In the domain of chemometrics, multiblock data analysis is widely performed for exploring or fusing data from multiple sources. Commonly used methods for multiblock predictive analysis are the ...extensions of latent space modelling approaches. However, recently, deep learning (DL) approaches such as convolutional neural networks (CNNs) have outperformed the single block traditional latent space modelling chemometric approaches such as partial least-square (PLS) regression. The CNNs based DL modelling can also be performed to simultaneously deal with the multiblock data but was never explored until this study. Hence, this study for the first time presents the concept of parallel input CNNs based DL modelling for multiblock predictive chemometric analysis. The parallel input CNNs based DL modelling utilizes individual convolutional layers for each data block to extract key features that are later combined and passed to a regression module composed of fully connected layers. The method was tested on a real visible and near-infrared (Vis-NIR) large data set related to dry matter prediction in mango fruit. To have the multiblock data, the visible (Vis) and near-infrared (NIR) parts were treated as two separate blocks. The performance of the parallel input CNN was compared with the traditional single block CNNs based DL modelling, as well as with a commonly used multiblock chemometric approach called sequentially orthogonalized partial least-square (SO-PLS) regression. The results showed that the proposed parallel input CNNs based deep multiblock analysis outperformed the single block CNNs based DL modelling and the SO-PLS regression analysis. The root means squared errors of prediction obtained with deep multiblock analysis was 0.818%, relatively lower by 4 and 20% than single block CNNs and SO-PLS regression, respectively. Furthermore, the deep multiblock approach attained ∼3% lower RMSE compared to the best known on the mango data set used for this study. The deep multiblock analysis approach based on parallel input CNNs could be considered as a useful tool for fusing data from multiple sources.
Display omitted
•A deep multi-block data modelling method is presented.•An example of fusing two blocks, i.e., visible, and near-infrared data, is demonstrated.•Proposed method outperformed classical multi-block chemometric modelling.
•Illumination effects limits the close-range hyperspectral imaging of whole plants.•Light scattering, shadowing, multiple reflections are the result of illumination effects.•6 different approaches to ...illumination correction are available.•Spectral normalization is the easiest way to correct illumination effects.
Digital plant phenotyping is emerging as a key research domain at the interface of information technology and plant science. Digital phenotyping aims to deploy high-end non-destructive sensing techniques and information technology infrastructures to automate the extraction of both structural and physiological traits from plants under phenotyping experiments. One of the promising sensor technologies for plant phenotyping is hyperspectral imaging (HSI). The main benefit of utilising HSI compared to other imaging techniques is the possibility to extract simultaneously structural and physiological information on plants. The use of HSI for analysis of parts of plants, e.g. plucked leaves, has already been demonstrated. However, there are several significant challenges associated with the use of HSI for extraction of information from a whole plant, and hence this is an active area of research. These challenges are related to data processing after image acquisition. The hyperspectral data acquired of a plant suffers from variations in illumination owing to light scattering, shadowing of plant parts, multiple scattering and a complex combination of scattering and shadowing. The extent of these effects depends on the type of plants and their complex geometry. A range of approaches has been introduced to deal with these effects, however, no concrete approach is yet ready. In this article, we provide a comprehensive review of recent studies of close-range HSI of whole plants. Several studies have used HSI for plant analysis but were limited to imaging of leaves, which is considerably more straightforward than imaging of the whole plant, and thus do not relate to digital phenotyping. In this article, we discuss and compare the approaches used to deal with the effects of variation in illumination, which are an issue for imaging of whole plants. Furthermore, future possibilities to deal with these effects are also highlighted.
The increasing need to develop a rapid understanding of plant functional dynamics has led to the employment of sensor technology for non-destructive assessment of plants. Hyperspectral Imaging (HSI) ...being an integration of two modalities, imaging and point spectroscopy, is nowadays emerging as a potential tool for rapid, non-destructive and automated close range assessment of plants functional dynamics both in terms of structure and physiology.
Firstly, this paper presents an overview of some basic concepts of close range HSI on plants, concerning the plant–light interaction, instrumental setup, and spectral data analysis. Furthermore, the work reviews recent advances of HSI for plant related studies under controlled experimental conditions as well as in natural agricultural settings. Applications are discussed on foliar content estimation, variety identification, growth monitoring, stress and disease-related studies, phenotyping and adoption of HSI in high-throughput phenotyping platforms (HTPPs).
Close range HSI is a challenging task and suffers from technical complexities related to external factors (e.g. illumination effects) as well as plant-related factors (e.g. complex plant geometry). The paper finally discusses some of the technical challenges related to the implementation of HSI in the close range assessment of plant traits.
