Abstract
Leaf senescence, an indicator of plant age and ill health, is an important phenotypic trait for the assessment of a plant’s response to stress. Manual inspection of senescence, however, is time consuming, inaccurate and subjective. In this paper we propose an objective evaluation of plant senescence by color image analysis for use in a high throughput plant phenotyping pipeline. As high throughput phenotyping platforms are designed to capture whole-of-plant features, camera lenses and camera settings are inappropriate for the capture of fine detail. Specifically, plant colors in images may not represent true plant colors, leading to errors in senescence estimation. Our algorithm features a color distortion correction and image restoration step prior to a senescence analysis. We apply our algorithm to two time series of images of wheat and chickpea plants to quantify the onset and progression of senescence. We compare our results with senescence scores resulting from manual inspection. We demonstrate that our procedure is able to process images in an automated way for an accurate estimation of plant senescence even from color distorted and blurred images obtained under high throughput conditions.
Highlights
Even though image processing and computer vision methods have been applied in a range of plant biology contexts and over a span of years [1,2,3,4,5,6], the use of these techniques in a fully automated and high-throughput setting is still being established
We propose a new approach for color distortion correction in blurred images for the specific purpose of analyzing plant senescence
Given the absence of actual ground-truth information and given that a correction matrix is constructed from distorted data, it is prudent to first assess the performance of our approach based on an image of a young plant that we have used for training
Summary
Even though image processing and computer vision methods have been applied in a range of plant biology contexts and over a span of years [1,2,3,4,5,6], the use of these techniques in a fully automated and high-throughput setting is still being established. This applies to the topic addressed in this paper: the automated phenotypic analysis of leaf senescence, one of the trademark indicators of plant age and ill health. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript
Sign-in/Register to view highlights & summary 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.
