Color segmentation for visual attention of mobile robots

Segmentation is a very important step in most of the machine vision applications. A method for color segmentation is proposed as a basic requirement of visual attention system of mobile robots. After experiments with different color spaces, the HSI color space is selected for segmentation with minimum influence of shades and shadows. The proposed method first classifies the color of seed pixel and then selects a method for region growing according to nature of the seed pixel. Results obtained are very encouraging and according to requirements of the application area.