Mobile high-throughput phenotyping using watershed segmentation algorithm

Date

2017-05-01

Journal Title

Journal ISSN

Volume Title

Publisher

Kansas State University

Abstract

This research is a part of BREAD PHENO, a PhenoApps BREAD project at K-State which combines contemporary advances in image processing and machine vision to deliver transformative mobile applications through established breeder networks. In this platform, novel image analysis segmentation algorithms are being developed to model and extract plant phenotypes. As a part of this research, the traditional Watershed segmentation algorithm has been extended and the primary goal is to accurately count and characterize the seeds in an image. The new approach can be used to characterize a wide variety of crops. Further, this algorithm is migrated into Android making use of the Android APIs and the first ever user-friendly Android application implementing the extended Watershed algorithm has been developed for Mobile field-based high-throughput phenotyping (HTP).

Description

Keywords

High-Throughput phenotyping, Plant breeding, Phenotype, PhenoApps, OpenCV with Android, Watershed segmentation

Graduation Month

May

Degree

Master of Science

Department

Department of Computing and Information Sciences

Major Professor

Mitchell L. Neilsen

Date

2017

Type

Thesis

Citation