Journal of Spectral Imaging
Latest Publications


TOTAL DOCUMENTS

82
(FIVE YEARS 46)

H-INDEX

6
(FIVE YEARS 3)

Published By Im Publications Open Llp

2040-4565, 2040-4565

Author(s):  
Tatireddy Reddy ◽  
Jonnadula Harikiran

Hyperspectral imaging is used in a wide range of applications. When used in remote sensing, satellites and aircraft are employed to collect the images, which are used in agriculture, environmental monitoring, urban planning and defence. The exact classification of ground features in the images is a significant research issue and is currently receiving greater attention. Moreover, these images have a large spectral dimensionality, which adds computational complexity and affects classification precision. To handle these issues, dimensionality reduction is an essential step that improves the performance of classifiers. In the classification process, several strategies have produced good classification results. Of these, machine learning techniques are the most powerful approaches. As a result, this paper reviews three different types of hyperspectral image machine learning classification methods: cluster analysis, supervised and semi-supervised classification. Moreover, this paper shows the effectiveness of all these techniques for hyperspectral image classification and dimensionality reduction. Furthermore, this review will assist as a reference for future research to improve the classification and dimensionality reduction approaches.


Author(s):  
Mohamed Aref ◽  
Abdallah Hussein ◽  
Abou-Bakr Youssef ◽  
Ibrahim Aboughaleb ◽  
Amr Sharawi ◽  
...  

Thermal ablation modalities, for example radiofrequency ablation (RFA) and microwave ablation, are intended to prompt controlled tumour removal by raising tissue temperature. However, monitoring the size of the resulting tissue damage during the thermal removal procedures is a challenging task. The objective of this study was to evaluate the observation of RFA on an ex vivo liver sample with both a commercial and a low-cost system to distinguish between the normal and the ablated regions as well as the thermally affected regions. RFA trials were conducted on five different ex vivo normal bovine samples and monitored initially by a custom hyperspectral (HS) camera to measure the diffuse reflectance (Rd) utilising a polychromatic light source (tungsten halogen lamp) within the spectral range 348–950 nm. Next, the light source was replaced with monochromatic LEDs (415, 565 and 660 nm) and a commercial charge-coupled device (CCD) camera was used instead of the HS camera. The system algorithm comprises image enhancement (normalisation and moving average filter) and image segmentation with K-means clustering, combining spectral and spatial information to assess the variable responses to polychromatic light and monochromatic LEDs to highlight the differences in the Rd properties of thermally affected/normal tissue regions. The measured spectral signatures of the various regions, besides the calculation of the standard deviations (δ) between the generated six groups, guided us to select three optimal wavelengths (420, 540 and 660 nm) to discriminate between these various regions. Next, we selected six spectral images to apply the image processing to (at 450, 500, 550, 600, 650 and 700 nm). We noticed that the optimum image is the superimposed spectral images at 550, 600, 650 and 700 nm, which are capable of discriminating between the various regions. Later, we measured Rd with the CCD camera and commercially available monochromatic LED light sources at 415, 565 and 660 nm. Compared to the HS camera results, this system was more capable of identifying the ablated and the thermally affected regions of surface RFA than the side-penetration RFA of the investigated ex vivo liver samples. However, we succeeded in developing a low-cost system that provides satisfactory information to highlight the ablated and thermally affected region to improve the outcome of surgical tumour ablation with much shorter time for image capture and processing compared to the HS system.


