i
Development of electronic monitoring (EM) computer vision systems and machine learning algorithms for automated catch accounting in Alaska Fisheries
-
2023
-
-
Series: AFSC processed report ; 2023-01
[PDF-7.96 MB]
Details:
-
Personal Author:
-
Corporate Authors:
-
NOAA Program & Office:
-
Description:EMI research consists of the development and deployment of camera systems for acquiring imagery and the development and integration of automated CV machine learning algorithms and applications. The purpose of these systems is to detect and identify catch events from fishing imagery and to classify those detection to species or larger taxonomic groups. Once these detections are made, further data analysis about the catch event can be obtained, such as length estimation and count information. Algorithms have different requirements based on the detection types and the fisheries environment involved. EMI identified multiple fisheries applications where CV can be of use. These include 1) automated species detection, identification, and length estimation of fish as it is caught at the rail of fixed gear (longline) vessels; 2) species identification of fish images collected in controlled environments; 3) detection, count, and length estimation of Pacific halibut (Hippoglossus stenolepis) bycatch; and 4) detection, count, and distinction of salmon from processing plant belts containing multiple species of fish. Additional algorithm functionality for the detection and monitoring of crew member activity on vessel decks is also presented.
-
Keywords:
-
Series:
-
DOI:
-
Format:
-
Document Type:
-
License:
-
Rights Information:CC0 Public Domain
-
Compliance:Submitted
-
Main Document Checksum:
-
Download URL:
-
File Type:
