Profile based image analysis for identification of chopped biomass stem nodes and internodes

Abstract

Because of their significant variation in chemical composition, segregation of chopped biomass into nodes and internodes helps in efficient utilization of these feedstocks. Stem internodes having low ash content are a better feedstock for biofuel and bioenergy applications than nodes. However, separation of these components is challenging because their physical characteristics are similar. We applied an image processing technique to identify nodes and internodes of chopped biomass from scanned digital images. In this study, we utilized the object profile identified differences in the node and internode components and tested on chopped corn stalks and switchgrass stems. We considered four methods of image processing including rectangularity, solidity, width-, and slope-variation and developed an ImageJ plugin for the node–internode identification. Digital chopping of the ends of the objects was necessary for identification, especially dealing with projecting fibers and chipped rough ends, and an algorithm was developed for this. Among the methods tested, width-variation gave the best identification accuracy (97–98%), followed by rectangularity (93–96%), solidity (86–91%), and slope-variation (69–82%). Rectangularity – a relatively simpler method, and solidity – a standard ImageJ output, can be directly used to perform identification. The developed approach of node–internode identification can be easily applied to other chopped biomass and similar materials, and its application may lead to efficient biomass end use in biofuel and bioproduct industries.