Article

Multipurpose spectral imager

The University Courses on Svalbard, 9170 Longyearbyen, Norway.
Applied Optics (Impact Factor: 1.78). 07/2000; 39(18):3143-53. DOI: 10.1364/AO.39.003143
Source: PubMed

ABSTRACT A small spectral imaging system is presented that images static or moving objects simultaneously as a function of wavelength. The main physical principle is outlined and demonstrated. The instrument is capable of resolving both spectral and spatial information from targets throughout the entire visible region. The spectral domain has a bandpass of 12 A. One can achieve the spatial domain by rotating the system's front mirror with a high-resolution stepper motor. The spatial resolution range from millimeters to several meters depends mainly on the front optics used and whether the target is fixed (static) or movable relative to the instrument. Different applications and examples are explored, including outdoor landscapes, industrial fish-related targets, and ground-level objects observed in the more traditional way from an airborne carrier (remote sensing). Through the examples, we found that the instrument correctly classifies whether a shrimp is peeled and whether it can disclose the spectral and spatial microcharacteristics of targets such as a fish nematode (parasite). In the macroregime, we were able to distinguish a marine vessel from the surrounding sea and sky. A study of the directional spectral albedo from clouds, mountains, snow cover, and vegetation has also been included. With the airborne experiment, the imager successfully classified snow cover, leads, and new and rafted ice, as seen from 10.000 ft (3.048 m).

0 Followers
 · 
96 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Reflection spectra obtained from hyperspectral imaging can be used as a bio-optical taxonomic identification tool if the pigment composition and the corresponding optical absorption signatures of an organism are known. In this study we elucidate species-specific absorption and corresponding reflection signatures of marine organisms and discuss optical fingerprints from underwater hyperspectral imaging (UHI) for future automated identification of organisms on the seafloor. When mounted on underwater robots, UHI has the potential to be a time- and cost-efficient identification and mapping method covering large areas over a short time. Hyperspectral imaging in vivo and in situ were used to obtain species-specific reflection signatures (optical fingerprints). High performance liquid chromatography, liquid chromatography-mass spectroscopy and nuclear magnetic resonance were used for pigment identification and to obtain species-specific absorption signatures of four marine benthic species; the spoonworm Bonellia viridis, and the sponges Isodictya palmata, Hymedesmia paupertas and Hymedesima sp. Species-specific optical fingerprints based on a UHI-based reflectance signature were verified successfully in the organisms investigated.
    Organisms Diversity & Evolution 06/2014; 14(2):237-246. DOI:10.1007/s13127-013-0163-1 · 3.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We discuss a new versatile setup for goniometric measurements of spectral radiances with two modes of operation: (1) it can operate as a 2-D goniometer for measurements in a horizontal plane of the singly scattered radiance from particles in suspension and (2) it can be used as a 3-D goniometer for measuring spectral radiances over an entire hemisphere. In our setup, various kinds of light sources and detectors can easily be inserted. Among the detectors, a spectral imager is designed and used. Proper hardware and software is chosen so as to make our setup fully automated and easy to operate. We present results from two different investigations to demonstrate the utilization of our setup. The first investigation is concerned with measurements of the volume scattering function (VSF) over a large forward and backward angular range. Our experimental results for the VSF show good agreement with theoretical simulations. We also use our setup to obtain a series of 1-D angular spectral images of the skin on the dorsal side of a human hand in vivo by employing various illumination angles. Our setup provides a robust, highly automated, and flexible framework for carrying out goniometric measurements in a variety of applications.
    Optical Engineering 05/2006; 45(5). DOI:10.1117/1.2205287 · 0.96 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Automatic fish fillet inspection, i.e. detection of flaws and irregularities, is often done on fast moving conveyer belts using line scan imaging. Recent work shows that imaging spectroscopy is a promising method for automatic fish fillet inspection. This requires fast algorithms that can handle large amount of data and make reliable decisions in fractions of a second. One important step in these algorithms is segmentation, where different regions in the image are labelled. For cod fillets, this can be to identify which part of the fillet belongs to the loin, belly flap, centre cut and tail. How severe a detected flaw is depends on which part of the fillet it is located in. Segmentation requires a robust spatial reference system which is invariant to rotation and warping of the fillet. The centreline, consisting of veins and arteries cut off during filleting, is always visible on cod fillets and hence a good reference for segmentation. We show how to enhance the centreline by using the absorption characteristics of haemoglobin, and how a novel ridge detection method can detect the centreline in cod fillets. The results show that the centreline can be detected with an average accuracy of 1 mm from the tail and 77% into the fillet relative to its total length. The average error increases rapidly in the neck region and typical errors of 4 mm is reported.
    Journal of Food Engineering 02/2009; DOI:10.1016/j.jfoodeng.2008.06.035 · 2.58 Impact Factor