Accession Number:

AD1014344

Title:

DCL System Using Deep Learning Approaches for Land-Based or Ship-Based Real Time Recognition and Localization of Marine Mammals

Descriptive Note:

Technical Report

Corporate Author:

Bioacoustics Research Program, Cornell University Ithaca United States

Personal Author(s):

• Dugan,Peter J.
• Clark,Christopher W.
• LeCun,Yann A.
• Van Parijs,Sofie M.

Report Date:

2015-09-30

Pagination or Media Count:

15.0

Abstract:

The ONR DCL grant focuses on advancing state-of-the-art of data-mining for the bioacoustics community through researching and creating new technologies, algorithms and systems to decipher and understand very large passive acoustic datasets. The long-term goal is to develop advanced computational systems and algorithms that will provide scientists the ability to efficiently harvest animal localizations from large, complex datasets. The newly developed systems provides efficient, high performance processing of acoustic sounds by allowing a state-of-the-art technology to host algorithms for advanced detection and classification and other data-mining strategies.

Descriptors:

• bioacoustics
• data mining
• marine animals
• relational data bases
• algorithms
• acoustic detection
• classification
• noise sound

Subject Categories:

Distribution Statement:

APPROVED FOR PUBLIC RELEASE