Analyzing architectural diversity in maize plants using the skeleton-image-based method

Shoot architecture in maize is widely regarded as being of critical importance since it determines resource use, impacts wind and rain damage tolerance, and affects yield stability. It is important to be able to quantify the diversity among inbred lines in heterosis breeding, especially when describing germplasm resources. However, traditional geometric description methods over simplify shoot architecture and ignore the plants overall architecture, making it difficult to reflect and illustrate diversity. Here we present a new method to describe maize shoot architecture and quantify its diversity by combining computer vision algorithms and persistent homology. Our results revealed that persistent homology can capture key characteristics of shoot architecture in maize and other details that are often overlooked by traditional geometric analysis. Based on this method, the morphological diversity of shoot architecture can be mined (quantified), and the main shoot architecture types can be obtained. Consequently, this method can be used to easily describe the diversity of shoot architecture in a large number of maize materials.