Additional file 1 of A two-step registration-classification approach to automated segmentation of multimodal images for high-throughput greenhouse plant phenotyping

Additional file 1: Table S1. Summary of evaluation of 288 case-scenario models for plant image segmentation against ground truth data. 288 case-scenarios result from training of eight classification models (i.e. bayes, da, glm, gpr, linmod, svm, svmreg, net) for three different plant types (arabidopsis, wheat and maize), two different camera views (top/side view), three image modalities (FLU, VIS and FLU+VIS) and three plant developmental stages including: I—juvenile/small, II—mid-stage and III—adult/large shoots, as well as their combinations, i.e. I + II, I + III, II + III, I + II + III.附加文件1：表S1。288个植物图像分割案例场景模型的真值（ground truth）对比评估汇总。该288个案例场景由8种分类模型（即贝叶斯（bayes）、判别分析（da）、广义线性模型（glm）、高斯过程回归（gpr）、线性模型（linmod）、支持向量机（svm）、支持向量回归（svmreg）、神经网络（net））训练构建而来，涵盖3类不同植物：拟南芥（arabidopsis）、小麦（wheat）与玉米（maize）；2种拍摄视角：顶视/侧视；3种成像模态：荧光成像（FLU）、可见光成像（VIS）以及荧光+可见光复合成像（FLU+VIS）；3个植物发育阶段：I期——幼龄/小型植株，II期——中期，III期——成株/大型植株，以及上述阶段的全部组合场景，即I+II、I+III、II+III与I+II+III。