A new method of macrotextural analysis using run length histogram is introduced. The method can be used to extract effictive features of macrotexture directly from images. The adaptive threshold and the average shape are keys to over-come difficulties of segmentation and describtion of textures. Those are described in detail in the paper and an experimental study in which the grasses were classified successfuly is also given.The accuracy of the classification can reach up 90%. The method introduced in the paper is intuitive and simple to use. It could be applied to industrial automation

robot and other fields where the time for image processing and pattern recognition is very strict.