Hyperspectral imaging sensors are promising tools for monitoring crop plants or vegetation in different environments. Information on physiology, architecture or biochemistry of plants can be assessed non-invasively and on different scales. For instance, hyperspectral sensors are implemented for stress detection in plant phenotyping processes or in precision agriculture. Up to date, a variety of non-imaging and imaging hyperspectral sensors is available. The measuring process and the handling of most of these sensors is rather complex. Thus, during the last years the demand for sensors with easy user operability arose. The present study introduces the novel hyperspectral camera Specim IQ from Specim (Oulu, Finland). The Specim IQ is a handheld push broom system with integrated operating system and controls. Basic data handling and data analysis processes, such as pre-processing and classification routines are implemented within the camera software. This study provides an introduction into the measurement pipeline of the Specim IQ as well as a radiometric performance comparison with a well-established hyperspectral imager. Case studies for the detection of powdery mildew on barley at the canopy scale and the spectral characterization of Arabidopsis thaliana mutants grown under stressed and non-stressed conditions are presented.
The severity of plant diseases, traditionally the proportion of the plant tissue exhibiting symptoms, is a key quantitative variable to know for many diseases and is prone to error. Good quality disease severity data should be accurate (close to the true value). Earliest quantification of disease severity was by visual estimates. Sensor-based image analysis including visible spectrum and hyperspectral and multispectral sensors are established technologies that promise to substitute, or complement visual ratings. Indeed, these technologies have measured disease severity accurately under controlled conditions but are yet to demonstrate their full potential for accurate measurement under field conditions. Sensor technology is advancing rapidly, and artificial intelligence may help overcome issues for automating severity measurement under hyper-variable field conditions. The adoption of appropriate scales, training, instruction and aids (standard area diagrams) has contributed to improved accuracy of visual estimates. The apogee of accuracy for visual estimation is likely being approached, and any remaining increases in accuracy are likely to be small. Due to automation and rapidity, sensor-based measurement offers potential advantages compared with visual estimates, but the latter will remain important for years to come. Mobile, automated sensor-based systems will become increasingly common in controlled conditions and, eventually, in the field for measuring plant disease severity for the purpose of research and decision making.
The application of hyperspectral imaging technology for plant disease detection in the field is still challenging. Existing equipment and analysis algorithms are adapted to highly controlled environmental conditions in the laboratory. However, only real time information from the field scale is able to guide plant protection measures and to optimize the use of resources. At the field scale, many parameters such as the optimal measurement distance, informative feature sets, and suitable algorithms have not been investigated. In this study, the hyperspectral detection and quantification of yellow rust in wheat was evaluated using two measurement platforms: a ground-based vehicle and an unmanned aerial vehicle (UAV). Different disease development stages and disease severities were provided in a plot-based field experiment. Measurements were performed weekly during the vegetation period. Data analysis was performed by three prediction algorithms with a focus on the selection of optimal feature sets. In this context, the across-scale application of optimized feature sets, an approach of information transfer between scales, was also evaluated. Relevant aspects for an on-line disease assessment in the field integrating affordable sensor technology, sensor spatial resolution, compact analysis models, and fast evaluation have been outlined and reflected upon. For the first time, a hyperspectral imaging observation experiment of a plant disease was comparatively performed at two scales, ground canopy and UAV.
This study establishes a method to detect and distinguish between brown rust and yellow rust on wheat leaves based on hyperspectral imaging at the leaf scale under controlled laboratory conditions. A major problem at this scale is the generation of representative and correctly labelled training data, as only mixed spectra comprising plant and fungal material are observed. For this purpose, the pure spectra of rust spores of Puccinia triticina and P. striiformis, causal agents of brown and yellow rust, respectively, were used to serve as a spectral fingerprint for the detection of a specific leaf rust disease. A least-squares factorization was used on hyperspectral images to unveil the presence of the spectral signal of rust spores in mixed spectra on wheat leaves. A quantification of yellow and brown rust, chlorosis and healthy tissue was verified in time series experiments on inoculated plants. The detection of fungal crop diseases by hyperspectral imaging was enabled without pixel-wise labelling at the leaf scale by using reference spectra from sporescale observations. For the first time, this study shows an interpretable decomposition of the spectral reflectance mixture during pathogenesis. This novel approach will support a more sophisticated and precise detection of foliar diseases of wheat by hyperspectral imaging.
The characterization of plant disease symptoms by hyperspectral imaging is often limited by the missing ability to investigate early, still invisible states. Automatically tracing the symptom position on the leaf back in time could be a promising approach to overcome this limitation. Therefore we present a method to spatially reference time series of close range hyperspectral images. Based on reference points, a robust method is presented to derive a suitable transformation model for each observation within a time series experiment. A non-linear 2D polynomial transformation model has been selected to cope with the specific structure and growth processes of wheat leaves. The potential of the method is outlined by an improved labeling procedure for very early symptoms and by extracting spectral characteristics of single symptoms represented by Vegetation Indices over time. The characteristics are extracted for brown rust and septoria tritici blotch on wheat, based on time series observations using a VISNIR (400–1000 nm) hyperspectral camera.
This work established a hyperspectral library of important foliar diseases of wheat in time series to detect spectral changes from infection to symptom appearance induced by different pathogens. The data was generated under controlled conditions at the leaf-scale. The transition from healthy to diseased leaf tissue was assessed, spectral shifts were identified and used in combination with histological investigations to define developmental stages in pathogenesis for each disease. The spectral signatures of each plant disease that are indicative of a certain developmental stage during pathogenesis - defined as turning points - were combined into a spectral library. Different machine learning analysis methods were applied and compared to test the potential of this library for the detection and quantification of foliar diseases in hyperspectral images. All evaluated classifiers provided a high accuracy for the detection and identification for both the biotrophic fungi and the necrotrophic fungi of up to 99%. The potential of applying spectral analysis methods, in combination with a spectral library for the detection and identification of plant diseases is demonstrated. Further evaluation and development of these algorithms should contribute to a robust detection and identification system for plant diseases at different developmental stages and the promotion and development of site-specific management techniques of plant diseases under field conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.