Using the Hough transform to detect fish in Freshwater Creek

Graduation Date

2007

Document Type

Thesis

Program

Other

Program

Thesis (M.S.)--Humboldt State University, Environmental Systems, Mathematical Modeling, 2007

Committee Chair Name

Kenneth D. Owens

Committee Chair Affiliation

HSU Faculty or Staff

Keywords

Hough transform, Fish detection, Humboldt State University -- Theses -- Mathematical Modeling

Abstract

We present a fish detection method capable of detecting fish profiles in underwater video while filtering out complicated unwanted images such as fish tails, otters, juvenile fish and turbulent water. Also, fish attributes, such as size and orientation, are automatically displayed as fish are detected. This method utilizes the Hough transform to automatically detect fish profiles. We display the results of this method applied to images from Freshwater Creek and discuss benefits from using the Hough transform to detect fish. We also discuss strengths of the method along with future applications, including the possibility of implementing a real-time detection scheme based on this method.

https://scholarworks.calstate.edu/concern/theses/bc386m49q

Share

 
COinS