Segmentation of pores in wood microscopic images based on mathematical morphology with a variable structuring element

A set of pores is one of the key features in the process of wood identification. A novel method to segment pores in a wood microscopic image is proposed in this paper. The method has three steps. In the first step, structuring elements with several sizes are employed in a mathematical morphology algorithm to enhance the pores and remove other tissues in a wood microscopic image. Then the best structuring element size is determined according to the results obtained from the previous step. Finally, a binary image that includes only pores is obtained by an adaptive thresholding. Experimental results show the efficiency of the method.

Research highlights▶ The application field is new. There hardly exists any work with this kind of approach. ▶ A novel algorithm based on mathematical morphology to segment pores in a wood microscopic image is proposed. ▶ Different algorithms are compared by experiments and the experimental results show the efficiency of the proposed algorithm. ▶ The results of segmentation can be used for further research, such as recognition of pores distribution shape.