Share Email Print
cover

Proceedings Paper

Development of the marine-aggregated-particle profiling and enumerating rover
Author(s): David K. Costello; Kendall L. Carder; Robert G. Steward
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The in situ imaging of marine particles shares many of the signal-to-noise difficulties of underwater imaging in general. Natural and traditional artificial illumination, for example, allow light scattered from particles outside the imaging volume, reducing the image contrast. Sizing and classification of small-particle images (magnification approaching 1 or more) have additional difficulties associated with a limited depth-of-field and the resulting noise from illuminated but unfocused targets in the field of view. Moreover, target sizing and classification are uncertain without individual target range information. The new marine particle imaging instrument to be discussed employs diode laser illumination (675 nm) with line-generator optics to produce a thin light sheet at the system focal plane. This light sheet and narrow-band, optical filters are utilized to minimize noise associated with diffuse ambient illumination since significant red ambient illumination is lost below 5 m depth. It also removes the uncertainty involved in the determination of the three-dimensional position and size of a target in a two-dimensional image. An additional problem inherent in marine particle research is that the size of the particles of interest ranges over several orders of magnitude (micrometers to centimeters diameter). The instrument addresses this problem of scale with coincident video imaging systems of high and low spatial resolution. Shape-generated feature vectors and particle optical attributes are extracted from digitized particle images and utilized in an automatic particle classification scheme. The strategy is multidimensional and incorporates a pattern recognition algorithm rooted in the theory of moment invariants.

Paper Details

Date Published: 1 December 1991
PDF: 12 pages
Proc. SPIE 1537, Underwater Imaging, Photography, and Visibility, (1 December 1991); doi: 10.1117/12.48881
Show Author Affiliations
David K. Costello, Univ. of South Florida (United States)
Kendall L. Carder, Univ. of South Florida (United States)
Robert G. Steward, Univ. of South Florida (United States)


Published in SPIE Proceedings Vol. 1537:
Underwater Imaging, Photography, and Visibility
Richard W. Spinrad, Editor(s)

© SPIE. Terms of Use
Back to Top