Dispersive Spectrometer Design and Analysis

Passive, hyperspectral image data and bathymetric lidar data are complimentary data types that can be used effectively in tandem. Hyperspectral data contain information related to water quality, depth, and bottom type; and bathymetric lidar data contain precise information about the depth of the water and qualitative information about water quality and bottom reflectance. The two systems together provide constraints on each other. For example, lidar-derived depths can be used to constrain spectral radiative transfer models for hyperspectral data, which allows for the estimation of bottom reflectance for each pixel. Similarly, depths can be used to calibrate models, which permit the estimation of depths from the hyperspectral data cube on the raster defined by the spectral imagery. We demonstrate these capabilities by fusing hyperspectral data from the LASH and AVIRIS spectrometers with depth data from the SHOALS bathymetric laser to achieve bottom classification and increase the density of depth measurements in Kaneohe Bay, Hawaii. These capabilities are envisioned as operating modes of the next-generation SHOALS system, CHARTS, which will deploy a bathymetric laser and spectrometer on the same platform.

Recently, a joint Swiss/Belgian initiative started a project to build a new generation airborne imaging spectrometer, namely APEX (Airborne Prism Experiment) under the ESA funding scheme named PRODEX. APEX is a dispersive pushbroom imaging spectrometer operating in the spectral range between 380 - 2500 nm. The spectral resolution will be better then 10 nm in the SWIR and < 5 nm in the VNIR range of the solar reflected range of the spectrum. The total FOV will be ± 14 deg, recording 1000 pixels across track with max. 300 spectral bands simultaneously. APEX is subdivided into an industrial team responsible for the optical instrument, the calibration homebase, and the detectors, and a science and operational team, responsible for the processing and archiving of the imaging spectrometer data, as well as for its operation. APEX is in its design phase and the instrument will be operationally available to the user community in the year 2006.

CRISM (Compact Reconnaissance Imaging Spectrometer for Mars) hyperspectral data have a spatial resolution ranging from 12 to 36m/pixel allowing the high resolution mapping of minerals at the surface of Mars. However, the signal-to-noise ratio (SNR) makes challenging the discrimination of minerals spectrally close such as certain phyllosilicates and carbonates. Here, we discuss different processing of data reduction used to improve the signal-to-noise ratio and to highlight the alteration minerals at the surface of Mars and their limit. We show that our tool allows to understand trends in global mineralogy present in hyperspectral data cube.

This study explores the utilization of Binary Embedded Zero Tree Wavelet Algorithms (BEZW) to compress hyperspectral images. The primary goal is to enhance the representation of spectral data while minimizing storage and transmission requirements. The BEZW algorithm employs wavelet transformations to leverage both spectral and spatial redundancies found in hyperspectral data cubes. Its embedded zero tree structure ensures efficient encoding of small coefficients with minimal overhead. The study evaluates the performance of the BEZW method utilizing various hyper-spectral datasets and compression settings, comparing it to other compression techniques. The results illustrate that the BEZW algorithm offers a promising approach to hyperspectral image compression, achieving competitive compression ratios while preserving spectral accuracy. This makes it a valuable option for applications where efficient hyperspectral data storage and transmission are crucial. The research contributes to the field of hyperspectral imaging by introducing an effective compression method that enhances data accessibility, simplifying the utilization of hyperspectral data in resource-constrained environments.