Author(s):  
Austin Hayes ◽  
T. David Reed

Flue-cured tobacco (Nicotiana tabacum L.) is a high value-per-acre crop that is intensively managed to optimise the yield of high-quality cured leaf. A 15-day study assessed the potential of hyperspectral reflectance data for detecting Phytophthora nicotianae (black shank) incidence in flue-cured tobacco. Hyperspectral reflectance data were taken from a commercial flue-cured tobacco field with a progressing black shank infestation. The effort encompassed two key objectives. First, develop hyperspectral indices and/or machine learning classification models capable of detecting Phytophthora nicotianae (black shank) incidence in flue-cured tobacco. Second, evaluate the model’s ability to separate pre-symptomatic plants from healthy plants. Two hyperspectral indices were developed to detect black shank incidence based on differences in the spectral profiles of asymptomatic flue-cured tobacco plants compared to those with black shank symptoms. While one of the indices is a broad-band index and the other uses narrow wavelength values, the statistical difference between the two indices was not significant and both provided an accurate classification of symptomatic plants. Further analysis of the indices showed significant differences between the index values of healthy and symptomatic plants (α = 0.05). In addition, the indices were able to detect black shank symptoms pre-symptomatically (α = 0.09). Subspace linear discriminant analysis, a machine learning classification, was also used for prediction of black shank incidence with up to 85.7% classification accuracy. The implications of using either spectral indices or machine learning for classification for future black shank research are discussed.


Author(s):  
Binu Devassy ◽  
Sony George

Firmness is one of the most important quality measures of strawberries, and is related to other aspects of the fruit, such as flavour, ripeness and internal characteristics. The most popular method for measuring firmness is puncturing with a penetrometer, which is destructive and time-consuming. In the present study, we make an attempt to predict the firmness of strawberries in a fast, non-destructive and non-contact way using hyperspectral imaging (HSI) and data analysis with various regression techniques. The primary goal of this research is to investigate and compare the firmness prediction capability of seven prominent regression techniques. We have performed HSI data acquisition of 150 strawberries and optimised seven regression models using the spectral information to predict strawberry firmness. These models are linear, ridge, lasso, k-neighbours, random forest, support vector and partial least square regression. The res ults show that HSI data with regression models has the potential to predict firmness in a rapid, non-destructive manner. Out of these seven regression models, the k-neighbours regression model outperformed all other methods with a standard error of prediction of 0.14, which is better than that of the state-of-the-art results.


Author(s):  
Cara Murphy ◽  
John Kerekes

The classification of trace chemical residues through active spectroscopic sensing is challenging due to the lack of physics-based models that can accurately predict spectra. To overcome this challenge, we leveraged the field of domain adaptation to translate data from the simulated to the measured domain for training a classifier. We developed the first 1D conditional generative adversarial network (GAN) to perform spectrum-to-spectrum translation of reflectance signatures. We applied the 1D conditional GAN to a library of simulated spectra and quantified the improvement in classification accuracy on real data using the translated spectra for training the classifier. Using the GAN-translated library, the average classification accuracy increased from 0.622 to 0.723 on real chemical reflectance data, including data from chemicals not included in the GAN training set.


Author(s):  
Ramy Abdlaty ◽  
Qiyin Fang

It is highly demanding to identify healthy and non-healthy species in a heterogeneous environment such as human tissues. In such a case, one identifier, such as a spectral fingerprint, might be inadequate. Therefore, additional identification is required, for instance, a polarisation measurement. In view of that, the development of a spectropolarimeter that captures two cross-polarised arrays of spectral images is a key requirement. To meet this requirement, an imager optical setup has been designed to provide spatial, spectral and polarisation preference information for species that exist in a heterogeneous environment, such as in medical tissue samples. The spectral and polarisation information is obtained employing an acousto-optic tunable filter and a polarising beam splitter, respectively. The optical imager is designed to operate in the visible-near infrared range (450–850 nm) with a spectral resolution of 3 nm. The spectropolarimeter design along with optical characterisation results are reported.


Author(s):  
Jun-Li Xu ◽  
Cecilia Riccioli ◽  
Ana Herrero-Langreo ◽  
Aoife Gowen

Deep learning (DL) has recently achieved considerable successes in a wide range of applications, such as speech recognition, machine translation and visual recognition. This tutorial provides guidelines and useful strategies to apply DL techniques to address pixel-wise classification of spectral images. A one-dimensional convolutional neural network (1-D CNN) is used to extract features from the spectral domain, which are subsequently used for classification. In contrast to conventional classification methods for spectral images that examine primarily the spectral context, a three-dimensional (3-D) CNN is applied to simultaneously extract spatial and spectral features to enhance classificationaccuracy. This tutorial paper explains, in a stepwise manner, how to develop 1-D CNN and 3-D CNN models to discriminate spectral imaging data in a food authenticity context. The example image data provided consists of three varieties of puffed cereals imaged in the NIR range (943–1643 nm). The tutorial is presented in the MATLAB environment and scripts and dataset used are provided. Starting from spectral image pre-processing (background removal and spectral pre-treatment), the typical steps encountered in development of CNN models are presented. The example dataset provided demonstrates that deep learning approaches can increase classification accuracy compared to conventional approaches, increasing the accuracy of the model tested on an independent image from 92.33 % using partial least squares-discriminant analysis to 99.4 % using 3-CNN model at pixel level. The paper concludes with a discussion on the challenges and suggestions in the application of DL techniques for spectral image classification.


