In Computer Vision, shadow free object recognition is a wide phrase covering a range of applications such as human motion capture, video surveillance, traffic monitoring, segmentation and tracking of foreground objects. Unfortunately, shadows in these applications may appear as foreground objects, when in fact they are caused by the interaction between light and objects. The inability to distinguish between foreground objects and shadows can cause malicious problems such as object merging, false segmentation, misclassified as foreground objects and identification failure, all of which significantly affect the performance of detection and tracking systems. However in most situations, it is essential to avoid shadow as it becomes undesired and unwanted part which deteriorates the outcome. Therefore, an effective shadow detection method is necessary for accurate object segmentation. One of the main challenging problems is identifying insignificant shadow from natural images by computing systems. Though many researchers try to deal with these problem using different methodologies, yet it is intriguing problem. This paper deals with the problem of identifying and extracting regions that correspond to shadow from natural scenes. Also, it aims to produce a comprehensive evaluation on the state-of-the-art methods of detecting shadows from natural images.
CITATION STYLE
Muthukumar, S., Subban, R., Krishnan, N., & Pasupathi, P. (2014). Real time insignificant shadow extraction from natural sceneries. Advances in Intelligent Systems and Computing, 235, 391–399. https://doi.org/10.1007/978-3-319-01778-5_40
Mendeley helps you to discover research relevant for your work.