Angelo Cardellicchio

An Human Perceptive Model for Person Re-identification

Person re-identification has increasingly become an interesting task in the computer vision field, especially after the well known terroristic attacks on the World Trade Center in 2001. Even if video surveillance systems exist since the early 1950s, the third generation of such systems is a relatively modern topic and refers to systems formed by multiple fixed or mobile cameras geographically referenced or not whose information have to be handled and processed by an intelligent system. In the last decade, researchers are focusing their attention on the person re-identification task because computers (and so video surveillance systems) can handle a huge amount of data reducing the time complexity of the algorithms. Moreover, some well known image processing techniques i.e. background subtraction can be embedded directly on cameras, giving modularity and flexibility to the whole system. The aim of this work is to present an appearance-based method for person re-identification that models the chromatic relationship between both different frames and different areas of the same frame. This approach has been tested against two public benchmark datasets (ViPER and ETHZ) and the experiments demonstrate that the person re-identification processing by means of intra frame relationships is robust and shows great results in terms of recognition percentage.

Evaluation of smartphone usage in neurological pathologies diagnosis

Population ageing, i.e. the increase of the median age in the population, has become a pressing matter, due to a significant rise in health-care expenses, which, if not controlled, may destabilize welfare mechanisms. To face this problem, various approaches have already been presented in literature: some of these methods acquire pathology-relevant data using ad-hoc systems (i.e. sensor arrays), while others employs commercial devices, like wearable or smartphones; however, there are several issues that such systems must address, i.e. the meaningfulness of sensed data, or the wellness of the patients in the follow-up phase. The aim of this work is to propose a medical decision support system, specifically oriented to the acquisition of data related to neurological pathologies, whose final aim is to build an extended medical knowledge base, which will be used by skilled professionals in the diagnosis phase of the pathology; to improve the comfort of the patients, a smartphone medical app has been developed and tested, therefore evaluating the relevance of the data acquired by sensors embedded in an Android smartphone, and assessing the overall feasibility of such a decision support system.

IoT and CPS applications based on wearable devices. A case study: Monitoring of elderly and infirm patients

After briefly discussing the loT and Cyber physical device main features, the application of wearable devices was examined. The constrains of this class of application are in the greatest interest of the researchers and enterprises for the interested in many possible uses of it. The Apulia Region has issued a call for proposals for “Innonetwork sostegno alle attività di R&S per lo sviluppo di nuove tecnologie sostenibili, di nuovi prodotti e servizi” by requesting innovative projects in the field of medical monitoring. The paper focuses on the defining of some guidelines for a possible application of a wearable devices, with sensor and internet connections suitable for monitoring of elderly and infirm peoples also in the remote follow-up phase. The issue of monitoring long-term care patients introduces an array of technical challenges in designing systems and algorithms.

Identification of “Die Hard” Nodes in Complex Networks: A Resilience Approach

The topology of a network defines the structure on which physical processes dynamically evolve. Even though the topological analysis of these networks has revealed important properties about their organization, the components of real complex networks can exhibit other significant characteristics. In this work we focus in particular on the distribution of the weights associated to the links. Here, a novel metric is proposed to quantify the importance of both nodes and links in weighted scale-free networks in relation to their resilience. The resilience index takes into account the complete connectivity patterns of each node with all the other nodes in the network and is not correlated with other centrality metrics in heterogeneous weight distributions.

An opportunistic sensor network approach to wide area environmental sensing

As the availability of frequency spectrum is constantly being reduced by the growing diffusion of portable applications, the need for high speed and low power wireless transmissions of large amounts of data has become a challenging topic, especially in scenarios like geographically spread environmental sensor networks. To deal with this, opportunistic transmission approaches, i.e. methods which leverage on spectrum sensing techniques to transfer data without the need for pre-assigned frequency slots, can be exploited; however, these methods often require computationally expensive spectrum estimation techniques, which may be unsuitable for low power devices. In this work, a spectrum sensing method which uses a simple and efficient chirp-based demodulation approach is presented: this technique allows preservation of both time and frequency resolution in spectrum estimation, while reducing computational load on sensor nodes, therefore making the deployment of an opportunistic environmental sensor network feasible.

Comparative Analysis of PRID Algorithms Based on Results Ambiguity Evaluation

The re-identification of a subject among different cameras (namely Person Re-Identification or PRID) is a task that implicitly defines ambiguities. Two individuals dressed in a similar manner or with a comparable body shape are likely to be misclassified by a computer vision system, especially when only poor quality images are available (i.e. the case of many surveillance systems). For this reason we introduce a method to find, exploit and classify ambiguities among the results of PRID algorithms. This approach is useful to analyze the results of a classical PRID pipeline on a specific dataset evaluating its effectiveness in re-identification terms with respect to the ambiguity rate (AR) value. Cumulative Matching Characteristic curves (CMC) can be consequently split according to the AR, using the proposed method to evaluate the performance of an algorithm in low, medium or high ambiguity cases. Experiments on state-of-art algorithms demonstrate that ambiguity-wise separation of results is an helpful tool in order to better understand the effective behaviour of a PRID approach.

Real time Artificial Auditory Systems for cluttered environments

In this paper, a three dimensional sound steering system which exploits pointing error minimization techniques is presented. The system uses a three-microphones array located on the vertexes of an equilateral triangle, which capture audio signals generated from a sound source to control a stepper motor based steering unit. The system works in reverberant and noisy environments, and is able to estimate the Direction Of Arrival (DOA) of an audio source located in a semispace delimited by the plane of the array. The DOA of the audio source is estimated in a two-step process: the first step determines the Time Delay Of Arrival (TDOA) for each pair of microphones, the second step uses the TDOAs previously estimated to compute the DOA. The system works in real time and can achieve state-of-the-art performances using only three microphones, representing a significant improvement with respect to the existing approaches.