Maturity determination of single maize seed by using near-infrared hyperspectral imaging coupled with comparative analysis of multiple classification models

2021 ◽  
Vol 112 ◽  
pp. 103596
Author(s):  
Zheli Wang ◽  
Xi Tian ◽  
Shuxiang Fan ◽  
Chi Zhang ◽  
Jiangbo Li
Keyword(s):  
Comparative Analysis ◽  
Hyperspectral Imaging ◽  
Near Infrared ◽  
Classification Models ◽  
Maize Seed ◽  
Multiple Classification

scholarly journals Comparison of spectral selection methods in the development of classification models from visible near infrared hyperspectral imaging data

2019 ◽  
Author(s):  
Aoife Gowen ◽  
Jun-Li Xu ◽  
Ana Herrero-Langreo
Keyword(s):  
Hyperspectral Imaging ◽  
Near Infrared ◽  
Spatial Information ◽  
Hyperspectral Image ◽  
Imaging Spectroscopy ◽  
Model Performance ◽  
Data Sampling ◽  
Classification Models ◽  
Imaging Data ◽  
Selection Of

Applications of hyperspectral imaging (HSI) to the quantitative and qualitative measurement of samples have grown widely in recent years, due mainly to the improved performance and lower cost of imaging spectroscopy instrumentation. Data sampling is a crucial yet often overlooked step in hyperspectral image analysis, which impacts the subsequent results and their interpretation. In the selection of pixel spectra for the calibration of classification models, the spatial information in HSI data can be exploited. In this paper, a variety of sampling strategies for selection of pixel spectra are presented, exemplified through five case studies. The strategies are compared in terms of the proportion of global variability captured, practicality and predictive model performance. The use of variographic analysis as a guide to the spatial segmentation prior to sampling leads to the selection of representative subsets while reducing the variation in model performance parameters over repeated random selection.

scholarly journals Hyperspectral imaging of the degradation of meat and comparison with necrotic tissue in human wounds

2019 ◽  
Author(s):  
Amadeus Holmer ◽  
Christoph Homberger ◽  
Thomas Wild ◽  
Frank Siemers
Keyword(s):  
Hyperspectral Imaging ◽  
Near Infrared ◽  
Objective Evaluation ◽  
Infrared Range ◽  
Spectral Features ◽  
Necrotic Tissue ◽  
Tissue Discrimination ◽  
Near Infrared Range ◽  
Imaging Setup

The objective evaluation of scattering tissue and the discrimination of tissue types is an issue that cannot be solved with colour cameras and image processing alone in many cases. Examples can be found in the determination of freshness and ageing of meat, and the discrimination of tissue types in food technology. In medical applications tissue discrimination is also an issue, e.g. in wound diagnostics. A novel hyperspectral imaging setup with powerful signal analysis algorithms is presented which is capable of addressing these topics. The spectral approach allows the chemical analysis of material and tissues and the measurement of their temporal change. We present a method of hyperspectral imaging in the visible-near infrared range which allows both the separation and spatial allocation of different tissue types in a sample, as well as the temporal changes of the tissue as an effect of ageing. To prove the capability of the method, the ageing of meat (slices of pork) was measured and, as a medical example, the application of the hyperspectral imaging setup for the recording of wound tissue is presented. The method shows the ability to discriminate the different tissue components of pork meat, and the ageing of the meat is observable as changes in spectral features. An additional result of our study is the fact that some spectral features, which seem to be typical for the ageing of the meat, are similar to those observed in the necrotic tissue from wound diagnostics in medicine.

NIR hyperspectral imaging (NIR-HI) and μXRD for determination of the transition between juvenile and mature wood of Pinus sylvestris L.

Holzforschung ◽  
2019 ◽  
Vol 73 (7) ◽  
pp. 621-627 ◽  
Author(s):  
Antonio Ruano ◽  
Andreas Zitek ◽  
Barbara Hinterstoisser ◽  
Eva Hermoso
Keyword(s):  
Pinus Sylvestris ◽  
Hyperspectral Imaging ◽  
Transition Point ◽  
Near Infrared ◽  
Juvenile Wood ◽  
Principal Component ◽  
Unsupervised Classification ◽  
Mature Wood ◽  
X Ray

AbstractThe ratio of juvenile wood (jW) to mature wood (mW) is relevant for structural wood applications because of their different properties. Near infrared-hyperspectral imaging (NIR-HI) indicates after calibration, the spatial distribution of jW and mW, and this approach is less time consuming than the established micro X-ray densitometry (μXRD). In the present study, a comparative detection of the jW and mW ofPinus sylvestrisL. was performed by NIR-HI and μXRD and the NIR-HI results were evaluated in combination with three chemometric approaches, namely, the principal component analysis (PCA), partitional k-means unsupervised classification (p-k-mUC) and partial least squares discriminant analysis (PLS-DA) in the range of 900–1700 nm. The best NIR-HI results can be obtained when the transition point of earlywood (EW) and latewood (LW) are assessed separately by PLS-DA. The presented results are useful for an automating data evaluation and simplified data collection.

Sign in / Sign up

Export Citation Format

Share Document