Scoring Shape Characters of Monocot Leaves

Biologists regularly collect images of leaves for their further studies. One such biological study of leaves is scoring the phenomic characters of leaves for the construction of the Tree of Life (ToL), i.e. the evolutionary lineage of taxa in botany. There is an opportunity for computer vision to help biologists automate this character scoring. In this master’s report, we describe an automatic system for scoring shape characters of monocot leaves. This system uses computer vision to speed up the current manual scoring process. Each leaf is processed using following computational steps: Segmentation, Orientation Alignment and Scoring. We developed two main frameworks for shape character scoring. One uses global leaf descriptors and then applies a Support Vector Machine (SVM). The other matches the leaf shape contour to exemplar shape contours using Dynamic Time Warping (DTW). Our evaluation shows that both frameworks give high performance on scoring leaves. To the best of our knowledge, this is the first computer vision system that addresses the problem of leaf shape character scoring.