Luigi Cinque

Color-based image retrieval using spatial-chromatic histograms

Abstract The paper describes a new indexing methodology for image databases integrating color and spatial information for content-based image retrieval. This methodology, called Spatial-Chromatic Histogram (SCH), synthesizing in few values information about the location of pixels having the same color and their arrangement within the image, can be more satisfactory than standard techniques when the user would like to retrieve from the database the images that actually resemble the query image selected in their color distribution characteristics. Experimental trials on a database of about 3000 images are reported and compared with more standard techniques, like Color Coherence Vectors, on the basis of human perceptual judgments.

A clustering fuzzy approach for image segmentation

Abstract Segmentation is a fundamental step in image description or classification. In recent years, several computational models have been used to implement segmentation methods but without establishing a single analytic solution. However, the intrinsic properties of neural networks make them an interesting approach, despite some measure of inefficiency. This paper presents a clustering approach for image segmentation based on a modified fuzzy approach for image segmentation (ART) model. The goal of the proposed approach is to find a simple model able to instance a prototype for each cluster avoiding complex post-processing phases. Results and comparisons with other similar models presented in the literature (like self-organizing maps and original fuzzy ART) are also discussed. Qualitative and quantitative evaluations confirm the validity of the approach proposed.

Tuning range image segmentation by genetic algorithm

Several range image segmentation algorithms have been proposed, each one to be tuned by a number of parameters in order to provide accurate results on a given class of images. Segmentation parameters are generally affected by the type of surfaces (e.g., planar versus curved) and the nature of the acquisition system (e.g., laser range finders or structured light scanners). It is impossible to answer the question, which is the best set of parameters given a range image within a class and a range segmentation algorithm? Systems proposing such a parameter optimization are often based either on careful selection or on solution space-partitioning methods. Their main drawback is that they have to limit their search to a subset of the solution space to provide an answer in acceptable time. In order to provide a different automated method to search a larger solution space, and possibly to answer more effectively the above question, we propose a tuning system based on genetic algorithms. A complete set of tests was performed over a range of different images and with different segmentation algorithms. Our system provided a particularly high degree of effectiveness in terms of segmentation quality and search time.

Shape description using cubic polynomial Bezier curves

Abstract In this paper we present a new method for shape description consisting of an approximation of a shape by a variable number of Bezier curve segments. In our method we can control the accuracy of the Bezier approximation by a parameter thus controlling the complexity and resolution of the approximation process. The technique described here is suited to a variety of shape-based image retrieval applications and matching processes.

A multiresolution approach for page segmentation

In this work we propose a new page segmentation method for recognizing text and graphics based on a multiresolution representation of the page image. Our approach is based on the analysis of a set of feature maps available at different resolution levels. The final output is a description of the physical structure of a page. A page image is broken down into several blocks which represent components of a page, such as text, line-drawings, and pictures. The result, which uses only a small amount of memory in addition to that for the image, may be the first step for a more detailed analysis such as optical character recognition.

2-D object recognition by multiscale tree matching

Abstract In this paper we present an efficient 2D object recognition method that uses multiscale tree representations. A planar object is represented by means of a tree, in which each node corresponds to a boundary segment at some level of resolution and an arc connects nodes corresponding to segments at successive levels that are spatially related. The problem of matching an object against a model is formulated as the one of determining the best mapping between nodes at all levels of the two associated trees. The proposed matching algorithm is based on dynamic programming and has optimal O(∣T ∣∣T′∣) time complexity, where ∣ T ∣ and ∣ T ′∣ are the number of nodes in the two trees.

A Statistical Method for People Counting in Crowded Environments

In this paper we present the results of a two-years research project on automatic people counting in public crowded environments. The aim of the proposed system is to estimate the number of people passing through a gate in a public area such as a metro or a railway station. The problem is particularly challenging due to both the presence of crowd which makes it difficult the use of previous systems based on detection of isolated passengers and to the high level of statistic accuracy requested by traffic monitoring applications (error rate less then 5%).

Improvements to image magnification

The main limitation of state-of-the-art magnifying techniques is that they do not introduce any new information to the original image. This lack of information, more precisely the absence of high spatial frequency components is responsible for the perceptible degradation of the enlarged image. The idea underlying this work is to estimate the phases and frequencies of absent waveforms of absent frequencies from the original low resolution image and then to synthesize them in the high resolution image. The developed technique takes advantage of sub-pixel edge estimations from the low resolution image to direct the subsequent polynomial interpolation step. To improve perceptible image quality, the n-degree polynomial interpolating curve is user-controllable allowing both sharp and smooth edges to be synthesized. The proposed approach has been applied both to gray-level images and to color images.

Shape description and recognition by a multiresolution approach

This paper presents a multiresolution approach that uses a diffusion process to describe the shape of a 2D object. As a result, shape recognition can be achieved: shape contours may be recognized independently from orientation or size. The method proposed relies on the concept of a structural coding of an object at varying levels of resolution. A tree structure represents the evolution of the contour at increasing levels of detail, where each tree node represents a contour segment via a set of attributes to provide a richer description of the image shape. The shape recognition is based on a matching of the attributed tree representation of the candidates with that of the model.

Segmentation of page images having artifacts of photocopying and scanning

Abstract The analysis of scanned documents is important in the construction of digital libraries and paperless offices. One significant challenge is coping with artifacts of photocopying and scanning. We present a series of simple techniques for handling these difficulties. Using 125 images of the University of Washington scanned documents database, we demonstrate the effectiveness of these methods in preparing the images for segmentation by a multiresolution algorithm.