Shape Analytics for In Situ Fine Root Measurements

The quantification of in situ fine root phenotypes is an important task in the comprehensive study of natural and agricultural environments and critical to under- standing how plants respond to various changes in those environments, especially as those changes relate to energy and environmental challenges Mini-rhizotron experiments enable nondestructive imaging of in situ root systems, but their analysis is complicated by a labor and expertise intensive manual image segmentation task We propose to develop a gener- alizable automated image analysis software pipeline for the segmentation and topological quantification of root system phenotypes in order to enable high throughput mini-rhizotron experiments We intend to utilize unsupervised image processing algorithms and ad- vanced shape analytics methods to overcome a number of barriers to automation of mini- rhizotron image analysis These include unpredictable soil environments, multiple over- lapping root and fungal species, partial occlusions by soil or particulate matter, and even seasonal variations in soil layer heights Unsupervised wavelet decompositions of images produce representations that are robust to many of these obstacles Coupled with advanced shape analytics (robust measures of topological shape in non-traditional data objects) we expect to develop a software analysis approach that can process large amounts of image data with minimal user intervention In Phase I we will develop a software implementation of our proposed image segmentation and phenotype quantification approach We will perform a targeted data collection effort to validate our algorithms and software Criteria for validation include ease of use, degree of user intervention required, and reliability and applicability of results Root phenotyping is of growing interest in the commercial agriculture sector Lowering the cost to entry (in terms of captital, time, and expertise) to high-throughput mini-rhizotron analysis will be of immense impact in a sec- tor where the value of big data is being increasingly leveraged Also, adaptable automated processing will increase data output of mini-rhizotron experiments, advancing scientific un- derstanding of the nature and dynamics of a critical part of our ecosystems