•HSI is an automated, non-destructive and rapid alternative to explore plant traits.•It acquires both chemical and structural information related with plant traits.•Phenotyping, disease, species identification, foliar chemistry estimation.•Illumination effects are the major technical challenge to be dealt with.
This work develops a dual-layer energy management (DLEM) model for a microgrid (MG) consisting of a community, distributed energy resources (DERs), and a grid. It ensures the participation of all ...these energy entities of MG in the market and their interaction with each other. The first layer performs the scheduling operation of the community with the goal of minimizing its net-billing cost and sends the obtained schedule to the DER operator and grid. Further, the second layer formulates a power scheduling algorithm (PSA) to minimize the net-operating cost of DERs and takes into account the load demand requested by the community operator (COR). This PSA aims to achieve optimal operation of MG by considering solar PV power, requested demand, per unit grid energy prices, and state of charge of the battery energy storage system of the DER layer. Moreover, to study the impact of electric vehicles (EVs) load programs on DLEM, the advanced probabilistic EV load profile model is developed considering practical and uncertain events. The EV load is modelled for grid to vehicle mode, and a new mode of EV operation is introduced, i.e., vehicle to grid with EV demand response strategy (V2G_DRS) mode. The solar PV and load demand data are obtained from the MG setup installed and buildings present at the university campus. However, a scenario reduction technique is used to deal with the uncertainties of the obtained data. In order to evaluate the efficacy of the developed DLEM, its results are compared to previously reported energy management models. The results reveal that DLEM is superior to the existing models as it decreases the net-billing cost of COR by 13% and increases the profit of the DER operator by 17%. Further, it is found that for the highest EV penetration, i.e., 30 EVs, the V2G_DRS mode of EV operation reduces the total energy imported by COR by 11.39% and the net-billing cost of COR by 7.88%. Therefore, it can be concluded that the proposed model with the introduced V2G_DRS mode of EV makes the operation of all the entities of MG more economical and sustainable.
This letter explores geometrical optics-based raytracing in combination with modal analysis to realize an open cavity structure as an advanced variant of the resonant cavity antenna (RCA). Additional ...ray confinement has been ensured by determining accurate phase-locking conditions as demonstrated for the first time. Its combination with the analysis of the cavity mode has also been introduced to determine the optimum design parameters, achieving improved radiation properties. With respect to a reference RCA reported earlier, the proposed approach exhibits a remarkable improvement in gain by 4-9 dB, resulting in about 17 dBi peak value consistently over the full 18% matching bandwidth. This is resultant from an increase in aperture efficiency, typically from 27% to 74%. This is achieved without any compromise with the cross-polarization property. The sidelobe level improves over the band except in higher frequency in H-plane. The proposed concept is commercially viable, showing significantly advantageous features.
Visible and near-infrared (Vis-NIR) spectral imaging is appearing as a potential tool to support high-throughput digital agricultural plant phenotyping. One of the uses of spectral imaging is to ...predict non-destructively the chemical constituents in the plants such as nitrogen content which can be related to the functional status of plants. However, before using high-throughput spectral imaging, it requires extensive calibration, just as needed for any other spectral sensor. Calibrating the high-throughput spectral imaging setup can be a challenging task as the resources needed to run experiments in high-throughput setups are far more than performing measurements with point spectrometers. Hence, to supply a resource-efficient approach to calibrate spectral cameras integrated with high-throughput plant phenotyping setups, this study proposes the use of chemometric calibration transfer (CT) and model update. The main idea was to use a point spectrometer to develop the primary model and transfer it to the spectral cameras integrated into the high-throughput setups. The potential of the approach was showed using a real Vis-NIR dataset related to nitrogen prediction in wheat plants measured with point spectrometer, tabletop spectral cameras and spectral cameras integrated with a high-throughput plant phenotyping setup. For CT and model update, direct standardization and parameter-free calibration enhancement approaches were explored. A key aim of this study was to only use and compare techniques that does not require any further optimization as they can be easily implemented by the plant biologist in future applications. The proposed approach based on the transfer of point spectroscopy models to spectral cameras in a high-throughput setup can allow spectral calibrations to be sharable and widely applicable, thus helping the global digital plant phenotyping community.
Display omitted
•A new strategy to calibrate high-throughput spectral imaging setup was proposed.•Calibration transfer from point spectrometer to spectral cameras was demonstrated.•Standard-based and standard-free calibration techniques were explored.•A case of nitrogen prediction in wheat leaves was presented.
