Plant Phenomics / 2019 / Article / Fig 6

Research Article

A High-Throughput Phenotyping System Using Machine Vision to Quantify Severity of Grapevine Powdery Mildew

Figure 6

Relationship between severity of infection as determined by the Automated Phenotyping System (APS) neural network versus three independent human experts. Human experts rated the same subimages analyzed by the neural network, and results were analyzed by linear regression. The coefficients of correlation of neural network scores and human observer scores were 0.9877, 0.9768, and 0.9786 respectively for Experts 1, 2, and 3.