High-Resolution Aerial Imaging Based Estimation of Crop Emergence in Potatoes

• Published in: American journal of potato research 2017-12, Vol.94 (6), p.658-663
• Main Authors: Sankaran, Sindhuja, Quirós, Juan José, Knowles, N. Richard, Knowles, Lisa O.
• Format: Article
• Language: English
• Subjects: Agriculture; Agronomy; Biomedical and Life Sciences; Correlation coefficient; Correlation coefficients; Crop development; Crop diseases; Crop improvement; Crops; Emergence; Global positioning systems; GPS; Ground level; Growth regulators; High resolution; Image processing; Image resolution; Life Sciences; Noise threshold; Plant Breeding/Biotechnology; Plant Genetics and Genomics; Plant Pathology; Plant Sciences; Plants (botany); Potatoes; Remote sensing; Seeds; Short Communication; Unmanned aerial vehicles; Vegetables; Vegetation
• ISSN: 1099-209X; 1874-9380
• DOI: 10.1007/s12230-017-9604-2

Plant emergence and stand establishment are key indicators of early crop development that are routinely assessed in potato agronomy and crop improvement research. The standard method for evaluating emergence is through manual plant counts at regular intervals. In this proof-of-concept study, unmanned aerial vehicles integrated with multispectral imaging were used for high-throughput evaluation of crop emergence under field conditions. High-resolution aerial imaging was performed at 15 m above ground level to capture data from potato plots of two varieties (‘Alturas’ and ‘Payette Russet’) in which the seed had been treated with different concentrations of growth regulators (including non-treated controls). The treatments resulted in differences in plant emergence and establishment. The images were collected at 32, 37, and 43 days after planting (DAP). Image-based features such as plant count, SUM-NDVI, and SUM-BINARY were computed from normalized difference vegetation index (NDVI) images for each treatment plot using ArcGIS ® . The Pearson’s correlation coefficients ( r ) were significant ( p