Abstract
This study describes the evaluation of a range of approaches to semantic segmentation of hyperspectral images of sorghum plants, classifying each pixel as either nonplant or belonging to one of the three organ types (leaf, stalk, panicle). While many current methods for segmentation focus on separating plant pixels from background, organ-specific segmentation makes it feasible to measure a wider range of plant properties. Manually scored training data for a set of hyperspectral images collected from a sorghum association population was used to train and evaluate a set of supervised classification models. Many algorithms show acceptable accuracy for this classification task. Algorithms trained on sorghum data are able to accurately classify maize leaves and stalks, but fail to accurately classify maize reproductive organs which are not directly equivalent to sorghum panicles. Trait measurements extracted from semantic segmentation of sorghum organs can be used to identify both genes known to be controlling variation in a previously measured phenotypes (e.g., panicle size and plant height) as well as identify signals for genes controlling traits not previously quantified in this population (e.g., stalk/leaf ratio). Organ level semantic segmentation provides opportunities to identify genes controlling variation in a wide range of morphological phenotypes in sorghum, maize, and other related grain crops.
Highlights
A wide range of plant morphological traits are of interest and of use to plant breeders and plant biologists
Forty days after planting (DAP), all plants were placed on previously described conveyor belt imaging and automatic watering system [8]
Annotated using the Zooniverse crowdsourcing platform, which substantially reduced the amount of tool development necessary to able to record both the locations of the clicked pixels and the corresponding label information in order to generate ground truth data
Summary
A wide range of plant morphological traits are of interest and of use to plant breeders and plant biologists. Leaf number at reproductive maturity, and the size and area of the largest leaf are all parameters employed in crop growth models to estimate plant performance in different environments [6]. These parameters are currently quantified using low-throughput and labor-intensive methodologies, limiting the feasibility of constructing models for large numbers of genotypes [7]. Accuracy was defined as the number of pixels assigned the same label by manual classifiers and the algorithm being evaluated divided by the total number of pixels classified by both manual classifiers and the algorithm. As this was a balanced dataset with four total classes, the null expectation for accuracy from an algorithm which assigned labels randomly is 0.25
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