Background clutter and target signatures have traditionally been described by parameters derived form measurements of spatial structure and spectral ratios derived from fixed spectral bandpass images. The advent of hyperspectral imagery requires descriptions of background clutter in a mixed wavelength-spatial or Fourier-transformed (FT) spectral - spatial framework because the data stream may contain simultaneous spatial - spectral, or FT spectral - spatial clutter components. We have developed and tested analytical routines for characterizing the background clutter and target signatures observed by Fourier-transform instruments, without requiring production of a hyperspectral data cube having spectra wavelength and 2D spatial image dimensions. The Kestrel Fourier-transform hyper spectral imager, a Sagnac format interferometer produces a data stream consisting of the Fourier spectra of the background in the in-track focal plane dimension and the spatial information int he cross-track dimension. The temporal data stream thus consists of a time series of frames in FT- spectral vs. spatial dimensions. Spectral wavelength filtering and guard-band subtraction can be accomplished in FT space by binary shift and add algorithms without prior transformation of the data into a hyperspectral data cube. Spatial filtering in the cross track dimension can similarly benefit from efficient binary operations. This paper summarizes some of the target and background clutter characterization algorithms developed and their evaluation against an example atmospheric gas detection scenario.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Summary form only given. This paper presents the current on-going research efforts in which a real-time hyperspectral compression system developed and demonstrated for a military customer is being ported to an embedded platform fit for deployment onto a tactical platform such as an unmanned aerial vehicle (UAV). The original system consists of a PC host containing multiple PCI boards with SHARC processors interfaced to a state-of-the-art hyperspectral image (HSI) sensor. The resulting embedded implementation will leverage a scalable multiprocessing architecture. Processing nodes based on PowerPC processors with AltiVec technology provide the compute power, while the scalable standard RACEway fabric (ANSI/VITA 5-1994) handles the large interprocessor communication bandwidth. The motivation for this effort is derived from the increased interest in fielding hyperspectral sensors in the intelligence, surveillance, and reconnaissance missions of the military. Historically, there has been significant work performed to develop various link systems. Data transmission requirements have grown quickly to whatever capacity was available in the link. With hyperspectral data, this problem becomes even more significant. Sensors such as the EO/IR packages generate large two-dimensional (2-D) sets. There are many standards developed to compress 2-D sets, including the ubiquitous JPEG family of routines. With hyperspectral data, there is now a third dimension contained in the collection, that being the spectral components associated with each spatial pixel element. No longer do 2-D approaches apply efficiently. The data produced by an HSI sensor has correlation components in spatial, temporal, and spectral dimension. The principle component transformation algorithm is one such routine that can work within the cube environment. The results of this port to a deployable, embedded system architecture will be a scalable product that can be integrated into a larger system that may provide actual exploitation either on the unmanned platform or on the ground element. Performance characteristics between the two implementations are compared. An attempt to generalize the parallelism to increase the scalability to any number of available processing elements is a critical objective to increase the utility of this approach. The final product of this work will be the creation of a commercial off-the-shelf (COTS) subsystem that can be leveraged by system developers.

Dimensionality reduction is one of the most important tasks in improving hyperspectral image (HSI) classification performance and has been widely studied. In traditional dimensionality reduction methods, the hyperspectral data cube is first converted to a matrix, and then, the matrix-algebra is employed, in which the spatial structure is usually not taken into consideration. To jointly utilize spectral and spatial information, researches considering HSI as a tensor have attracted more and more attention. In this letter, we propose a group-based tensor model for HIS dimensionality reduction. The local and nonlocal spatial information of HSI cubes is explored by segmenting the original HSI tensors into a lot of small tensors and grouping them into clusters. Finally, the clusters are projected into low-rank subspace to obtain a proper feature space. Experimental results on two real hyperspectral data sets exhibit the effectiveness of the proposed algorithm for HSI-dimensionality reduction.

Invasive plants are causing massive economic and environmental troubles for our societies worldwide. The aim of this study is to employ a set of machine learning classifiers for detecting invasive plant species using remote sensing data. The target species is Kudzu vine, which mostly grows in the south-eastern states of the US and quickly outcompetes other plants, making it a relevant and threatening species to consider. Our study area is Atlanta, Georgia and the surrounding area. Five different algorithms: Boosted Logistic Regression (BLR), Naive Bayes (NB), Neural Network (NN), Random Forest (RF) and Support Vector Machine (SVM) were tested with the aim of testing their performance and identifying the most optimal one. Furthermore, the influence of temporal, spectral and spatial resolution in detecting Kudzu was also tested and reviewed. Our finding shows that random forest, neural network and support vector machine classifiers outperformed. While the achieved internal accuracies were about 97%, an external validation conducted over an expanded area of interest resulted in 79.5% accuracy. Furthermore, the study indicates that high accuracy classification can be achieved using multispectral Sentinel-2 imagery and can be improved while integrating with airborne visible/infrared imaging spectrometer (AVIRIS) hyperspectral data. Finally, this study indicates that dimensionality reduction methods such as principal component analysis (PCA) should be applied cautiously to the hyperspectral AVIRIS data to preserve its utility. The applied approach and the utilized set of methods can be of interest for detecting other kinds of invasive species as part of fulfilling UN sustainable development goals, particularly number 12: responsible consumption and production, 13: climate action, and 15: life on land.

