Francesco G. B. De Natale

LINEAR ANTENNA SYNTHESIS WITH A HYBRID GENETIC ALGORITHM

An optimization problem for designing non-uniformly spaced, linear arrays is formulated and solved by means of an improved genetic algorithm (IGA) procedure. The proposed iterative method aims at array thinning and optimization of element positions and weights by minimizing the side-lobes level. Selected examples are included, which demonstrate the effectiveness and the design flexibility of the proposed method in the framework of electromagnetic synthesis of linear arrays.

Intelligent extended floating car data collection

The elaboration of data collected by vehicles moving on road network is relevant for traffic management and for private service providers, which can bundle updated traffic information with navigation services. Floating data, in its extended acceptation, contains not only time and location provided by a positioning system, but also information coming from various vehicle sensors. In this paper we describe our extended data collection system, in which vehicles are able to collect data about their local environment, namely the presence of roadworks and traffic slowdowns, by analyzing visual data taken by a looking forward camera and data from the on-board Electronic Control Unit. Upon detection of such events, a packet is set up containing time, position, vehicle data, results of on-board elaboration, one or more images of the road ahead and an estimation of the local traffic level. Otherwise, the transmitted packet containing only the minimal data, making its size adaptive to the environment surrounding the vehicle.

Quality evaluation and nonuniform compression of geometrically distorted images using the quadtree distortion map

The paper presents an analysis of the effects of lossy compression algorithms applied to images affected by geometrical distortion. It will be shown that the encoding-decoding process results in a nonhomogeneous image degradation in the geometrically corrected image, due to the different amount of information associated to each pixel. A distortion measure named quadtree distortion map (QDM) able to quantify this aspect is proposed. Furthermore, QDM is exploited to achieve adaptive compression of geometrically distorted pictures, in order to ensure a uniform quality on the final image. Tests are performed using JPEG and JPEG2000 coding standards in order to quantitatively and qualitatively assess the performance of the proposed method.

A versatile enhanced genetic algorithm for planar array design

In order to synthesize planar, sparse, and aperiodic arrays, a numerical procedure based on an enhanced genetic algorithm is proposed. The method maximizes a suitably defined singleobjective fitness function iteratively acting on the states and the weights of the elements of the array. Such a cost function is related to the shape of the desired beam pattern, to the number of active elements and to others user-defined array-pattern constraints. To preliminarily assess the effectiveness of the approach, selected numerical experiments are performed. The obtained results seem to confirm its feasibility. Moreover, given the heterogeneity of the test benchmarks, the versatility is pointed out as a key-feature of the implemented methodology.

Error concealment in video transmission over packet networks by a sketch-based approach

Abstract The transmission of coded visual information over packet networks introduces fidelity problems related to the loss of frames during transmission. In standard block-based coding, such losses result in a wrong reconstruction of long block sequences, also due to the use of predictive and variable length source coding techniques. In video transmission, artifacts are even more visible due to the temporal propagation caused by prediction and interpolation schemes. In order to reduce the impact of these errors on visual quality, appropriate concealment algorithms should be applied, aimed at minimizing the appearance of block artifacts due to transmission errors. In this paper, a new concealment technique is presented, which aims at restoring the lost visual information by using a synthetic reconstruction of the high-frequency content of the damaged blocks. The method is funded on the theory of sketch-based encoders: for each block to be interpolated, the sketch information of the available surrounding blocks is extracted and propagated to the missing area. Then, the low-pass content is easily interpolated from the sketch. The proposed method uses only the spatial correlation, and has been applied with good results in the transmission of video data over non-reliable packet networks.

Object Trajectory Analysis in Video Indexing and Retrieval Applications

The focus of this chapter is to present a survey on the most recent advances in representation and analysis of video object trajectories, with application to indexing and retrieval systems. We will review the main methodologies for the description of motion trajectories, as well as the indexing techniques and similarity metrics used in the retrieval process. Strengths and weaknesses of different solutions will be discussed through a comparative analysis, taking into account performance and implementation issues. In order to provide a deeper insight on the exploitation of these technologies in real world products, a selection of exampleswill be introduced and examined. The set of possible applications is very wide and includes (but it is not limited to) generic browsing of video databases, as well as more specific and context-dependent scenarios such as indexing and retrieval in visual surveillance, traffic monitoring, sport events analysis, video-on-demand, and video broadcasting.

A hybrid approach for retrieving diverse social images of landmarks

In this paper, we present a novel method that can produce a visual description of a landmark by choosing the most diverse pictures that best describe all the details of the queried location from community-contributed datasets. The main idea of this method is to filter out non-relevant images at a first stage and then cluster the images according to textual descriptors first, and then to visual descriptors. The extraction of images from different clusters according to a measure of users credibility, allows obtaining a reliable set of diverse and relevant images. Experimental results performed on the MediaEval 2014 “Retrieving Diverse Social Images” dataset show that the proposed approach can achieve very good performance outperforming state-of-art techniques.

Exploitation of time constraints for (sub-)event recognition

The aim of this paper is threefold: (a) to introduce a dataset for the recognition of events and sub-events in photographs taken by common users; (b) to propose event-based classification to achieve a more accurate labeling of event-related photo collections; (c) to use time clustering information to improve the sub-event recognition in an efficient Bag of Features classification approach. The dataset is organized according to event models and provides a collection of sample instances that allow the comparison of different recognition systems. On this basis, we will demonstrate how the use of time clustering together with multiple image visual features can outperform single image classification.

A mesh-interpolation scheme for very-low bitrate coding of video sequences

A very-low bitrate video codec is proposed that processes an image sequence as a two-component information source. The main difference from state-of-the-art approaches lies in the use of a 3D spatio-temporal interpolation; it is based on a non-uniform sampling where frames are processed as 3D blocks, so achieving an adaptive allocation for the information to be transmitted. Residuals are then compressed through a transform coder. Such a method allows for a concurrent and efficient exploitation of intra- and inter-frame correlations; experimental results show a good fidelity in decoded sequences, both in terms of subjective and objective quality.

RANK-ORDER FUNCTIONS FOR THE FAST DETECTION OF TEXTURE FAULTS

Simple approaches to texture discrimination based on histogram analysis are useful in real-time applications but often yield inadequate results. On the other hand, methods based on higher-order statistics (e.g., co-occurrence matrices) provide a more complete statistical characterisation but are extremely time-consuming. In this paper, methods based on first order statistical analysis are reviewed and the significance of the relevant representative features analyzed. Then, rank functions are considered and appropriate distance functions are introduced that prove to have substantial advantages over classical histogram-based approaches.