HabCam Data processing for Leg 3 of the 2009 Scallop Survey

Scott M. Gallager and Amber York, Woods Hole Oceanographic Institution, Richard Taylor Arnie’s Fisheries, Inc.

   Print  PDF  Change text to small (default) Change text to medium Change text to large

This proposal is to support image processing, scallop and ground fish identification and enumeration, and comparison of HabCam data with that obtained by standard dredge tows during Leg 3 of the 2009 NMFS scallop survey. The HabCam imaging system was installed on the R/V Hugh R. Sharp for Leg 3 of the survey to conduct joint tows for the purpose of addressing dredge efficiency, logistical difficulties related to operation of an optical survey tool under full survey conditions, and to evaluate the abilities for HabCam to survey groundfish, particularly yellow tail flounder. HabCam surveyed 23 standard stations in parallel with the Sharp and collected a total of 485,451 images with a footprint of about 1 m^2 each. By area, 85,572 images were collected in Closed Area I, 216,809 images in Closed Area II, 183,070 on the Canadian side of the Northeast Peak of Georges Bank, and 302,381 images between stations. The objectives of this proposal are to: 1) Process all images from HS_20090623 for color correction, light field correction, and conversion to jpeg. 2) Manually count and size images along each track for scallops, all fish, and substrate characterization. 3) Plot ship tracks, images and data products in Google Earth for ease of visualization. 4) Establish a training set of images manually processed for use with an automated target classifier and compare results with manual classification. 5) Compare results from HabCam with those of the dredge survey to obtain a dredge efficiency for both scallops and yellow tail flounder, where possible.

Last updated: February 25, 2015