An automated, high-throughput method for standardizing image color profiles to improve image-based plant phenotyping

• DOI: 10.7717/peerj.5727
• Publication Date: 2018-10-04T15:15:30Z
• Authors: Jeffrey C. Berry; Noah Fahlgren; Alexandria A. Pok...; Rebecca S. Bart; Kira M. Veley
• Keywords: Data set

High-throughput phenotyping has emerged as a powerful method for studying plant biology. Large image-based datasets are generated and analyzed with automated image analysis pipelines. A major challenge associated these analyses is variation in quality that can inadvertently bias results. Images made up of tuples data called pixels, which consist R, G, B values, arranged grid. Many factors, example brightness, influence the captured. These factors alter values pixels within images consequently downstream analyses. Here, we provide an to adjust dataset so contrast, color profile standardized. The correction collection linear models adjusts pixel based on reference panel colors. We apply this technique set taken high-throughput imaging facility successfully detect variance dataset. In case, resulted from temperature-dependent light intensity throughout experiment. Using method, were able standardize dataset, show enhanced our ability accurately quantify morphological measurements each image. implement pipeline available paper, it also implemented PlantCV.