1,721,135 research outputs found
Adaptive tracking of multiple non rigid objects in cluttered scenes
No full textTracking of non-rigid objects (e.g. humans) is a crucial application for understanding the behavior of objects. Different methods have been presented in literature, whose main drawback is low robustness or high computational load in analysis of cluttered scenes. In the paper a low computational algorithm for tracking non-rigid objects in cluttered scenes is presented. The proposed approach models the shape of the objects by using corners. A learning algorithm is introduced in order to automatically extract the model of the object from a short video sequence acquired immediately before merging of more objects in the scene. The adaptive model extraction mechanism strongly improves method robustness. The method is tested on an existing video-surveillance system in order to track moving objects in cluttered scenes. Results show that the proposed approach gives good performances with low-processing times
HOS-based noise models for signal-detection optimization in non-Gaussian environments
No full textTwo pdf models suitable for describing nonGaussian iid noise are introduced. The models are used in the design of a LOD test for detecting weak signals in real
non Gaussian noise. Results obtained in tlie context of an
underwater acoustic application are encouraging
VISUAL SURVEILLANCE BY DEPTH FROM FOCUS
No full textThis paper aims at describing a visual surveillance prototype system able to monitor a bounded (in depth) area by using a depth-from-focus approach. The single visual camera utilized is characterized by an adjustable lens objective controlled by a host computer. In this way, the system is capable to detect possible intruders entering in the controlled zone, recovering the range information by a depth-from-focus method, and tracking the incoming entity by using a simple strategy. Tracking is performed by computing the range distance such as to keep in focus the moving object that has violated the area. This method is based on an adaptive regulation strategy of the camera parameters which makes the system robust to lighting variations occurring in real environment. Preliminary results show the goodness of the proposed approach, especially concerning the depth information recovery.This paper aims at describing a visual surveillance prototype system able to monitor a bounded (in depth) area by using a depth-from-focus approach. The single visual camera utilized is characterized by an adjustable lens objective controlled by a host computer. In this way, the system is capable to detect possible intruders entering in the controlled zone, recovering the range information by a depth-from-focus method, and tracking the incoming entity by using a simple strategy. Tracking is performed by computing the range distance such as to keep in focus the moving object that has violated the area. This method is based on an adaptive regulation strategy of the camera parameters which makes the system robust to lighting variations occurring in real environment. Preliminary results show the goodness of the proposed approach, especially concerning the depth information recovery
Multilevel GMRF-based segmentation of image sequences
No full textA probabilistic method for obtaining a complete image representation on the basis of spatial-temporal knowledge is presented. The main goal of the algorithm is to obtain a consistent segmentation of a noisy image sequence. Consistent means that the same region must maintain the same label in all consequent images of the sequence where it appears. To this end, a processing scheme is presented which extends Bayesian networks of Gibbs-Markov random fields (GMRF) to segmentation of dynamic scene
Real-time robust detection of moving objects in cluttered scenes
No full textObject recognition is a very important task in computer vision and different techniques have been presented to solve it. In this paper a Hough-type low-computational algorithm for detection of objects in cluttered scenes is presented. The approach is based on the detection of the shape of an object, modeled by means of a set of corners. An automatically model learning method is introduced. The method is used in an existing video-surveillance system in order to increase its detection performances. Results show that the proposed approach provides good performances with low processing times
A hierarchical approach to feature extraction and grouping
No full textIn this paper, the problem of extracting and grouping image features from complex scenes is solved by a hierarchical approach based on two main processes: voting and clustering. Voting is performed for assigning a score to both global and local features. The score represents the evidential support provided by input data for the presence of a feature. Clustering aims at individuating a minimal set of significant local features by grouping together simpler correlated observations, It is based on a spatial relation between simple observations on a fixed level, i.e., the definition of a distance in an appropriate space. As the multilevel structure of the system implies that input data for an intermediate level are outputs of the lower level, voting can be seen as a functional representation of the "part-of" relation between features at different abstraction levels. The proposed approach has been tested on both synthetic and real images and compared with other existing feature grouping methods
A robust method for reflections analysis in color image sequences
No full textA robust algorithm for the detection of reflections in colour image sequences is here presented. The algorithm directly works on the RGB space and it performs an analysis at two different resolution levels, i.e. the pixel level and the region level. The proposed algorithm is used to improve the localisation and tracking capabilities of a video-based surveillance system. Results showing the goodness of the proposed approach are presented
- …
