Modified artificial bee colony algorithm with multilevel threshold segmentation and boundaries evaluation for shadow detection

Abstract

The appearance of shadows typically causes severe problems in pc vision. There are various methods have already put forward but scope in this field is open. In this article Shadow Detection and Removal Using Modified artificial bee colony (MABC) Algorithm with Multilevel Threshold segmentation is proposed. The proposed method uses three threshold and corresponding boundaries, associated curvature, edge response, gradient, and MABC algorithm. First data pre-processing is applied to find the correlation between the pixels then three threshold and corresponding boundaries evaluated to accurately differentiate pixels as foreground. The edge response, curvature, gradient are applied to find the boundaries. Finally, MABC has been applied for detecting the shadow. The results show improvement in comparison with other existing methods.