High-throughput plant phenotyping has revealed its importance by illuminating complex questions about plant growth, development, and response to the environment. Recently emerging technologies including un-maned vehicle as an image acquiring technology and machine learning as image analyzing technology have enabled the use of high-throughput phenotyping of plants for basic and applied sciences. Digital image analysis is one of methods for high throughput phenotyping. However, those techniques are not accessible because of high costs in general and the lack of man power to efficiently operate them. Here, we propose an easily accessible high-throughput phenotyping tool for biomass for many researchers. Four sorghum cultivars were evaluated for their height, a variable well documented for its high correlation with biomass, with both hand-collection and image taking in order to determine if image analysis for biomass in sorghum could replace hand data collection of plant height. Vertical plant images were obtained using a digital camera at one-week interval for 5 weeks. Plant height and node height were hand-collected at the same time. Images were analyzed using a tool, called Canopeo (http://www.canopeoapp.com). Biomass was measured indirectly as percentage units by counting the green pixels in each image. Strong and significant correlations between the percentage of green color and plant height and between the percentage of green color and node height were found for each cultivar and measurement. The results of this proof-of-concept study strongly indicate that digital image analysis could replace the intense labor needed to collect data related to biomass. (C) 2017 Elsevier Ltd. All rights reserved.