Home About us Contact
|
Home About us Contact | ||
|
|
||
Hyperspectral Images (hyperspectral + image)
Selected AbstractsHyperspectral imaging combined with principal component analysis for bruise damage detection on white mushrooms (Agaricus bisporus)JOURNAL OF CHEMOMETRICS, Issue 3-4 2008A. A. Gowen Abstract Hyperspectral imaging (HSI) combines conventional imaging and spectroscopy to simultaneously acquire both spatial and spectral information from an object. This technology has recently emerged as a powerful process analytical tool for rapid, non-contact and non-destructive food analysis. In this study, the potential application of HSI for damage detection on the caps of white mushrooms (Agaricus bisporus) was investigated. Mushrooms were damaged by controlled vibration to simulate damage caused by transportation. Hyperspectral images were obtained using a pushbroom line-scanning HSI instrument, operating in the wavelength range of 400,1000,nm with spectroscopic resolution of 5,nm. The effective resolution of the CCD detector was 580,×,580,pixels by 12 bits. Two data reduction methods were investigated: in the first, principal component analysis (PCA) was applied to the hypercube of each sample, and the second PC (PC 2) scores image was used for identification of bruise-damaged regions on the mushroom surface; in the second method PCA was applied to a dataset comprising of average spectra from regions normal and bruise-damaged tissue. In this case it was observed that normal and bruised tissue were separable along the resultant first principal component (PC 1) axis. Multiplying the PC 1 eigenvector by the hypercube data allowed reduction of the hypercube to a 2-D image, which showed maximal contrast between normal and bruise-damaged tissue. The second method performed better than the first when applied to a set of independent mushroom samples. The results from this study could be used for the development of a non-destructive monitoring system for rapid detection of damaged mushrooms on the processing line. Copyright © 2008 John Wiley & Sons, Ltd. [source] 2324: Comparison of algorithms for oximetry in vivo and ex vivoACTA OPHTHALMOLOGICA, Issue 2010D DE BROUWERE Purpose Several authors have proposed a number of algorithms to extract the oxygen saturation in retinal blood vessels based on multispectral image analysis. We evaluated the outcomes of seven known algorithms based on hyperspectral retinal images. Methods Hyperspectral images are acquired using a fundus camera where a slit spectrograph is registered onto a retinal image. This combination compromises both accurate spatial and spectral information over the selected slit. Hyperspectral image analysis was used as input for the oximetry calculations described in the literature. We used a model eye to evaluate the different techniques in a controlled setup. Defibrinated horse blood was perfused through microtubules placed in front of a white (spectralon) background. Oxygen saturation was controlled by mixing different concentrations of sodium dithianate in the blood. Results Oxygen saturation was varied in five equidistant steps between 0 and 1. We correlated the outcomes to the metric of Harvey et al. [Biomed Optics 6631, 2007] Linear correlation with other algorithms resulted in r2 values between 0.881 and 0.985, however we observed a large discrepancy of the slope of each correlation line. The algorithms were also evaluated in images recorded in five healthy volunteers. In all techniques, veins could be separated from arteries by their reduces oxygen saturation, although values varied strongly between the different techniques. Conclusion Our findings confirm the working of a number of noninvasive retinal oximetry algorithms. Different readings can be can be attributed to an offset caused by an uncertainty of pigmentation and scattering parameters in the calibration procedure. [source] Parallel heterogeneous CBIR system for efficient hyperspectral image retrieval using spectral mixture analysisCONCURRENCY AND COMPUTATION: PRACTICE & EXPERIENCE, Issue 9 2010Antonio J. Plaza Abstract The purpose of content-based image retrieval (CBIR) is to retrieve, from real data stored in a database, information that is relevant to a query. In remote sensing applications, the wealth of spectral information provided by latest-generation (hyperspectral) instruments has quickly introduced the need for parallel CBIR systems able to effectively retrieve features of interest from ever-growing data archives. To address this need, this paper develops a new parallel CBIR system that has been specifically designed to be run on heterogeneous networks of computers (HNOCs). These platforms have soon become a standard computing architecture in remote sensing missions due to the distributed nature of data repositories. The proposed heterogeneous system first extracts an image feature vector able to characterize image content with sub-pixel precision using spectral mixture analysis concepts, and then uses the obtained feature as a search reference. The system is validated using a complex hyperspectral image database, and implemented on several networks of workstations and a Beowulf cluster at NASA's Goddard Space Flight Center. Our experimental results indicate that the proposed parallel system can efficiently retrieve hyperspectral images from complex image databases by efficiently adapting to the underlying parallel platform on which it is run, regardless of the heterogeneity in the compute nodes and communication links that form such