Imaging spectrometer, such as MAIS produces a tremendous volume of image data with up to 5.12 Mbps raw data rate, which needs urgently a real-time, efficient and reversible compression implementation. Between the lossy scheme with high compression ratio and the lossless scheme with high fidelity, we must make our choice based on the particular information content analysis of each imaging spectrometer's image data. In this paper, we present a careful analysis of information-preserving compression of imaging spectrometer MAIS with an entropy and autocorrelation study on the hyperspectral images. First, the statistical information in an actual MAIS image, captured in Marble Bar Australia, is measured with its entropy, conditional entropy, mutual information and autocorrelation coefficients on both spatial dimensions and spectral dimension. With these careful analyses, it is shown that there is high redundancy existing in the spatial dimensions, but the correlation in spectral dimension of the raw images is smaller than expected. The main reason of the nonstationarity on spectral dimension is attributed to the instruments's discrepancy on detector's response and channel's amplification in different spectral bands. To restore its natural correlation, we preprocess the signal in advance. There are two methods to accomplish this requirement: onboard radiation calibration and normalization. A better result can be achieved by the former one. After preprocessing, the spectral correlation increases so high that it contributes much redundancy in addition to spatial correlation. At last, an on-board hardware implementation for the lossless compression is presented with an ideal result.

Lycopene is a major carotenoid in tomatoes and its content varies considerably during postharvest ripening. Hence evaluating lycopene changes can be used to monitor the ripening of tomatoes. Raman chemical imaging technique is promising for mapping constituents of interest in complex food matrices. In this study, a benchtop point-scanning Raman chemical imaging system was developed to evaluate lycopene content in tomatoes at different maturity stages. The system consists of a 785 nm laser, a fiber optic probe, a dispersive imaging spectrometer, a spectroscopic CCD camera, and a two-axis positioning table. Tomato samples at different ripeness stages (i.e., green, breaker, turning, pink, light red, and red) were selected and cut before imaging. Hyperspectral Raman images were acquired from cross sections of the fruits in the wavenumber range of 200 to 2500 cm -1 with a spatial resolution of 1 mm. The Raman spectrum of pure lycopene was measured as reference for spectral matching. A polynomial curve-fitting method was used to correct for the underlying fluorescence background in the Raman spectra of the tomatoes. A hyperspectral image classification method was developed based on spectral information divergence to identify lycopene in the tomatoes. Raman chemical images were created to visualize quantity and spatial distribution of the lycopene at different ripeness stages. The lycopene patterns revealed the mechanism of lycopene generation during the postharvest development of the tomatoes. The method and findings of this study form a basis for the future development of a Raman-based nondestructive approach for monitoring internal maturity of the tomatoes.

We describe a new noninvasive microscopic near infrared reflectance hyperspectral imaging method for visualizing, in vivo, spatially distributed contributions of oxyhemoglobin perfusing the microvasculature within dermal tissue. Microscopic images of the dermis are acquired, generating a series of spectroscopic images formatted as a function of wavelength consisting of one spectral and two spatial dimensions; a hyperspectral image data cube. The data thus collected can be considered as a series of spatially resolved spectra. For data collection, images are acquired by a system consisting of a near infrared liquid crystal tunable filter (LCTF) and a Focal plane array detector (FPA) integrated with a microscope. The LCTF is continuously tunable over a useful near infrared spectral range (650-950 nm) with an average full width at half-height bandwidth of 6.78 nm. To provide high quantum efficiency without etaloning we utilized a back-illumination FPA with deep -depletion technology. A 30W halogen light source illuminates a dermal tissue area of approximately 18 mm in diameter. Reflected light from the dermal tissue is first passed through the microscope, the LCTF, and then imaged onto the FPA. The acquired hyperspectral data is deconvoluted using a multivariate least squares approach that requires at least two reference spectra, oxy- and deoxyhemoglobin. The resulting images are gray scale encoded to directly represent the varying spatial distributions of oxyhemoglobin contribution. As a proof of principle example, we examined a clinical model of vascular occlusion and reperfusion.

