Journal: Plant Methods

Article Title: A two-step registration-classification approach to automated segmentation of multimodal images for high-throughput greenhouse plant phenotyping

doi: 10.1186/s13007-020-00637-x

Figure Lengend Snippet: Overview of binary classification models and corresponding MATLAB functions trained using the input data X (i.e. 20 dimensional multi-color space representation of manually segmented FLU+VIS plant and background regions) and the vector of region labels Y (i.e. Y=0 for background, Y=1 for plant image regions)

Article Snippet: Consequently, a set of classification models based on eight classifiers from Table was trained to detect small background (i.e. non-plant) structures on the basis of the object’s color ratios (R/B, R/G, G/B), size and vicinity to the largest image structure assessed by the Euclidean distance map using the bwdist MATLAB function.

Techniques: Plasmid Preparation