Extraction of Phenotypic Traits for Drought Stress Study Using Hyperspectral Images
05 Dec 2017-pp 608-614
TL;DR: This work proposes a novel framework for phenotypic discovery based on autoencoders, which is trained using Simple Linear Iterative Clustering (SLIC) superpixels and shows potential by separating the plant responses into three classes with a finer granularity.
Abstract: High-throughput identification of digital traits encapsulating the changes in plant’s internal structure under drought stress, based on hyperspectral imaging (HSI) is a challenging task. This is due to the high spectral and spatial resolution of HSI data and lack of labelled data. Therefore, this work proposes a novel framework for phenotypic discovery based on autoencoders, which is trained using Simple Linear Iterative Clustering (SLIC) superpixels. The distinctive archetypes from the learnt digital traits are selected using simplex volume maximisation (SiVM). Their accumulation maps are employed to reveal differential drought responses of wheat cultivars based on t-distributed stochastic neighbour embedding (t-SNE) and the separability is quantified using cluster silhouette index. Unlike prior methods using raw pixels or feature vectors computed by fusing predefined indices as phenotypic traits, our proposed framework shows potential by separating the plant responses into three classes with a finer granularity. This capability shows the potential of our framework for the discovery of data-driven phenotypes to quantify drought stress responses.
Citations
More filters
References
More filters
TL;DR: This review focuses on imaging methods used in the phenotyping of plant shoots including a brief survey of the sensors used, namely RGB, chlorophyll fluorescence, thermal and hyperspectral imaging.
Abstract: Current methods of in-house plant phenotyping are providing a powerful new tool for plant biology studies. The self-constructed and commercial platforms established in the last few years, employ non-destructive methods and measurements on a large and high-throughput scale. The platforms offer to certain extent, automated measurements, using either simple single sensor analysis, or advanced integrative simultaneous analysis by multiple sensors. However, due to the complexity of the approaches used, it is not always clear what such forms of plant phenotyping can offer the potential end-user, i.e. plant biologist. This review focuses on imaging methods used in the phenotyping of plant shoots including a brief survey of the sensors used. To open up this topic to a broader audience, we provide here a simple introduction to the principles of automated non-destructive analysis, namely RGB, chlorophyll fluorescence, thermal and hyperspectral imaging. We further on present an overview on how and to which extent, the automated integrative in-house phenotyping platforms have been used recently to study the responses of plants to various changing environments.
187 citations
TL;DR: This work applies for the first time a recent matrix factorisation technique, simplex volume maximisation (SiVM), to hyperspectral data, an unsupervised classification approach, optimised for fast computation of massive datasets.
Abstract: Early water stress recognition is of great relevance in precision plant breeding and production. Hyperspectral imaging sensors can be a valuable tool for early stress detection with high spatio-temporal resolution. They gather large, high dimensional data cubes posing a significant challenge to data analysis. Classical supervised learning algorithms often fail in applied plant sciences due to their need of labelled datasets, which are difficult to obtain. Therefore, new approaches for unsupervised learning of relevant patterns are needed. We apply for the first time a recent matrix factorisation technique, simplex volume maximisation (SiVM), to hyperspectral data. It is an unsupervised classification approach, optimised for fast computation of massive datasets. It allows calculation of how similar each spectrum is to observed typical spectra. This provides the means to express how likely it is that one plant is suffering from stress. The method was tested for drought stress, applied to potted barley plants in a controlled rain-out shelter experiment and to agricultural corn plots subjected to a two factorial field setup altering water and nutrient availability. Both experiments were conducted on the canopy level. SiVM was significantly better than using a combination of established vegetation indices. In the corn plots, SiVM clearly separated the different treatments, even though the effects on leaf and canopy traits were subtle.
130 citations
TL;DR: A method for the segmentation and the automated analysis of time-lapse plant images from phenotyping experiments in a general laboratory setting that can adapt to scene variability and is able to handle images with complicated and changing background in an automated fashion is proposed.
123 citations
26 Oct 2010
TL;DR: A linear time algorithm for the factorization of gigantic matrices that iteratively yields latent components that is efficient, well-grounded in distance geometry, and easily applicable to matrices with billions of entries.
Abstract: Matrix factorization methods are among the most common techniques for detecting latent components in data Popular examples include the Singular Value Decomposition or Non-negative Matrix Factorization Unfortunately, most methods suffer from high computational complexity and therefore do not scale to massive data In this paper, we present a linear time algorithm for the factorization of gigantic matrices that iteratively yields latent components We consider a constrained matrix factorization st~the latent components form a simplex that encloses most of the remaining data The algorithm maximizes the volume of that simplex and thereby reduces the displacement of data from the space spanned by the latent components Hence, it also lowers the Frobenius norm, a common criterion for matrix factorization quality Our algorithm is efficient, well-grounded in distance geometry, and easily applicable to matrices with billions of entries In addition, the resulting factors allow for an intuitive interpretation of data: every data point can now be expressed as a convex combination of the most extreme and thereby often most descriptive instances in a collection of data Extensive experimental validations on web-scale data, including 80 million images and 15 million twitter tweets, demonstrate superior performance compared to related factorization or clustering techniques
48 citations
08 May 2017
TL;DR: An image processing pipeline is developed that comprises of low level processing which enables high-throughput detection of xylem vessels and successfully captures the phenotypic difference between MTU-1010 (d drought susceptible rice cultivar) and Sahbhagi Dhan (drought tolerant Rice cultivar).
Abstract: Xylem vessels play a pivotal role in plant adaptation to drought stress. In this paper, we propose a novel framework that associates automatic segmentation of xylem vessels with its morphological features as a quantitative proxy to predict drought stress response (DSR). We develop an image processing pipeline that comprises of low level processing which enables high-throughput detection of xylem vessels. With no prior information about its size and location, the proposed detection methodology gives an accuracy of 98%. The labelled data for DSR are either not available or are subjectively developed, which is a low-throughput and error prone task. We resolve this problem by employing simplex volume maximization (SiVM) algorithm. The convex representations obtained from SiVM for each xylem in microscopic images based on its shape factors are aggregated to get an automated scoring of the whole plant. Bhattacharya distance is then employed to obtain the divergence of these responses w.r.t. the control group. The proposed framework successfully captures the phenotypic difference between MTU-1010 (drought susceptible rice cultivar) and Sahbhagi Dhan (drought tolerant rice cultivar).
5 citations