Biological plant root growth detection from spatial and temporal resolution image sequences
This thesis describes the development of a new approach to measuring the growth of plant roots. Work on changing the growth patterns of plants by the introduction of the right materials into their feed as well as the process of genetic manipulation is enhanced by being able to measure the growth of...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/12039/ |
| _version_ | 1848791417322733568 |
|---|---|
| author | Chen, Xiaolin |
| author_facet | Chen, Xiaolin |
| author_sort | Chen, Xiaolin |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | This thesis describes the development of a new approach to measuring the growth of plant roots. Work on changing the growth patterns of plants by the introduction of the right materials into their feed as well as the process of genetic manipulation is enhanced by being able to measure the growth of the plants roots in real time. Previous work in doing this has been subject to low reliability due in part to the nature of the problem. Plant root growth rates are of the order of 0.1 μm per second and thus have to be captured under the microscope. The plant surfaces show low contrast and have few predictable features so many methods prove to be inappropriate. Previous work in the measurement made use of the RootFlowRT software that uses a combination of a tensor based method and a correspondence method. However, the results from these methods have a high level of unreliability. The tensor method as applied shows a reliability of less than 10% and work carried out in this thesis shows that the correspondence method on its own cannot reliably predict the growth rates for large areas in any root.
The work has introduced the use of Scale Space Optical Flow method to replace the previous tensor method and this has been shown to have a reliability of greater than 30% in almost all cases. The results of this method are then used to refine the search space for the correspondence method and again increase the reliability of the measurements.
The validity of the final results using the current method are thus shown to be a great improvement on the previous method. For comparison:
Percentage of measurements in the correct direction and size
• RootFlowRT 70%
• Current method 95%
Maximum spread of invalid results
• RootFlowRT +/-200% in size and 100% in direction
• Current method +/-10% in size or direction |
| first_indexed | 2025-11-14T18:28:11Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-12039 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T18:28:11Z |
| publishDate | 2011 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-120392025-02-28T11:17:09Z https://eprints.nottingham.ac.uk/12039/ Biological plant root growth detection from spatial and temporal resolution image sequences Chen, Xiaolin This thesis describes the development of a new approach to measuring the growth of plant roots. Work on changing the growth patterns of plants by the introduction of the right materials into their feed as well as the process of genetic manipulation is enhanced by being able to measure the growth of the plants roots in real time. Previous work in doing this has been subject to low reliability due in part to the nature of the problem. Plant root growth rates are of the order of 0.1 μm per second and thus have to be captured under the microscope. The plant surfaces show low contrast and have few predictable features so many methods prove to be inappropriate. Previous work in the measurement made use of the RootFlowRT software that uses a combination of a tensor based method and a correspondence method. However, the results from these methods have a high level of unreliability. The tensor method as applied shows a reliability of less than 10% and work carried out in this thesis shows that the correspondence method on its own cannot reliably predict the growth rates for large areas in any root. The work has introduced the use of Scale Space Optical Flow method to replace the previous tensor method and this has been shown to have a reliability of greater than 30% in almost all cases. The results of this method are then used to refine the search space for the correspondence method and again increase the reliability of the measurements. The validity of the final results using the current method are thus shown to be a great improvement on the previous method. For comparison: Percentage of measurements in the correct direction and size • RootFlowRT 70% • Current method 95% Maximum spread of invalid results • RootFlowRT +/-200% in size and 100% in direction • Current method +/-10% in size or direction 2011-07-13 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en arr https://eprints.nottingham.ac.uk/12039/1/XiaolinChenPhD.pdf Chen, Xiaolin (2011) Biological plant root growth detection from spatial and temporal resolution image sequences. PhD thesis, University of Nottingham. plant root growth optical flow scale space |
| spellingShingle | plant root growth optical flow scale space Chen, Xiaolin Biological plant root growth detection from spatial and temporal resolution image sequences |
| title | Biological plant root growth detection from spatial and temporal resolution image sequences |
| title_full | Biological plant root growth detection from spatial and temporal resolution image sequences |
| title_fullStr | Biological plant root growth detection from spatial and temporal resolution image sequences |
| title_full_unstemmed | Biological plant root growth detection from spatial and temporal resolution image sequences |
| title_short | Biological plant root growth detection from spatial and temporal resolution image sequences |
| title_sort | biological plant root growth detection from spatial and temporal resolution image sequences |
| topic | plant root growth optical flow scale space |
| url | https://eprints.nottingham.ac.uk/12039/ |