Applications of hyperspectral imaging in plant phenotyping

Hyperspectral imaging captures both spectral (λ) and spatial (x, y) information and merges these into a 3D data matrix termed a ‘hyperspectral data cube’ (hypercube). A 3D hyperspectral data cube consists of 2D of spatial information plus one spectral dimension that contains information for hundreds of spectral bands. Hyperspectral imaging has been applied to detect abiotic, biotic, and quality traits in plants in indoor and outdoor growing conditions. Hyperspectral imaging can be applied from a cellular to landscape scale to determine plant traits. Data processing and mining tools are still evolving, with machine learning and deep learning algorithms being used in order to assist scientists in predicting traits. Our ability to interrogate and manipulate the genome far exceeds our capacity to measure the effects of genetic changes on plant traits. Much effort has been made recently by the plant science research community to address this imbalance. The responses of plants to environmental conditions can now be defined using a variety of imaging approaches. Hyperspectral imaging (HSI) has emerged as a promising approach to measure traits using a wide range of wavebands simultaneously in 3D to capture information in lab, glasshouse, or field settings. HSI has been applied to define abiotic, biotic, and quality traits for optimisation of crop management. Our ability to interrogate and manipulate the genome far exceeds our capacity to measure the effects of genetic changes on plant traits. Much effort has been made recently by the plant science research community to address this imbalance. The responses of plants to environmental conditions can now be defined using a variety of imaging approaches. Hyperspectral imaging (HSI) has emerged as a promising approach to measure traits using a wide range of wavebands simultaneously in 3D to capture information in lab, glasshouse, or field settings. HSI has been applied to define abiotic, biotic, and quality traits for optimisation of crop management. process of finding anomalies, patterns, and correlations within large datasets to predict outcomes. a subset of machine learning referring to the use of multiple and ‘deep’ layers of a neural network, which are cascaded to extract high-level information, and for pattern recognition and data predictions. an imaging technique that collects spectral and temporal information of reflected light arriving from the imaged surface through hundreds of spectral channels. Often used to interpret the chemical and physical properties of the imaged object. the resulting dataset produced by an HSI camera, which consists of two spatial and one spectral dimension. The main limitation of the data cube is extremely large dimensions because of the high resolution of spectral data. phenotyping undertaken in an enclosed and partially or fully controlled environment, which includes glasshouses/greenhouses, growth chambers, and labs. Plants can be illuminated entirely or partially by artificial lighting. Typically, imaging is undertaken using automated systems and handheld devices. an algorithm that can be trained and self-adjust for progressive learning based on experience and input data in the form of text, numbers, images, video, and so forth. nondestructive imaging of plants using one or more techniques on the electromagnetic spectrum in order to observe and measure plant traits. phenotyping undertaken outdoors, usually on farm fields or natural ecosystems, using only a natural light source. Typically, imaging is undertaken by ground-based vehicles, drones, aircraft, and satellites. arising from the emission or absorption of a photon with energy corresponding to the difference between initial and final states of the transition. In the instance of phenotyping images, a spectral feature refers to an observable change in the electromagnetic signature corresponding to one or more pixels. A spectral feature may indicate an area of interest such as the disease or abiotic/biotic stress. a number that quantifies vegetation biomass or plant vigour presented in a single pixel. a range of wavelengths falling between two given limits in the electromagnetic spectrum corresponding to an HSI system.