Presentation
Time:
Master’s Project Defense
Date:
2:00 pm
Schorr Center
Room: 211
1100 T St
Lincoln NE 68588
Lincoln NE 68588
Additional Info: SHOR
Master’s Project Defense: Renjie Gui
“Stress Detection and Phenotype Computation using Chlorophyll Fluorescence Image Sequences”
Committee Members: Dr. Ashok Samal and Dr. Sruti Das Choudhury (Co-Advisors), Dr. Hongfeng Yu
Thursday, November 21, 2019, 2 p.m.
211 Schorr Center
Abstract: Plant phenotyping is an emerging multidisciplinary research field that involves plant science, computer science, statistics, and genomics. A phenotype is a biophysical trait and is a function of a plant’s genotype and its growth environment. Understanding their relationship can lead to increased plant yield and improved resource utilization. Using image-based high-throughput plant phenotyping systems, the phenotypes can be computed noninvasively during the plant’s life cycle. With different imaging techniques, a diverse set of plant phenotypes can be computed.
Chlorophyll fluorescence is the light re-emitted by chlorophyll molecules in plants. The fluorescence can be used as an indicator of the performance of a plant’s photosynthesis process. With high-resolution fluorescent imaging, the plant’s photochemical information can be captured. Under stress conditions such as drought, darkness, or lack of nutrition, the plant’s photosynthesis process will likely be affected, and hence, the plant will show a different response than non-stress conditions. This physiological change is difficult to detect in the early stages of stress induction but may be detected from the fluorescent images by examining the change of the fluorescent intensity.
In this study, we focus on drought stress detection using chlorophyll fluorescence image sequences. A data-driven approach has been developed with a stress detection classifier for identifying the stressed parts in the plant. Stress detection is performed using fluorescent image sequences using a classifier that classifies image pixels into 3 different stress classes. Three stress-based phenotypes are proposed and may be computed from the classified plant images. The phenotypes track the temporal change of the impact of stress on the plant and may be used to compare the resistance to drought between different genotypes.
“Stress Detection and Phenotype Computation using Chlorophyll Fluorescence Image Sequences”
Committee Members: Dr. Ashok Samal and Dr. Sruti Das Choudhury (Co-Advisors), Dr. Hongfeng Yu
Thursday, November 21, 2019, 2 p.m.
211 Schorr Center
Abstract: Plant phenotyping is an emerging multidisciplinary research field that involves plant science, computer science, statistics, and genomics. A phenotype is a biophysical trait and is a function of a plant’s genotype and its growth environment. Understanding their relationship can lead to increased plant yield and improved resource utilization. Using image-based high-throughput plant phenotyping systems, the phenotypes can be computed noninvasively during the plant’s life cycle. With different imaging techniques, a diverse set of plant phenotypes can be computed.
Chlorophyll fluorescence is the light re-emitted by chlorophyll molecules in plants. The fluorescence can be used as an indicator of the performance of a plant’s photosynthesis process. With high-resolution fluorescent imaging, the plant’s photochemical information can be captured. Under stress conditions such as drought, darkness, or lack of nutrition, the plant’s photosynthesis process will likely be affected, and hence, the plant will show a different response than non-stress conditions. This physiological change is difficult to detect in the early stages of stress induction but may be detected from the fluorescent images by examining the change of the fluorescent intensity.
In this study, we focus on drought stress detection using chlorophyll fluorescence image sequences. A data-driven approach has been developed with a stress detection classifier for identifying the stressed parts in the plant. Stress detection is performed using fluorescent image sequences using a classifier that classifies image pixels into 3 different stress classes. Three stress-based phenotypes are proposed and may be computed from the classified plant images. The phenotypes track the temporal change of the impact of stress on the plant and may be used to compare the resistance to drought between different genotypes.