Automated classification of underwater multispectral imagery for coral reef monitoring

Survey protocols for monitoring the composition and status of coral reef benthic communities vary in the level of detail acquired, but fundamentally follow one of two approaches: 1) a diver identifies organisms in the field, 2) an analyst identifies organisms from underwater imagery (photos or video). Both methods are highly labor intensive and require a trained biologist/geologist. A method for automated classification of reef benthos would improve coral reef monitoring by reducing the cost of data analysis. Spectral classification of standard (three-band) color underwater imagery does not work well for distinguishing major bottom types. Recent publications of hyperspectral reflectance of corals, algae, and sediment, on the other hand, suggest that careful choice of narrow (-10 nm) spectral bands might improve classification accuracy relative to the three wide bands available on commercial cameras. We built an underwater multispectral camera to test whether narrow spectral bands were actually superior to standard RGB cameras for automated classification of underwater images. A filter wheel was used to acquire imagery in six 10 nm spectral bands, which were chosen from suggestions in the literature. Results indicate that the algorithms suggested in the literature require very careful compensation for variable illumination and water column attenuation for even marginal success in classifying underwater imagery. On the other hand, a new algorithm, based on the normalized difference ratio of images at 568 nm and 546 nm can reliably segment photosynthetic organisms (corals and algae) from non-photosynthetic background. Moreover, when this new algorithm is combined with very simple texture segmentation, the general cover classes of coral and algae can be discriminated from the image background with accuracies on the order of 80%. These results suggest that a combination of high spectral resolution and texture-based image segmentation may be an optimal methodology for automated classification of underwater coral reef imagery.

[1]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[2]  Lawrence R. Griffing,et al.  An evaluation of image analysis at benthic sites based on color segmentation , 2001 .

[3]  Heather Holden,et al.  Hyperspectral identification of coral reef features , 1999 .

[4]  Kevin E. Kohler,et al.  Coral Point Count with Excel extensions (CPCe): A Visual Basic program for the determination of coral and substrate coverage using random point count methodology , 2006, Comput. Geosci..

[5]  Emanuele Trucco,et al.  Computer and Robot Vision , 1995 .

[6]  Linda G. Shapiro,et al.  Computer and Robot Vision , 1991 .

[7]  Ellsworth F. LeDrew,et al.  Spectral Discrimination of Healthy and Non-Healthy Corals Based on Cluster Analysis, Principal Components Analysis, and Derivative Spectroscopy , 1998 .

[8]  Andrew J. Nevis,et al.  High‐resolution determination of coral reef bottom cover from multispectral fluorescence laser line scan imagery , 2003 .

[9]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[10]  Kenneth Voss,et al.  Upwelling radiance distribution camera system, NURADS. , 2005, Optics express.

[11]  Curtis D. Mobley,et al.  A Spectrum-Matching and Look-Up-Table Approach to Interpretation of Hyperspectral Remote-Sensing Data , 2004 .

[12]  Robert A. Leathers,et al.  Optical remote sensing of benthic habitats and bathymetry in coastal environments at Lee Stocking Island, Bahamas: A comparative spectral classification approach , 2003 .

[13]  Serge Andréfouët,et al.  Spectral reflectance of coral reef bottom-types worldwide and implications for coral reef remote sensing , 2003 .

[14]  Eric J. Hochberg,et al.  Capabilities of remote sensors to classify coral, algae, and sand as pure and mixed spectra , 2003 .

[15]  Ellsworth F. LeDrew,et al.  Hyperspectral Discrimination of Healthy Versus Stressed Corals Using In Situ Reflectance , 2001 .

[16]  J. Jaubert,et al.  Spectral discrimination of coral mortality states following a severe bleaching event , 2000 .

[17]  E. Hochberg,et al.  Spectral discrimination of coral reef benthic communities , 2000, Coral Reefs.

[18]  Ellsworth F. LeDrew,et al.  Measuring and modeling water column effects on hyperspectral reflectance in a coral reef environment , 2002 .

[19]  Trijntje Valerie Downes,et al.  Interpretation of hyperspectral remote-sensing imagery by spectrum matching and look-up tables. , 2005, Applied optics.

[20]  Robert van Woesik,et al.  Coral reef texture classification using support vector machines , 2007, VISAPP.

[21]  Rangachar Kasturi,et al.  Machine vision , 1995 .