Dielectric metasurfaces composed of arrays of subwavelength structures are capable of completely manipulating the phase, amplitude, and polarisation of light. In this Letter, a dielectric metasurface grating imitating the blazing behaviour of saw‐tooth gratings for imaging spectrometer is proposed. The authors present a method for designing the metasurface blazed grating by extending the effective medium theory to the two‐dimensional pattern. The designed metasurface blazed grating consists of periodical unit cells in both orthogonal directions and there contains seven sub‐periods within one unit cell. Duty cycles of each sub‐period in two orthogonal directions are independent, so that the effective index of both TE and TM polarisations can be manipulated simultaneously, and the grating's polarisation property can be optimised. Simulation results show that, for normal incident light, diffraction efficiencies of TE and TM polarisations at reference wavelength 0.7 μm are 79.2 and 79.3%, respectively. Within wavelength range from 0.6 to 0.8 μm, diffraction efficiencies are above 70% and the maximum degree of polarisation is 2.8%. It shows high efficiency and polarisation‐independent diffraction characteristics, and it may find various applications in optical frequencies, especially for quantitative imaging spectrometers.

The demand for safe and edible meat has led to the advancement of freeze-storage techniques, but falsely labeled thawed meat remains an issue. Many methods have been proposed for this purpose, but they all destroy the sample and can only be performed in the laboratory by skilled personnel. In this study, hyperspectral image data were used to construct a machine learning (ML) model to discriminate between freshly refrigerated, long-term refrigerated, and thawed beef meat samples. With four pre-processing methods, a total of five datasets were prepared to construct an ML model. The PLS-DA and SVM techniques were used to construct the models, and the performance was highest for the SVM model applying scatter correction and the RBF kernel function. These results suggest that it is possible to construct a prediction model to distinguish between fresh and non-fresh meat using the spectra obtained by purifying hyperspectral image data cubes, which can be a rapid and non-invasive method for routine analyses of the meat storage state.

To achieve perfect teeth color reproduction, we propose the design of a snapshot hyperspectral imaging spectrometer that can accurately measure the teeth spectrum. In this paper, the optical system of a snapshot hyperspectral dental imaging spectrometer is designed. In particular, to encode images in both the spatial and spectral dimensions, a one-dimensional random coding mask rotated 45 degrees around the optical axis is used. The system has a spectral resolution of approximately 2.2∼10 nm in the 450∼700 nm spectral range, and the total length of the system is less than 150 mm. The whole design can be used as a miniaturized handheld instrument. To verify the feasibility of whiteness measuring, a simulation experiment is carried out. The target spectral information is reconstructed through the reconstruction algorithm. The simulation result shows that spectral imaging has the potential to improve the accuracy and precision of teeth whiteness measurements.

The primary compact high resolution imaging spectrometer was developed and reported. Due to its numerous wave bands the original image data is always in a huge scale and costs a tremendous process overhead, but the data amount of the region of interest is as a rule in the order of thousandth, if not less, of that of the whole push-broom region. With a digital micromirror device (DMD), only the region of interesting object is imaged by the imaging spectrometer, which results in a distinct reduction of data quantity and a high data compression ratio. A DMD of high turning rate and residential time adjustability is used as a spatial light modulator to fulfill the object selection function. It is placed after the fore objective and able to reflect the object to either the panchromatic CCD camera channel or the imaging spectrometer channel. The position of the object can be firstly determined through the image interpretation from panchromatic imaging channel and a DMD control command is executed to switch the corresponding mirrors to the imaging spectrometer channel, thus only the object region of interest is imaged by the spectrometer. The multiple objects of both printed patterns and real leaves are accurately determined and selected according to their different locality and shape features. The panchromatic and hyperspectral image data are both collected for further effective object recognition.