritage 19 (2016): 511–521. doi:10.1016/j.culher.2015.11.005
DOI: 10.1016/j.culher.2015.11.005

See at: ISTI Repository Open Access | Journal of Cultural Heritage Restricted | Journal of Cultural Heritage Restricted | Journal of Cultural Heritage Restricted | CNR ExploRA Restricted | Journal of Cultural Heritage Restricted


2016 Conference article Open Access OPEN

Building a digital library containing digital elaborations of ancient documents
Debole F., Savino P., Tonazzini A.
Digital archives containing digitized images and detailed descriptions of cultural heritage objects are of primary importance in order to guarantee the preservation and to foster the fruition of many fragile artifacts of our culture and history. Digital processing of these images is frequently needed in order to improve their readability, to correct degradations and damages, and to analyze their contents. This paper presents a metadata schema and a metadata editor supporting the description and the archiving of all elaboration activities performed. The archive allows one to perform content based searches of the original object's descriptions as well as of the results of the elaboration activities.Source: Tenth International Conference on Digital Information Management, pp. 124–131, Jeju Island, South Korea, 21-23/10/2015
DOI: 10.1109/icdim.2015.7381855

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | ieeexplore.ieee.org Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2016 Conference article Open Access OPEN

Joint non-rigid registration and restoration of recto-verso ancient manuscripts
Savino P., Bedini L., Tonazzini A.
Ancient manuscripts written on both pages of the sheet are frequently affected by ink bleeding from the reverse side, which produces a significant degradation of the text. Effective digital image restoration techniques may require the use of the content of both document sides, thus needing their perfect alignment. Usually, recto and verso are not aligned either for rigid misalignments occurring during acquisition, or for non-rigid deformations of the sheet. In this paper we propose a novel method to jointly register and restore color recto-verso manuscript images in a piecewise manner, by subdividing the images into sub-images that exhibit apparent, different deformations of one with respect to the other. For each pair of corresponding sub-images, a specific projective transformation is computed, the two sub-images are registered, and then restored with a pixel-by-pixel algorithm that returns free of interferences versions of the images in their original acquisition layout. The projective transformation is estimated exploiting the precise computation of the shifts of a large number of small corresponding recto and verso patches, via correlation of their gradients. The experiments show that this combined procedure of local registration plus restoration can provide an excellent removal of bleed-through, while leaving unaltered the salient features of the original manuscripts.Source: International Workshop on Computational Intelligence for Multimedia Understanding, Reggio Calabria, Italy, 27-28 October 2016
DOI: 10.1109/iwcim.2016.7801180

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2015 Journal article Open Access OPEN

A non-stationary density model to separate overlapped texts in degraded documents
Tonazzini A., Savino P., Salerno E.
We address the problem of the removal of a text superimposed to a more important one, in a document image, considering the two instances of canceling back-to-front interferences from recto and verso images of archival documents and of recovering the erased text in palimpsests from multispectral images. Both problems are approached through a model where the ideal images of the two texts are considered as individual source patterns, mixed through some parametric operator. To cope with occlusions, ink saturation, and space variability of the mixing operator, a data model for this problem should be nonlinear and space variant. Here, we show that if a pointwise non-stationarity is allowed, a linear model can compensate for the lack of a suitable nonlinearity and for other modeling errors.Source: Signal, image and video processing (Print) 9 (2015): 155–164. doi:10.1007/s11760-014-0735-3
DOI: 10.1007/s11760-014-0735-3

See at: link.springer.com Open Access | Signal Image and Video Processing Open Access | ISTI Repository Open Access | CNR ExploRA Open Access | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted | Signal Image and Video Processing Restricted


2015 Contribution to conference Open Access OPEN

Integrating optical imaging and digital processing for nondestructive diagnosis of artifacts
Console E., Tonazzini A., Salerno E., Savino P., Bruno F.
Optical imaging is one of the less expensive and less invasive techniques to capture information from tangible cultural heritage. An additional advantage is that it is faster than many other methods, providing comprehensive maps of the quantities under study in just one or a few steps. More specialized techniques, such as chemical analysis or spectroscopy, are capable of much more accurate and reliable measurements; on the one hand, however, these pointwise techniques often need a panoramic pre-diagnostics to find the areas of interest; on the other hand, diversity imaging complemented by appropriate processing techniques is often able to extract information that can well provide for the diagnostic needs at hand. In this presentation 3D and multispectral imaging are treated, since 3D spatial and spectral data allow us to extract information on the conservation, the structure and the composition of the object studied. Moreover, details that are invisible to the naked eye can emerge without damaging the object. For example, IR light can penetrate the surfaces, thus revealing invisible details such as pentimentos, sinopias or underdrawings in paintings. Also, by IR light we can distinguish different materials, such as iron- and carbon-based inks in ancient documents, or reveal the presence of biological or non-biological contaminants. As another example, an UV-fluorescence image can discriminate between apparently similar pigments, thus helping the historical and critical study of the artifact. 3D imaging, finally, allows us to study an object through its digital model, thus determining its deformation, the presence of flaws or cracks, and other features. Many of the possibilities mentioned are now long established [1]. What is relatively new in cultural heritage conservation is the integrated use of all the available channels together, to extract information patterns that are not detectable in any of the channels if analyzed alone. To this end, we have been studying and employing a number of processing algorithms that are capable of transforming the "color" layers in the raw image to coherent "information" layers in the processed output, sometimes with the help of the associated 3D information [2]. These algorithms range from the necessary multiple spatial colocation, through simple color space manipulation, to sophisticated statistical procedures applied to both the spectral and the spatial features of the input image. We also built an integrated 3D-multispectral capture system [3]. In this presentation, we describe this system and the numerical procedures we have been using in several applications, and report the results obtained in some real case studies.Source: TECHNART 2015 - Non-destructive and microanalytical techniques in art and cultural heritage, Catania, Italy, 27-30 April 2015

