Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Canopy Center and Plant Quantity Detection of Sugarcane Based on Aerial Visible Images
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2018 ASABE Annual International Meeting 1800706.(doi:10.13031/aim.201800706)
Authors: Lijia Wang, Xiuhua Li, Ce Wang, Yonghua Zhou, Ce Yang, Jiaoyan Ai
Keywords: Canopy Center Detection, Image Recognition, Plant Quantity Counting, Sugarcane, UAV.
Abstract. Sugarcane cultivation believes that reasonable planting specifications can promote the rational distribution of ecological factors such as water, fertilizer, light and heat, so as to obtain higher yield and better quality. Therefore, obtaining the number of sugarcane plants in the field is of great significance for optimizing the planting densities and predicting the yield. A method for automatically detecting the canopy center and plant quantity in the tillering stage of sugarcane based on the aerial visible images was developed in this study. Sugarcane's canopy images were top shot in a sugarcane field by DJI UAV. First, the white veins of sugarcane were segmented from the complex background using morphological open operation, image reduction operation and Otsu's method. After thinning, hole-removal, deburring and small area deletion treatments on the binary image, a relatively clean white vein extraction map was obtained. Next, short veins and impurities were removed after reconnecting the broken vein lines. According to the dense distribution of endpoints in the center regions of plants, DBSCAN algorithm was implemented to identify plant centers based on the endpoints map of white veins. This method has a better performance on the sugarcane canopy taken in a top view, and the accuracy of counting can reach about 91.5% in the tillering stage of sugarcane. This study can automatically detect the quantity of sugarcane plants in early stageand provide technical support for the prediction of sugarcane quality and yield in the later stage.
(Download PDF) (Export to EndNotes)