For autonomous robot, it is very important to
perceive its environment for reacting environmental
change, such as road and obstacle. We proposed a method
to classify terrain using texture recognition technique. To
recognize material information of some part in an image,
we should divide an image into homogeneous regions
which have same material. For this purpose, we used multi
label segmentation algorithm to make homogeneous super
pixel. After that, we applied Leung and Malik’s method [1]
to recognize what target material is.
perceive its environment for reacting environmental
change, such as road and obstacle. We proposed a method
to classify terrain using texture recognition technique. To
recognize material information of some part in an image,
we should divide an image into homogeneous regions
which have same material. For this purpose, we used multi
label segmentation algorithm to make homogeneous super
pixel. After that, we applied Leung and Malik’s method [1]
to recognize what target material is.