Sergio Vitulano

Image Analysis and Processing – ICIAP 2005

This book constitutes the refereed proceedings of the 13th International Conference on Image Analysis and Processing, ICIAP 2005, held in Cagliari, Italy in September 2005. The 143 revised full papers presented together with 5 invited papers were carefully reviewed and selected from 217 submissions. The papers are organized in topical sections on pattern recognition for computer network security, computer vision for augmented reality and augmented environments, low and middle level processing, image segmentation, feature extraction and image analysis, graphs, shape and motion, image modelling and computer graphics, image communication, coding and security, computer architectures, technologies and tools, multimedia data bases, video processing and analysis, pattern classification and learning, stereo vision, 3D vision, medical applications, biometrics, and applications.

Image compression by B-tree triangular coding

This paper describes an algorithm for still image compression called B-tree triangular coding (BTTC). The coding scheme is based on the recursive decomposition of the image domain into right-angled triangles arranged in a binary tree. The method is attractive because of its fast encoding, O(n log n), and decoding, /spl Theta/(n), where n is the number of pixels, and because it is easy to implement and to parallelize. Experimental studies indicate that BTTC produces images of satisfactory quality from a subjective and objective point of view, One advantage of BTTC over JPEG is its shorter execution time.

Different methods to segment biomedical images

Abstract The paper describes three different techniques for the segmentation of biomedical images based on Entropy (ISE method), Fuzzy Entropy (FISE method) and the Least Square method (LHIS method). The relation between the entropy of an image and the entropy of its subdomains is explored as a uniformity predicate. Such entropy is obtained from an analysis of the image histogram associating a Gaussian distribution to the maximum frequency of grey levels. The experimental results show the comparison between the three proposed segmentation schemes, putting in evidence favourable time requirements and the subjective quality of the segmented images.

Edge detection : Local and global operators

The paper describes a technique called ISE for image segmentation using entropy. The relation between the entropy of an image domain and the entropy of its subdomains is explored as a uniformity predicate. Such entropy is obtained from the analysis of the image histogram associating a Gaussian distribution to the maximum frequency of gray levels. In order to implement the model, we have introduced a well-known technique of Problem Solving. In our model, the most important roles are played by the Evaluation Function (EF) and the Control Strategy. The EF is related to the ratio between the entropy of one region or zone of the picture and the entropy of the entire picture, while the Control Strategy determines the optimal path in the search tree (quadtree) so that the nodes in the optimal path have minimal entropy. The paper shows some comparisons between ISE and classical edge detection techniques.

Color image coding combining linear prediction and iterated function systems

Abstract This paper describes three color image coding schemes combining linear prediction and iterated function systems. Linear prediction is based on the known autoregressive model, while iterated function systems (IFS) are implemented using quadtree partitioning. After describing the coding schemes, comparisons are carried out in terms of objective results. Simulation studies indicate that the proposed coding schemes produce images of quality ranging from fairly good to good, while performing significantly better than IFS alone with respect to computing time.

Image segmentation by texture analysis

Texture segmentation is a significant and primary issue in texture analysis. It is concerned with automatically determining the boundaries between various textured regions in an image. In this paper we propose a region-based method for textured image segmentation. First, we consider the basic problem of texel identification and the determination of its shape. Further we describe the texture characterized by a texel, that is we identify the equivalence class of a certain structural description. Finally, we improve the accuracy of the boundaries provided by the segmentation algorithm using a least squares spline approximation.

A.I. Based Image Segmentation

The paper will show a possible model of the human perceptive process. With the aim to implement the model we have introduced a well know technique of Problem Solving.

The Role of Entropy: Mammogram Analysis

This paper introduces entropy as a feature for 1D signals. It proposes the ratio between signal perturbation (i.e. its part within minimum and maximum grey level) and the total signal energy as a measurement of entropy. Linear transformation of 2D signals into 1D signals is also illustrated together with the results. This paper also presents the experimentation carried out on different mammograms containing different pathologies (microcalcification and masses).A comparison between different entropy measures and ours is also illustrated in this study.

CONTEXT: a technique for image retrieval integrating CONtour and TEXTure information

Many intrinsically 2-dimensional visual signals can be effectively encoded in a 1-D form. This simpler representation is well-suited to both pattern recognition and image retrieval tasks. This paper deals with contour and texture, combined together in order to obtain an effective technique for content based image indexing. The data used for experimentally assessing CONTEXT were contours and textures from various application domains. The experiments reveal a high discriminating power which in turn yields a high perceived quality of the retrieval results.

Speeding Up Fractal Encoding of Images Using a Block Indexing Technique

This paper presents a novel block indexing technique (Brr) to speed up image fractal encoding. The technique assigns feature vectors to image blocks by establishing an analogy between gray level and mass. The experiments show that the BIT preserves bit rate and SNR values very close to exhaustive search, while providing speedups up to over 100.