Multiblock data sets and modeling techniques are widely encountered in the chemometric community. Although the currently available techniques, such as sequential orthogonalized partial least squares ...(SO-PLS) regression are mainly focused on the prediction of a single response and deal with the multiple response(s) case using PLS2 type approach. Recently, a new approach called canonical PLS (CPLS) was proposed for extracting the subspaces efficiently for multiple response(s) cases, supporting both regression and classification. 'Efficiently' here means more information in fewer latent variables. This work suggests a combination of SO-PLS and CPLS, sequential orthogonalized canonical partial least squares (SO-CPLS), to model multiple response(s) for multiblock data sets. The cases of SO-CPLS for modeling multiple response(s) regression and classification were demonstrated on several data sets. Also, the capability of SO-CPLS to incorporate meta-information related to samples for efficient subspace extraction is demonstrated. Furthermore, a comparison with the commonly used sequential modeling technique, called sequential orthogonalized partial least squares (SO-PLS), is also presented. The SO-CPLS approach can benefit both the multiple response(s) regression and classification modeling and can be of high importance when meta-information such as experimental design or sample classes is available.
Display omitted
•The multiblock method combining sequential modelling and canonical PLS is evaluated.•The method handles multiple responses more efficiently than PLS2.•The method allows using meta information to improve subspace extraction.•The method was tested on wide multiple responses prediction datasets.
An artificial intelligence approach based on deep generative neural networks for spectral imaging processing was proposed. The key idea was to treat different spectral image processing operations ...such as segmentation, regression, and classification as image-to-image translation tasks. For the image-to-image translation, the conditional generative adversarial networks were used. As a baseline comparison, the traditional chemometric approach based on pixels wise modelling was demonstrated. The analysis was presented with two real data sets related to fruit property prediction and kernel and shell classification of walnuts. The presented artificial intelligence approach for spectral image processing can provide benefits for any field of science where spectral imaging and processing is widely performed.
Display omitted
•Image translation was proposed for spectral image processing.•The generative adversarial network was used for image translation.•The image translation uses both spatial and spectral information.•The applications such as segmentation, regression and classification are demonstrated.
Modeling near‐infrared (NIR) spectral data to predict fresh fruit properties is a challenging task. The difficulty lies in creating generalized models that can work on fruits of different cultivars, ...seasons, and even multiple commodities of fruit. Due to intrinsic differences in spectral properties, NIR models often fail in testing, resulting in high bias and prediction errors. One current solution for achieving generalized models is to use large calibration sets measured over multiple cultivars and harvest seasons. However, current practice primarily focuses on calibration sets for single fruit commodities, disregarding the rich information available from other fruit commodities. This study aims to demonstrate the potential of locally weighted partial least‐squares an example of just‐in‐time (JIT) modeling to develop real‐time models based on calibration sets consisting of multiple fruit commodities. The study also explores JIT modeling for leveraging relevant information from other fruit commodities or adapting the model based on new samples. The application demonstrated here predicts the dry matter in fresh fruit using portable NIR spectroscopy. The results show that JIT modeling is particularly effective for multiple fruit commodities in a single calibration set. The JIT models achieved a root mean squared error of prediction (RMSEP) of 0.69% fresh weight (FW), while the traditional partial least squares (PLS) modeling RMSEP was 0.93% FW. JIT modeling can be particularly beneficial when the user has multiple fruit datasets and wants to combine them into a single dataset to utilize all the relevant information available.
The study proposes locally weighted PLS regression modeling for multifruit spectral data modeling. Local modeling allows using information in a weighted way from all different fruit types present.
Open resonant cavity antenna (ORCA) and its recent advances promise attractive features and possible applications, although the designs reported so far are solely based on the classical ...electromagnetic (EM) theory and general perception of EM circuits. This work explores machine learning (ML)-assisted antenna design techniques aiming to improve and optimize its major radiation parameters over the maximum achievable operating bandwidth. A state-of-the-art method, e.g., parallel surrogate model-assisted hybrid differential evolution for antenna synthesis (PSADEA) has been exercised upon a reference ORCA geometry revealing a fascinating outcome. This modifies the shape of the cavity which was not predicted by EM-based analysis as well as promising significant improvement in its radiation properties. The PSADEA-generated design has been experimentally verified indicating 3dB-11dB improvement in sidelobe level along with high broadside gain maintained above 17 dBi over the 18.5% impedance bandwidth of the ORCA. The new design has been theoretically interpreted by the theory of geometrical optics (GO). This investigation demonstrates the potential and possibilities of employing artificial intelligence (AI)-based techniques in antenna design where multiple parameters need to be adjusted simultaneously for the best possible performances.