Common Latent Space Exploration for Calibration Transfer across Hyperspectral Imaging-Based Phenotyping Systems

Hyperspectral imaging has increasingly been used in high-throughput plant phenotyping systems. Rapid advancement in the field of phenotyping has resulted in a wide array of hyperspectral imaging systems. However, sharing the plant feature prediction models between different phenotyping facilities becomes challenging due to the differences in imaging environments and imaging sensors. Calibration transfer between imaging facilities is crucially important to cope with such changes. Spectral space adjustment methods including direct standardization (DS), its variants (PDS, DPDS) and spectral scale transformation (SST) require the standard samples to be imaged in different facilities. However, in real-world scenarios, imaging the standard samples is practically unattractive. Therefore, in this study, we presented three methods (TCA, c-PCA, and di-PLSR) to transfer the calibration models without requiring the standard samples. In order to compare the performance of proposed approaches, maize plants were imaged in two greenhouse-based HTPP systems using two pushbroom-style hyperspectral cameras covering the visible near-infrared range. We tested the proposed methods to transfer nitrogen content (N) and relative water content (RWC) calibration models. The results showed that prediction R2 increased by up to 14.50% and 42.20%, while the reduction in RMSEv was up to 74.49% and 76.72% for RWC and N, respectively. The di-PLSR achieved the best results for almost all the datasets included in this study, with TCA being second. The performance of c-PCA was not at par with the di-PLSR and TCA. Our results showed that the di-PLSR helped to recover the performance of RWC, and N models plummeted due to the differences originating from new imaging systems (sensor type, spectrograph, lens system, spatial resolution, spectral resolution, field of view, bit-depth, frame rate, and exposure time) or lighting conditions. The proposed approaches can alleviate the requirement of developing a new calibration model for a new phenotyping facility or to resort to the spectral space adjustment using the standard samples.

Download Full-text

Calibration transfer across multiple hyperspectral imaging-based plant phenotyping systems: I – Spectral space adjustment

Computers and Electronics in Agriculture ◽

10.1016/j.compag.2020.105685 ◽

2020 ◽

Vol 176 ◽

pp. 105685 ◽

Cited By ~ 1

Author(s):

Tanzeel U. Rehman ◽

Libo Zhang ◽

Dongdong Ma ◽

Liangju Wang ◽

Jian Jin

Keyword(s):

Hyperspectral Imaging ◽

Plant Phenotyping ◽

Spectral Space ◽

Calibration Transfer

Download Full-text

PLS Subspace-Based Calibration Transfer for Near-Infrared Spectroscopy Quantitative Analysis

Molecules ◽

10.3390/molecules24071289 ◽

2019 ◽

Vol 24 (7) ◽

pp. 1289 ◽

Cited By ~ 3

Author(s):

Yuhui Zhao ◽

Jinlong Yu ◽

Peng Shan ◽

Ziheng Zhao ◽

Xueying Jiang ◽

...

Keyword(s):

Least Squares ◽

Near Infrared ◽

Least Squares Method ◽

Ordinary Least Squares ◽

Rank Test ◽

Transfer Model ◽

Calibration Model ◽

Least Squares Regression ◽

Standard Samples ◽

Calibration Transfer

In order to enable the calibration model to be effectively transferred among multiple instruments and correct the differences between the spectra measured by different instruments, a new feature transfer model based on partial least squares regression (PLS) subspace (PLSCT) is proposed in this paper. Firstly, the PLS model of the master instrument is built, meanwhile a PLS subspace is constructed by the feature vectors. Then the master spectra and the slave spectra are projected into the PLS subspace, and the features of the spectra are also extracted at the same time. In the subspace, the pseudo predicted feature of the slave spectra is transferred by the ordinary least squares method so that it matches the predicted feature of the master spectra. Finally, a feature transfer relationship model is constructed through the feature transfer of the PLS subspace. This PLS-based subspace transfer provides an efficient method for performing calibration transfer with only a small number of standard samples. The performance of the PLSCT was compared and assessed with slope and bias correction (SBC), piecewise direct standardization (PDS), calibration transfer method based on canonical correlation analysis (CCACT), generalized least squares (GLSW), multiplicative signal correction (MSC) methods in three real datasets, statistically tested by the Wilcoxon signed rank test. The obtained experimental results indicate that PLSCT method based on the PLS subspace is more stable and can acquire more accurate prediction results.

Download Full-text

The Performances of Hyperspectral Sensors for Proximal Sensing of Nitrogen Levels in Wheat

Sensors ◽

10.3390/s20164550 ◽

2020 ◽

Vol 20 (16) ◽

pp. 4550

Author(s):

Huajian Liu ◽

Brooke Bruning ◽

Trevor Garnett ◽

Bettina Berger

Keyword(s):

