Tracking colour object in presence of shadow

Segmentation is one of the most important steps leading to the analysis of images and image sequences. Its main goal is to divide an image into parts that have a strong correlation with objects or areas of the real world contained in the scene. A frequent occurrence in images is given by shadows. It is not difficult for human eyes to distinguish shadows from objects. However, identifying shadows by computer is a challenging research problem. Shadows are often integral parts of natural scenes and their identification is an important task in image and video analysis. On one hand, shadows are a valuable source of information about the shape and the relative position of objects in the scene. On the other hand, their presence may hinder the performance of object segmentation and interpretation systems for applications that require the identification of objects. Shadows can cause object merging, object shape distortion and even object losses (due to the shadow cast over another object). Moreover, the information about the shape and the colour of segmented objects may be distorted by shadows. Traditional motion-based tracking schemes cannot usually distinguish the shadow from the object itself, and this result in a falsely captured object shape. If we want to utilize the object’s shape information for a pattern recognition task, this poses a severe difficulty. In this context arises the problem of identifying shadows and extracting useful information from the visual data they represent. The objective is to provide an efficient approach for the recognition of different types of shadows in both still images and video sequences. One of the main challenges in these applications is tracking the real object in a general video environment even if there is strong shadow influence (the shadow’s effect is greatly attenuated). In video object tracking, shadows are commonly present for general lighting environments and present a confounding factor for correct object tracking. Tracking the real object in presence of shadow is a challenging problem. In this paper, we present a new method for objects tracking in a general video environment based on an illumination invariant colour space and a new external snake force called virtual electric field. Experimental results show that the proposed method can track the real object even if there is strong shadow influence.