Author(s):  
Yao Alvarez Kossonou ◽  
Jérémie Zoueu

In this paper, we present the progress made in developing multimodal and multispectral light microscopy for label-free malaria diagnosis. Our previously developed light emitting diode (LED) illumination system was replaced by laser diodes as light sources in order to narrow the spectral bands and improve the effectiveness of the contrast function for infected blood cell detection. The acquisition system is now equipped with an algorithm for automatic field scanning and best in-focus determination. We demonstrate the potential of this platform to provide multiple investigation modalities like transmission, reflection, scattering, fluorescence, excitation, emission and polarisation. The application of this platform on malaria-infected samples has shown the effectiveness of such a system in label-free and all-optical malaria detection by allowing the possibility of using a different type of imaging set-up for the samples analysed. Also, fewer illumination sources are used to characterise malaria-infected red blood cells compared to our previous works on malaria detection using LEDs illumination sources.


Author(s):  
Mohammad Al Ktash ◽  
Otto Hauler ◽  
Edwin Ostertag ◽  
Marc Brecht

Different types of raw cotton were investigated by a commercial ultraviolet-visible/near infrared (UV-Vis/NIR) spectrometer (210–2200 nm) as well as on a home-built setup for NIR hyperspectral imaging (NIR-HSI) in the range 1100–2200 nm. UV-Vis/NIR reflection spectroscopy reveals the dominant role proteins, hydrocarbons and hydroxyl groups play in the structure of cotton. NIR-HSI shows a similar result. Experimentally obtained data in combination with principal component analysis (PCA) provides a general differentiation of different cotton types. For UV-Vis/NIR spectroscopy, the first two principal components (PC) represent 82 % and 78 % of the total data variance for the UV-Vis and NIR regions, respectively. Whereas, for NIR-HSI, due to the large amount of data acquired, two methodologies for data processing were applied in low and high lateral resolution. In the first method, the average of the spectra from one sample was calculated and in the second method the spectra of each pixel were used. Both methods are able to explain ≥90 % of total variance by the first two PCs. The results show that it is possible to distinguish between different cotton types based on a few selected wavelength ranges. The combination of HSI and multivariate data analysis has a strong potential in industrial applications due to its short acquisition time and low-cost development. This study opens a novel possibility for a further development of this technique towards real large-scale processes.


Author(s):  
Carolina Blanch-Pérez del Notario ◽  
Carlos López-Molina ◽  
Andy Lambrechts ◽  
Wouter Saeys

The discrimination power of a hyperspectral imaging system for image segmentation or object detection is determined by the illumination, the camera spatial–spectral resolution, and both the pre-processing and analysis methods used for image processing. In this study, we methodically reviewed the alternatives for each of those factors for a case study from the food industry to provide guidance in the construction and configuration of hyperspectral imaging systems in the visible near infrared range for food quality inspection. We investigated both halogen- and LED-based illuminations and considered cameras with different spatial–spectral resolution trade-offs. At the level of the data analysis, we evaluated the impact of binning, median filtering and bilateral filtering as pre- or post-processing and compared pixel-based classifiers with convolutional neural networks for a challenging application in the food industry, namely ingredient identification in a flour–seed mix. Starting from a basic configuration and by modifying the combination of system aspects we were able to increase the mean accuracy by at least 25 %. In addition, different trade-offs in performance-complexity were identified for different combinations of system parameters, allowing adaptation to diverse application requirements.


Sign in / Sign up

Export Citation Format

Share Document