8 results
Search Results
Now showing 1 - 8 of 8
Item A Survey of Digital Mosaic Techniques(The Eurographics Association, 2006) Battiato, S.; Blasi, G. Di; Farinella, G.M.; Gallo, G.; S. Battiato and G. Gallo and F. StancoArt often provides valuable hints for technological innovations especially in the field of Image Processing and Computer Graphics. In this paper we survey in an unified framework several methods to transform a raster input image into good quality mosaics. The common and different ideas among the methods are reported and described. Finally we try to address the evaluation problem of an NPR algorithm.Item Ramses: a Visual Steganographic System(The Eurographics Association, 2007) Battiato, S.; Blasi, G. Di; Gallo, G.; Patti, S.; Raffaele De Amicis and Giuseppe ContiSteganography is the art of "secret communication". Its goal is to transmit a message (information) hidden inside another visible message. The typical visible message used in many steganographic systems is a digital image and the embedded message is usually hidden by working in the Fourier domain. In this paper we present Ramses, a novel approach to the steganography based on the Puzzle Image Mosaic (PIM) technique ( [DGP05]) which uses the metaphor of the hieroglyphicwriting to hide the message in the image. Themessage is first coded by a sequence of small irregular images and then merged inside another image together with many other small images.We prove that the Kerckhoff's principle required by a steganographic technique is satisfied and experimental results show how it is very difficult to detect the hidden message.Item Compressed SVG Representation of Raster Images Vectorized by DDT Triangulation(The Eurographics Association, 2006) Battiato, S.; Greco, F.; Nicotra, S.; S. Battiato and G. Gallo and F. StancoThis paper presents a portable compression algorithm applied to raster data images properly triangulated by using DDT (Data Dependent Triangulation). In particular the input source data are encoded to be rendered by standard SVG (Scalable Vector Graphics) engine. The proposed compression strategy works reducing the overall entropy implementing some heuristic strategies to properly re-code the redundancy inside the mesh representation. The compressed data are enclosed into an SVG player able to both decompress and show the original image. Results show the effectiveness of the approach.Item 3D CT Segmentation for Clinical Evaluation of Knee Prosthesis Operations(The Eurographics Association, 2006) Battiato, S.; Bosco, C.; Farinella, G. M.; Impoco, G.; S. Battiato and G. Gallo and F. StancoIn surgical knee replacement, the damaged knee joint is replaced with artificial prostheses. An accurate clinical evaluation must be carried out before applying knee prostheses to ensure optimal outcome from surgical operations and to reduce the probability of having long-term problems. Useful information can be inferred from CT scans of the knee area, that can be used to estimate the stress that will be acting on the prosthesis being implanted. This information can be exploited to tailor the prothesis to the patient s anatomy. We present some ideas for a system intended to help surgeons in evaluating the post-operative outcome of knee prosthesis implantation. We also show some preliminary results about the 3D segmentation of CT data, that is the main activity of our lab in the project.Item Color Palette Images Re-indexing by Self Organizing Motor Maps(The Eurographics Association, 2006) Battiato, S.; Rundo, F.; Stanco, F.; S. Battiato and G. Gallo and F. StancoPalette re-ordering is a well known and very effective approach for improving the compression of color-indexed images. If the spatial distribution of the indexes in the image is smooth, greater compression ratios may be obtained. As known, obtaining an optimal re-indexing scheme is not a simple problem. In this paper we provide a novel algorithm for palette re-ordering problem showing the advantages of using a neural network instead of classical heuristic methods. We propose to apply the Motor Map neural network which is considered an extension of the well-known SOM Kohonen neural network. Experiments confirm the effectiveness of the proposed technique.Item Modelling and Animation of Theatrical Greek Masks in an Authoring System(The Eurographics Association, 2007) Bertacchini, P. A.; Bilotta, E.; Pantano, P.; Battiato, S.; Cronin, M.; Blasi, G. Di; Talarico, A.; Tavernise, A.; Raffaele De Amicis and Giuseppe ContiThis paper describes an authoring system for modelling Greek virtual masks that express emotions for synchronising their facial movements with pre-recorded speech files, and for creating theatrical performances. The system is related to parametric modelling and includes three interfaces: Editor, Recorder and Virtual Theatre. In the Editor interface it is possible to create several different 3D masks from a unique mask basic model. In the Recorder it is possible to import the models created in the Editor and select eight expressions (neutral, anger, surprise, sadness, fear, joy, disgust, attention), as well as to create and save little alterations in these expressions, and to synchronise the facial movements with speech on the time-line of the system. In the Virtual Theatre the animated Greek masks can be imported and can perform.Item SVG Vectorization by Statistical Region Merging(The Eurographics Association, 2006) Battiato, S.; Farinella, G. M.; Puglisi, G.; S. Battiato and G. Gallo and F. StancoIn this paper a novel algorithm for raster to vector conversion is presented. The technique is mainly devoted to vectorize digital picture maintaining an high degree of photorealistic appearance specifically addressed to the human visual system. The algorithm makes use of an advanced segmentation strategy based on statistical region analysis together with a few ad-hoc heuristics devoted to track boundaries of segmented regions. The final output is rendered by Standard Vector Graphics. Experimental results confirm the effectiveness of the proposed approach both in terms of perceived and measured quality. Moreover, the final overall size of the vectorized images outperforms existing methods.Item A New Descreening Technique in the Frequency Domain(The Eurographics Association, 2006) Battiato, S.; Stanco, F.; S. Battiato and G. Gallo and F. StancoIn this paper a new algorithm to obtain a continuous tone image starting from a halftoned one is proposed. This descreening technique is based on Butterworth filtering in the frequency domain. It removes the pattern of the original screen leaving unchanged the colors in the image. The proposed algorithm ensures fast and effective results, and can be used also by non-qualified operators.