A Shackle Process for Shadow Detection
V. Suriya, Sona Poulose, S. Anila · 2013
The presence of shadows in images can represent a serious obstacle for their full exploitation. Shadows are a decrease in the amount of light that reaches a surface. They are a local change in the amount of light rejected by a surface towards the observer. Coping with shadows is a crucial challenge in object detection, scene understanding, recognition and tracking applications. In the proposed technique, detection of shadow region is performed by using morphological operations. Borders are identified by finding the difference between dilation and erosion processes. The classification process is implemented by means of the KNN (K-Nearest Neighbourhood) classifier. Colour segmentation is performed to compare with the results of the border image created. The comparison results of the colour segmented and border image are considered in terms of classification. Thus, using the proposed technique the classification of shadows and nonshadows is better than the segmentation technique. General Terms Geoscience and remote sensing, Thresholding, Morphological operations, KNN classifier and colour segmentation, LORENZI et al.