Hyperspectral Imaging ◽

High Throughput ◽

Management Practices ◽

Leaf Tissue ◽

Plant Phenotyping ◽

Proximal Sensing ◽

N Content ◽

Hyperspectral Sensors ◽

Wheat Leaves ◽

Non Destructive

The accurate and high throughput quantification of nitrogen (N) content in wheat using non-destructive methods is an important step towards identifying wheat lines with high nitrogen use efficiency and informing agronomic management practices. Among various plant phenotyping methods, hyperspectral sensing has shown promise in providing accurate measurements in a fast and non-destructive manner. Past applications have utilised non-imaging instruments, such as spectrometers, while more recent approaches have expanded to hyperspectral cameras operating in different wavelength ranges and at various spectral resolutions. However, despite the success of previous hyperspectral applications, some important research questions regarding hyperspectral sensors with different wavelength centres and bandwidths remain unanswered, limiting wide application of this technology. This study evaluated the capability of hyperspectral imaging and non-imaging sensors to estimate N content in wheat leaves by comparing three hyperspectral cameras and a non-imaging spectrometer. This study answered the following questions: (1) How do hyperspectral sensors with different system setups perform when conducting proximal sensing of N in wheat leaves and what aspects have to be considered for optimal results? (2) What types of photonic detectors are most sensitive to N in wheat leaves? (3) How do the spectral resolutions of different instruments affect N measurement in wheat leaves? (4) What are the key-wavelengths with the highest correlation to N in wheat? Our study demonstrated that hyperspectral imaging systems with satisfactory system setups can be used to conduct proximal sensing of N content in wheat with sufficient accuracy. The proposed approach could reduce the need for chemical analysis of leaf tissue and lead to high-throughput estimation of N in wheat. The methodologies here could also be validated on other plants with different characteristics. The results can provide a reference for users wishing to measure N content at either plant- or leaf-scales using hyperspectral sensors.

Download Full-text

A review of the medical hyperspectral imaging systems and unmixing algorithms’ in biological tissues

Photodiagnosis and Photodynamic Therapy ◽

10.1016/j.pdpdt.2020.102165 ◽

2020 ◽

pp. 102165

Author(s):

Aziz ul Rehman ◽

Shahzad Ahmad Qureshi

Keyword(s):

Hyperspectral Imaging ◽

Biological Tissues ◽

Imaging Systems

Download Full-text

Calibration Transfer between Miniature Photodiode Array-Based Spectrometers in the near Infrared Assessment of Mandarin Soluble Solids Content

Journal of Near Infrared Spectroscopy ◽

10.1255/jnirs.318 ◽

2002 ◽

Vol 10 (1) ◽

pp. 27-35 ◽

Cited By ~ 28

Author(s):

C.V. Greensill ◽

K.B. Walsh

Keyword(s):

Near Infrared ◽

Model Updating ◽

Soluble Solids ◽

Soluble Solid Content ◽

Calibration Model ◽

Calibration Transfer ◽

Orthogonal Signal Correction ◽

Significant Difference ◽

Photodiode Array ◽

Direct Standardisation

The transfer of predictive models among photodiode array based, short wave near infrared spectrometers using the same illumination/detection optical geometry has been attempted using various chemometric techniques, including slope and bias correction (SBC), direct standardisation (DS), piecewise direct standardisation (PDS), double window PDS (DWPDS), orthogonal signal correction (OSC), finite impulse transform (FIR) and wavelet transform (WT). Additionally, an interpolation and photometric response correction method, a wavelength selection method and a model updating method were assessed. Calibration transfer was attempted across two populations of mandarin fruit. Model performance was compared in terms of root mean squared error of prediction ( RMSEP), using Fearn's significance testing, for calibration transfer (standardisation) between pairs of spectrometers from a group of four spectrometers. For example, when a calibration model (Root Mean Square Error of Cross-Validation [ RMSECV = 0.26% soluble solid content (SSC)], developed on one spectrometer, was used with spectral data collected on another spectrometer, a poor prediction resulted ( RMSEP = 2.5% SSC). A modified WT method performed significantly better (e.g. RMSEP = 0.25% SSC) than all other standardisation methods (10 of 12 cases), and almost on a par with model updating (MU) (nine cases with no significant difference, one case and two cases significantly better for WT and MU, respectively).

Download Full-text

Development of Calibration Models to Predict Mean Fibre Diameter in Llama (Lama glama) Fleeces with Near Infrared Spectroscopy

Animals ◽

10.3390/ani11071998 ◽

2021 ◽

Vol 11 (7) ◽

pp. 1998

Author(s):

José Ignacio Amorena ◽

Dolores María Eugenia Álvarez ◽

Elvira Fernández-Ahumada

Keyword(s):

Textile Industry ◽

Near Infrared ◽

Prediction Models ◽

External Validation ◽

Fibre Diameter ◽

Near Infrared Reflectance ◽

Screening Programs ◽

Spectral Signatures ◽

Calibration Models ◽

Cross Validation Error

Llama fibre has the potential to become the most valuable textile resource in the Puna region of Argentina. In this study near infrared reflectance spectroscopy was evaluated to predict the mean fibre diameter in llama fleeces. Analyses between sets of carded and non-carded samples in combination with spectral preprocessing techniques were carried out and a total of 169 spectral signatures of llama samples in Vis and NIR ranges (400–2500 nm) were obtained. Spectral preprocessing consisted in wavelength selection (Vis–NIR, NIR and discrete ranges) and multiplicative and derivative pretreatments; spectra without pretreatments were also included, while modified partial least squares (M-PLS) regression was used to develop prediction models. Predictability was evaluated through R2: standard cross validation error (SECV), external validation error (SEV) and residual predictive value (RPD). A total of 54 calibration models were developed in which the best model (R2 = 0.67; SECV = 1.965; SEV = 2.235 and RPD = 1.91) was obtained in the Vis–NIR range applying the first derivative pretreatment. ANOVA analysis showed differences between carded and non-carded sets and the models obtained could be used in screening programs and contribute to valorisation of llama fibre and sustainable development of textile industry in the Puna territory of Catamarca. The data presented in this paper are a contribution to enhance the scarce information on this subject.

Download Full-text