Development of a Seed Treatment Technique for Automatic Identification of Soybean Plants

Abstract. The increasing labor costs of weed control and the continued emergences of herbicide-resistant weeds are threatening grower ability to manage weeds and maintain profitability. One strategy for rapid plant differentiation is to utilize a machine-detectable compound to mark the crop. This paper proposes a new systemic crop signaling technology based on seed treatment that can create machine-readable soybean crops to facilitate the automated removal of intra-row weeds in early growth stages. This study involves applying different solutions of Rh–B to the soybean seeds prior to planting to determine the spatial location of each soybean plant. The results showed that the fluorescence imaging system can identify and visualize the distribution of Rh–B in soybean plants during their early growth stages. The contributions of this research allow the computer vision system to reliably identify the soybean plants, and thus showing technical feasibility for outdoor weed control.