Machine Learning approaches for tassel identification
The tassel, the part of the corn plant containing the pollen, is affected by many factors in a complex manor. Furthermore, different genotypes of corn will react differently to their environment, releasing the pollen to fertilize other corn plants when conditions align. In order to advance efforts in plant breeding, it is important to understand the factors affecting the production and release of pollen, and how these factors interact with different genotypes. In this project we will develop a system for automated field-based phenotyping. We will design and implement a convolutional neural network that takes an input of thousands of pictures a day of many different genotypes of corn plants to identify the tassel from the pictures. Once isolated, we will analyze the pollination over time of the various genotypes through post processing to determine the number of spikelets on a tassel, the width of the main spike, and the average length of the antlers in order to track flowering and pollination.