See at: ISTI Repository Open Access | CNR ExploRA Open Access | technart2015.lns.infn.it Open Access


2015 Conference article Open Access OPEN

Removing achromatic reflections from color images with application to artwork imaging
Bedini L., Savino P., Tonazzini A.
We propose a general approach to remove reflections from a color image acquired through a semi-transparent medium, and show its application to the restoration of images of paintings framed behind glass and manuscripts laminated for conservation purposes, affected by the reflection of a light source. The problem is modeled by assuming that the unwanted reflection is an achromatic or monochromatic image that combines additively with the real transmitted image of the object of interest. In the absence of information about the mixing coefficients, we adopted a blind source separation technique that exploits the dependence of the three color channels of the original image, and the independence of the reflected image. In particular, these constraints are forced on the image gradients rather than on the intensity images. The algorithm is constituted of a step for the estimation, via independent component analysis of the model parameters, followed by a regularization technique to estimate the component images. The algorithm is very fast and provides promising results.Source: 9th International Symposium on Image and Signal Processing and Analysis, pp. 126–130, Zagreb, Croatia, 07-09/2015
DOI: 10.1109/ispa.2015.7306045

See at: ISTI Repository Open Access | academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | ieeexplore.ieee.org Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2014 Conference article Restricted

Non-stationary modeling for the separation of overlapped texts in documents
Tonazzini A., Savino P., Salerno E.
In this paper, we address the removal of severe back-to-front interferences in archival documents, when recto and verso images of the page are available. The problem is approached from a modeling point of view, considering the ideal images of the two separated texts as individual source patterns that overlap in the observed images through some parametric mixing operator. Earlier approaches were based on linear mixtures of the ideal reflectance maps, or of the ideal optical densities and absorptance maps, through unknown coefficients or blur kernels. Some approximations and/or partial user supervision were then adopted to jointly estimate the sources and the model parameters. Nevertheless, a feasible and reliable data model for this problem should at least be non-linear and space-variant, to cope with occlusions, ink saturation, and large variability of the mixing level. This is especially true for ancient documents affected by ink seeping (bleed-through). The search for such a model is still far from being concluded, or even impossible to pursue, due to the unavailability of information about the chemical and physical processes at the origin of the phenomenon. Hence, here, we propose the use of pixel-dependent parameters, within a model additive in the optical densities, to compensate not only for non-stationarity, but also for the lack or the imprecise knowledge of the non-linearity, and for modeling errors more in general.Source: SIU 2014 - 2014 22nd Signal Processing and Communications Applications Conference, pp. 2314–2318, Trabzon, Turkey, 23-25 April 2014
DOI: 10.1109/siu.2014.6830727

See at: academic.microsoft.com Restricted | dblp.uni-trier.de Restricted | doi.org Restricted | ieeexplore.ieee.org Restricted | ieeexplore.ieee.org Restricted | CNR ExploRA Restricted | xplorestaging.ieee.org Restricted


2014 Journal article Restricted

MI-File: using inverted files for scalable approximate similarity search
Amato G., Gennaro C., Savino P.
We propose a new efficient and accurate technique for generic approximate similarity searching, based on the use of inverted files. We represent each object of a dataset by the ordering of a number of reference objects according to their distance from the object itself. In order to compare two objects in the dataset, we compare the two corresponding orderings of the reference objects. We show that this representation enables us to use inverted files to obtain very efficiently a very small set of good candidates for the query result. The candidate set is then reordered using the original similarity function to obtain the approximate similarity search result. The proposed technique performs several orders of magnitude better than exact similarity searches, still guaranteeing high accuracy. To also demonstrate the scalability of the proposed approach, tests were executed with various dataset sizes, ranging from 200,000 to 100 million objects.Source: Multimedia tools and applications (Dordrecht. Online) 71 (2014): 1333–1362. doi:10.1007/s11042-012-1271-1
DOI: 10.1007/s11042-012-1271-1

See at: Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | Multimedia Tools and Applications Restricted | CNR ExploRA Restricted