In this paper we describe a people tracking system addressing the visual surveillance of outdoor environments. This context involves the detection of objects moving in a sequence of images acquired by a TV camera and the recognition and tracking of human figures to allow the analysis of their gestures. The proposed tracking approach does not consider the human body parts as features to be tracked, nor uses a priori body models. Moreover, also classical prediction models are useless in the specific context (visual surveillance of archaeological sites) because the behaviors identified as illegal imply a sequence of unpredictable gestures performed in a small area. The proposed method is based on the estimation of similarity between human figures segmented in the temporal sequence. The similarity score is estimated as a function of the number of corresponding feature points. The system has been verified on real image sequences acquired by a static TV camera while the gestures normally performed by intruders were simulated in a real archaeological site.

Our application context is the visual surveillance of archeological sites. In this context the main aim is to detect the presence of people and to scan them in order to recognize intruders on the basis of their gestures. Since an intruder needs some utensils indispensable to perform the illegal actions of excavating on the ancient ruins, intruder detection involves first of all to ascertain if a person is carrying some objects and then recognizing the kind. In this paper we concentrate on the recognition of the objects carried by the detected moving persons. An example-based learning technique is used to first detect people and successively to scan them to recognize the possible carried objects. The patterns to be analysed are represented through the approximation coefficients of their three level wavelet decomposition. Pattern classification is performed through a supervised three layer neural network.

We show that qualitative properties of optical flow fields suffice for realizing simple autonomous behaviours of a mobile robot in real-time. Such properties are derived by analyzing the motion components (expansion, rotation, X-Y translation and deformation), previously described by Subbarao (1990) and Cipolla (1992), for a local flow field. In particular, we show that such components can be reliably measured by translation invariant, linear and uncalibrated detectors, each one tuned on a particular motion component. The main difference between our scheme and the one proposed by Cipolla is that ours does not require any physical contour to be tracked for inferring qualitative properties of the perceived optical flow. Moreover, the measurement of the affine transformation coefficients does not involve any matrix inversion. The responses of the proposed motion component tuned detectors, expressed in terms of their relative excitation to the perceived visual stimulus, are used as controlled signals in closed feedback loop controllers, working in real-time (12.5 Hz), for realizing three types of opto-motor reflexes.

The goal of this work is to propose a method to solve the problem of underwater navigation of an autonomous vehicle with visual means. We refer to the ability of an autonomous system to determine its motion with respect to the environment. The method we propose is a two stage approach: matching of features extracted from 2D images acquired at different times and egomotion parameter computation. Both algorithms are based on optimization approaches minimizing appropriate energy functions. The matching stage recovers correspondences between sparse high interest feature points of two successive images useful to perform the second stage of egomotion parameter estimation. Experimental results obtained in real context show the robustness of the method.

This paper deals with oriented texture analysis applied to a problem of defect detection and classification in leather inspection, for industrial application. In this paper we present results obtained using a defects detector based on oriented texture analysis, which reveals itself useful for a few classes of leather defects, such as scars or folds. These defects can be detected by using a black and white camera running over the leather patch and by classifying textures, on the basis of their gradient orientations and local coherence. Leather defects are detected by segmenting the oriented texture map of the leather surface. The orientation in every point is expressed using a normalized B-spline basis whose knots are distributed on the leather surface. The resulting erector of coefficients can be used to perform the final segmentation and localize defective areas. A morphological segmentation procedure is applied to the regularized oriented texture field in order to extract probable defective areas.

Sparse optical flow estimated by an image sequence, often, is enough to obtain useful information about camera motion. Normally, matching among features on image sequence is estimated by radiometric similarity. When the camera is moving on a flat surface (planer navigation) additional constraints can be introduced in order to obtain a more robust estimation of the optical flow. In this paper, an energy based optimization method to compute a robust sparse optical flow using global constraints induced by camera motion, is proposed. Convergence is reached in few iterations, making the proposed method suitable for real-time applications.

The main aim of this paper is to describe an efficient approach for egomotion parameters estimation in order to solve the passive navigation problem. Passive navigation is the ability of an autonomous agent to determine its motion with respect to the environment. The two main egomotion parameters allowing to perform passive navigation are the heading direction and the time to collision with the environment. A lot of approaches have been proposed in literature in order to estimate the above parameters, most of which work well only if the motion is a predominant translation forward or small amounts of noise are present in the input data. In our work we propose a method computing in real time, using a least square approach, the egomotion parameters from same robust feature correspondences. The approach we propose is based on the observation that the radial structures of the 2D motion field generated by passive navigation can be approximated by a 2D motion field corresponding to the projection on the image plane of a translational motion on both X horizontal axis and Z axis. The aim is to show that the heading direction can be computed correctly from translational motion parameters, though large rotations are performed by the vehicle.

This paper deals with the problem of recovering directionality from compositional textures. It presents a new method that has proved to be effective for detection and classification of leather defects in the manufacturing industry. A flow field, estimated from texture, is represented as a set of projection coefficients onto suitable elementary basis vector fields. Since these bases are neither orthogonal nor complete, the coefficients must be estimated using a global optimization criterion. The method performs this optimization using a neural network. A region growing approach, based on single iterations of the same neural network, expands patches corresponding to oriented structures and produces the final segmentation map.

The goal of this work is to propose a method to solve the problem of passive navigation with visual means. The method is a two stage approach: matching of feature extracted from 2D images of a sequence at different times and egomotion parameter computation. Both algorithms are based on a least-square-error technique to minimize appropriate energy functions. Experimental results obtained in real context show the robustness of the method.

In this paper a neural algorithm for defect detection in industrial inspection is proposed. One of the most difficult problems in process control and automated inspection is the identification, description and classification of surface defects and anomalies. A critical role in surface inspection is played by texture because most of the defects are rich in textural content. The goal of this work is to propose a quantitative and qualitative technique to detect surface defects with oriented texture structure. The neural algorithm, that the authors propose to classify an oriented flow field, can classify each kind of defect with textural characteristics: once the system has been set to recognize a limited number of patterns, it is able to identify and analyze a broader family of patterns. To the surface image to be analyzed is associated a vector field which computes the dominant local orientations of the gradients of the image smoothed with a Gaussian filter. The different defective surface regions are recovered and classified minimizing an energy function by means of a neural network. Experimental results of tests performed on ferromagnetic surfaces show as the proposed neural at algorithm works.