Gian Luca Foresti

Active video-based surveillance system: the low-level image and video processing techniques needed for implementation

The importance of video surveillance techniques has considerably increased since the latest terrorist incidents. Safety and security have become critical in many public areas, and there is a specific need to enable human operators to remotely monitor the activity across large environments. For these reasons, multicamera systems are needed to provide surveillance coverage across a wide area, ensuring object visibility over a large range of depths. In the development of advanced visual-based surveillance systems, a number of key issues critical to its successful operation must be addressed. This article describes the low-level image and video processing techniques needed to implement a modern surveillance system. In particular, the change detection methods for both fixed and mobile cameras (pan and tilt) are introduced and the registration methods for multicamera systems with overlapping and nonoverlapping views are discussed.

Video Analysis in Pan-Tilt-Zoom Camera Networks

Pan-tilt-zoom (PTZ) cameras are able to dynamically modify their field of view (FOV). This functionality introduces new capabilities to camera networks such as increasing the resolution of moving targets and adapting the sensor coverage. On the other hand, PTZ functionality requires solutions to new challenges such as controlling the PTZ parameters, estimating the ego motion of the cameras, and calibrating the moving cameras.This tutorial provides an overview of the main video processing techniques and the currents trends in this active field of research. Autonomous PTZ cameras mainly aim to detect and track targets with the largest possible resolution. Autonomous PTZ operation is activated once the network detects and identifies an object as sensible target and requires accurate control of the PTZ parameters and coordination among the cameras in the network. Therefore, we present cooperative localization and tracking methods, i.e., multiagentand consensus-based approaches to jointly compute the targets properties such as ground-plane position and velocity. Stereo vision exploiting wide baselines can be used to derive three-dimensional (3-D) target localization. This tutorial further presents different techniques for controlling PTZ camera handoff, configuring the network to dynamically track targets, and optimizing the network configuration to increase coverage probability. It also discusses implementation aspects for these video processing techniques on embedded smart cameras, with a special focus on data access properties.

Grouping as a Searching Process for Minimum-Energy Configurations of Labelled Random Fields

A two-level probabilistic method for grouping edge-based descriptive primitives is proposed. At the lower level, a voting mechanism based on the Direct Hough Transform is employed in order to obtain a relational graph whose nodes correspond to a set of straight segments extracted from an edge image. The nodes are linked to each other according to geometrical relationships (i.e., parallelism, collinearity, convergence) among the detected segments. At the higher level, the grouping process consists of assigning a label to each node. Subsets of nodes with the same label are identified as consistent segment groups. A nonlinear cost function is used as a measure evaluating the goodness of every proposed configuration of labels. Such a measure is interpreted as an energy function related to the probability of a field configuration derived from the Gibbs distribution; this corresponds to modeling the label graph as a Markov Random Field (MRF). The energy function is formed by the interactions of local terms with precise geometrical significances. The label MRF is characterized by a multiple neighborhood system and, hence, by multiple cliques. A Simulated Annealing algorithm is used to find the best label configuration. The approach has been tested on a wide number of synthetic and real scenes and related results are provided to show the capabilities of the proposed approach.

Circular arc extraction by direct clustering in a 3D Hough parameter space

Abstract The Hough transform is a robust technique for analysis of straight lines in images containing noise and occlusions, but involves a considerable computational load and storage problems when it is used to recover circles, ellipses or more complex patterns. This paper presents an efficient technique for circular arc detection, called the circular direct Hough transform (CDHT), which aims to reduce the drawbacks affecting classical Hough-based approaches (i.e., low speed, loss of spatial information, and spurious-peak generation) without increasing the memory requirements. A modified parametrization is used to represent a circle by a couple of dependent equations of the first order (instead of the classical equation of the second order ( x − x 0 ) 2 + ( y − y 0 ) 2 − r 2 = 0) and a clustering phase is introduced to detect different circular arcs belonging to the same circle. Results are reported to describe and quantify the performances of the CDHT in terms of accuracy, robustness to noise, computational efficiency, and storage. Comparisons are made between the proposed method and some representative Hough-based algorithms (Yip et al., 1992; Duda and Hart, 1972), using both synthetic and real images. Circle detection in crowd images, where circular patterns are associated with human heads, is described as an application to show the robustness of the method.