parallel platform. Copyright © 2009 John Wiley & Sons, Ltd. [source] Weighting hyperspectral image data for improved multivariate curve resolution resultsJOURNAL OF CHEMOMETRICS, Issue 9 2008Howland D. T. Jones Abstract The combination of hyperspectral confocal fluorescence microscopy and multivariate curve resolution (MCR) provides an ideal system for improved quantitative imaging when multiple fluorophores are present. However, the presence of multiple noise sources limits the ability of MCR to accurately extract pure-component spectra when there is high spectral and/or spatial overlap between multiple fluorophores. Previously, MCR results were improved by weighting the spectral images for Poisson-distributed noise, but additional noise sources are often present. We have identified and quantified all the major noise sources in hyperspectral fluorescence images. Two primary noise sources were found: Poisson-distributed noise and detector-read noise. We present methods to quantify detector-read noise variance and to empirically determine the electron multiplying CCD (EMCCD) gain factor required to compute the Poisson noise variance. We have found that properly weighting spectral image data to account for both noise sources improved MCR accuracy. In this paper, we demonstrate three weighting schemes applied to a real hyperspectral corn leaf image and to simulated data based upon this same image. MCR applied to both real and simulated hyperspectral images weighted to compensate for the two major noise sources greatly improved the extracted pure emission spectra and their concentrations relative to MCR with either unweighted or Poisson-only weighted data. Thus, properly identifying and accounting for the major noise sources in hyperspectral images can serve to improve the MCR results. These methods are very general and can be applied to the multivariate analysis of spectral images whenever CCD or EMCCD detectors are used. Copyright © 2008 John Wiley & Sons, Ltd. [source] Hyperspectral NIR image regression part II: dataset preprocessing diagnosticsJOURNAL OF CHEMOMETRICS, Issue 3-4 2006James Burger Abstract When known reference values such as concentrations are available, the spectra from near infrared (NIR) hyperspectral images can be used for building regression models. The sets of spectra must be corrected for errors, transformed to reflectance or absorbance values, and trimmed of bad pixel outliers in order to build robust models and minimize prediction errors. Calibration models can be computed from small (<100) sets of spectra, where each spectrum summarizes an individual image or spatial region of interest (ROI), and used to predict large (>20,000) test sets of spectra. When the distributions of these large populations of predicted values are viewed as histograms they provide mean sample concentrations (peak centers) as well as uniformity (peak widths) and purity (peak shape) information. The same predicted values can also be viewed as concentration maps or images adding spatial information to the uniformity or purity presentations. Estimates of large population statistics enable a new metric for determining the optimal number of model components, based on a combination of global bias and pooled standard deviation values computed from multiple test images or ROIs. Two example datasets are presented: an artificial mixture design of three chemicals with distinct NIR spectra and samples of different cheeses. In some cases it was found that baseline correction by taking first derivatives gave more useful prediction results by reducing optical problems. Other data pretreatments resulted in negligible changes in prediction errors, overshadowed by the variance associated with sample preparation or presentation and other physical phenomena. Copyright © 2007 John Wiley & Sons, Ltd. [source] Detection of Fecal Contamination on Cantaloupes Using Hyperspectral Fluorescence ImageryJOURNAL OF FOOD SCIENCE, Issue 8 2005Angela M. Vargas ABSTRACT To determine whether detection of fecal contamination on cantaloupes is possible using fluorescence imaging, hyperspectral images of cantaloupes artificially contaminated with a range of diluted bovine feces were acquired from 425 to 774 nm in responses to ultraviolet-A (320 to 400 nm) excitation. Evaluation of images at emission peak wavelengths indicated that 675 nm exhibited the greatest contrast between feces contaminated and untreated surface areas. Two-band ratios compared with the single-band images enhanced the contrast between the feces contaminated spots and untreated cantaloupe surfaces. The 595/655-nm, 655/520-nm, and 555/655-nm ratio images provided relatively high detection rates ranging from 79% to 96% across all feces dilutions. However, both single band and ratio methods showed a number of false positives caused by such features as scarred tissues on cantaloupes. Principal component analysis (PCA) was performed using the entire hyperspectral images data; 2nd and 5th principal component (PC) image exhibited differential responses between feces spots and false positives. The combined use of the 2 PC images demonstrated the detection of feces spots (for example, minimum level of 16-,g/mL dry fecal matter) with minimal false positives. Based on the PC weighing coefficients, the dominant wavelengths were 465, 487, 531, 607, 643, and 688 nm. This research demonstrated the potential of multispectral-based fluorescence imaging for online applications for detection of fecal contamination on cantaloupes. [source] | ||||||||||