Mud and dirt separation method for floor cleaning robot

Floor cleaning robot has gained popularity in the European country. When people are busy with their work, the robot is to take over house cleaning task. Most of the service robot technologies become more and more matured so that applications far beyond fetch and carry tasks will be possible in near future. Most of the robot perform one task at the time. To be of any use, cleaning robots should be able to monitor the outcome of their work, or the need to remove dirt at the beginning for point-wise cleaning. This paper presents an approach towards the detection of various dirt on textured surfaces using a Camera image. The algorithm aims at removing the regular part of the image with spectral residual filtering leaving only the dirt and some noise and introduced a novel method to identify mud on the floor. With further post-processing, it is possible to localize many kinds of dirt in the image.