A distributed approach to 3D road scene recognition

A distributed 3D scene recognition system based on a multilevel representation of object models and signals is described. The solution to a recognition problem is obtained through a set of object-observation couples at the different abstraction levels. The various system modules exchange two kinds of information: (1) top-down messages, which are used to communicate to lower modules the predictions made on the basis of a priori knowledge on the application domain, (2) bottom-up messages, which are used to communicate to higher modules the evidence supporting possible local solutions. A local scheme for the combination of message flows is defined, and messages are interpreted by using a probabilistic network of estimators of random variables. The proposed model is suitable for addressing the problem of distributed geometric reasoning aimed at 3D road scene recognition by an autonomous vehicle. Recognition results include road detection and obstacle localization, together with a study of the relative computational load required by different modules of the system. The proposed approach is currently simulated on a workstation, while an effective implementation on board of an autonomous vehicle is under development in the contest of the CEC-EUREKA Prometheus project. >

Moving Object Recognition from an Image Sequence for Autonomous Vehicle Driving

Abstract This paper is focused on the recognition of moving objects for an autonomous vehicle driving. In particular, a method for detecting straight lines and their correspondences in successive image frames by means of a straight-line extraction and matching process is proposed. The main advantage offered by the described approach is the possibility of extracting and matching straight lines directly in the feature space, without requiring complex inverse transformations. In particular, the Direct Hough Transform (DHT) algorithm for straight-segment is proposed. It aims to avoid the most important problems associated with the classical HT: (a) spatial information loss, (b) spurious peaks, and (c) discretization effects. Then, a Straight-Line Matching (SLM) algorithm is presented, which utilizes only four attributes to describe a segment: (a) position ρ, (b) orientation θ, (c) length 1, and (d) midpoint m. Finally, results of some experiments performed using synthetic and real monocular image sequences are reported.

Associative and symbolic algorithms for viewpoint-independent object recognition

The authors address a specific problem related to image processing and understanding, i.e., recognition of three-dimensional (3-D) objects from a set of 2-D views. The chosen application is recognition of road scenes for semiautonomous vehicle driving. A recognition scheme is proposed that uses an associative mechanism as a guide sensor to estimate scene probabilities and probabilistic expectations of objects and their views in order to support a symbolic search process. Associative matching results are used to assess object probabilities of being present in the scene; such probabilities are utilized to define a top-down searching order to be followed by the symbolic recognition process. Some new results are reported on generation of specific keys for coding 2-D object views in associative memories.<<ETX>>

A numerical and symbolic fusion method for interpretation of image sequence

The problem of analyzing a sequence of images by taking into account the symbolic and numerical content of the signal is considered. An algorithm for segmentation and tracking of regions among images of a sequence is presented. The method is based on a Gibbs Markov random field (GMRF) model which couples the image process to a spatial-temporal region process. Optical flow field is used to adaptively decide the temporal clique to be used during annealing of energy. Displacement vectors of pixels belonging to recognized regions are predicted by using prior knowledge about object behavior.<<ETX>>

An ADAS Design based on IoT V2X Communications to Improve Safety - Case Study and IoT Architecture Reference Model.

Several technologies are used today to improve safety in transportation systems. The development of a system for drivability based on both V2V and V2I communication is considered an important task for the future. V2X communication will be a next step for the transportation safety in the nearest time. A lot of different structures, architectures and communication technologies for V2I based systems are under development. Recently a global paradigm shift known as the Internet-of-Things (IoT) appeared and its integration with V2I communication could increase the safety of future transportation systems. This paper brushes up on the state-of-the-art of systems based on V2X communications and proposes an approach for system architecture design of a safe intelligent driver assistant system using IoT communication. In particular, the paper presents the design process of the system architecture using IDEF modeling methodology and data flows investigations. The proposed approach shows the system design based on IoT architecture reference model.

Three-dimensional line segment extraction and grouping from image sequences

The problem of grouping 3D coplanar line segmented obtained from a single view is addressed. The proposed method is efficient and has been tested on both synthetic and real images. First, a Hough-based algorithm is used to detect 2D line segments in a sequence of images representing a 3D scene. Secondly, the 3D coordinates of the line segments are estimated, at each time instant, by means of an extended Kalman filter, based on the displacements (u,v) of the line segment endpoints on the image plane. Finally, 3D coplanar segments are grouped by a 3D voting approach. The novelty of this method lies in the possibility of using a simple voting scheme similar to that associated with the standard Hough transform for line extraction, where each edge point votes for a sheaf of rectilinear lines. In the proposed approach, each line segment votes for a sheaf of planes.