Presents a time contextual segmentation algorithm based on construction of an `object model' using evidence from several consecutive frames. The aim is to obtain more `correct' and stable segments in low contrast image sequences. Roughly speaking, for each object to be segmented the algorithm generates an object model as an aggregate of segments derived from the most recent frames. Due to object motion, the frame-to-frame displacement must be determined. This is done by a template matching technique. The segmentation is terminated when the object model has converged to a stable segment. Good results have been obtained in situations where single frame based algorithms due to either low local contrast or partial occlusion, have failed. The method may be implemented for real time applications

Many types of sensors may be used to gather information on the surrounding environment. It may be noted that different sensors possess distinct characteristics, are designed based on differing physical principles, operate in a wide range of the electromagnetic spectrum, and are geared toward a variety of applications. A single sensor operating alone provides a limited sensing range, and is inherently unreliable due to operational errors. However, a synergistic operation of many sensors provides a rich body of information on the sensed environment from a wide range of the electromagnetic spectrum. In this paper, we present a brief survey of the techniques/systems for sensor data fusion. We first address sensor data fusion in mobile robotics applications. Since mobile robots operate in an uncertain and constantly changing environment, a steady stream of rich and reliable information on the environment is needed for navigation and path planning. We discuss the feasibility of a few sensors in such applications and present several sensor data fusion techniques for road following and terrain analysis. Next, we review techniques for fusing data from visual, thermal, tactile, and structured-lighting sensors and discuss several general frameworks for sensor data fusion.