90 likes | 216 Views
Content Analysis and Restoration of Marine Video. Student :- Ken Sooknanan Supervisor :- Professor Naomi Harte in collaboration with Prof. Jim Wilson (Dept. of Zoology), the Marine Institute
E N D
Content Analysis and Restoration of Marine Video Student :- Ken Sooknanan Supervisor :- Professor Naomi Harte in collaboration with Prof. Jim Wilson (Dept. of Zoology), the Marine Institute Ireland, and Prof. Anil Kokaram (Google)
Content Analysis:- • (a) Identify Nephrop burrow • complexes • (b) Identify when major changes • along the seabed occur. Goals (a) Complex (b) Nephrop Video Light Footprint • Restoration:- • (a) Improve Visibility by correcting • illumination degradations due to • light source. (d) Muddy Video (c) Rocky
Improving Visibility r1 r2 G1 G2 (1) Degradations modelled as mixture c c Frame 1 Frame 2 (2) S, c estimated with ratio of pt. correspondences r ≤ rf G2(r2) = G(x) G1(r1) r > rf G(x) s1 s2 cx ; c = S = cy s2 s3 optimized alternately in Bayesian Framework cn+1=arg max p(c|G1,G2,Sn); Sn+1=arg max p(S|G1,G2,cn+1); (d) A(rx) > μ. Split into regions, footprint (red) (3) Footprint estimated as region where largest percentage increase in degradation occurred (c) A(rx) =
Results IMAGES :- (a) Original with footprint in blue (b) Our Result (c) Seon Joo Original Result (d) Seon Joo Result with our radii est. VIDEO :- [1] Seon Joo, et al.., “Robust radiometric calibration and vignetting correction," Pattern Analysis and Machine Intelligence, 2008
Identifying Nephrop Clusters • Problem broken up into two parts:- • - identifying all burrows • - clustering • Currently working on identification part (a) Sample Video Frame • Exploring the use of a Mosaic, as it:- • Eliminates tracking • Fixes geometric distortion. • - Computations reduced to a single image • - Provides a wide view of the seabed. • - Easy to spot clusters. (b) Generated Mosaic (using 1st 50 Frames)
Algorithm overview • Highlight all dark regions • - DOG Segment and label DOG – Intensity based (Bayesian framework - ICM). Identify and Split multiple burrows regions – shape modelling with GMMs (a) Original (b) Segmentation (c) Est. GMM (d) Split Region Classify based on shape, colour and shading features, f, - Cascade Classifier, p(f) > (T =0.5) SHAPE COLOUR SHADDING
Initial Results (a) Original Mosaic (b) Identified Burrows Results obtained from 3 Video Sequences (10 min., 15000 Frames) Compared with Ground Truth from:- (1) Video (existing method) (2) Mosaic
Accomplishments Year 1:- Attended and passed 3 courses to gain the necessary 15-credits Year 2:- Wrote year-1 transfer report and passed viva to enter PhD roster. Year 3:- (1) Gave a presentation in Google, San Francisco on 23rd Jan. 2012 (2) Published a paper [1] in SPIE conference, San Francisco Jan. 2012 (3) Submitted a paper [2] on Burrow Detection algorithm in ICIP 2012. (4) Submitted an abstract [3] on Mosaicing Algorithm in Oceans conference 2012. • Established good collaboration with the Marine Institute Galway • - keep in contact with Jennifer Doyle approximately once every month. [1] Sooknanan, et al.., “Improving Underwater Visibility using Vignetting correction," in proceedings of SPIE, 2012 [2] “A Bayesian Framework for Detection of Nephrop Burrows for Seabed Video Analysis,” ICIP 2012 [3] “A Bayesian Framework for Mosaic Creation of the Seabed from Underwater Video For Nephrop Burrow Detection”, Oceans, 2012
Future Work (1) Present work in Study Group on Nephrops Survey (SGNEPS) Meeting in Italy in 8th March 2012. (2) Do more testing with Burrow detection algorithm (3) Move onto the Burrow clustering part of the problem (4) Move onto detecting when major changes in seabed type occur (5) Write papers on (3) and (4) (6) Write up Thesis.