FishScale
October 3, 2024
FishScale estimates fish lengths, numeric abundance, and biomass density in underwater still images. This software package is further described in a journal article by Esselman et al. (in revision; Methods in Ecology and Evolution) entitled "A transferable approach for quantifying benthic fish sizes and densities in annotated underwater images". FishScale uses binary masks obtained from semantic segmentation of individual fish in images to provide accurate estimates of fish lengths, numeric abundance, numeric density, biomass, and biomass density after first correcting for lens distortion, biases caused by underwater magnification, distance from the camera to fish targets, and variation in fish curvature. Users must provide size calibration images and fish length-to-weight relationship(s) to fully utilize the software’s capability.
Citation Information
Publication Year | 2024 |
---|---|
Title | FishScale |
DOI | 10.5066/P13QVR2R |
Authors | Shadi (Contractor) Moradi, Peter C Esselman, Joseph (Contractor) K Geisz, Christopher Roussi |
Product Type | Software Release |
Record Source | USGS Asset Identifier Service (AIS) |
USGS Organization | Great Lakes Science Center |
Rights | This work is marked with CC0 1.0 Universal |
Related
A transferable approach for quantifying benthic fish sizes and densities in annotated underwater images
1. Benthic fishes are a common target of scientific monitoring but are difficult to quantify because of their close association to bottom habitats that are hard to access. Advances in image-acquisition technologies, machine vision, and deep learning have made capturing and quantifying fishes with cameras increasingly feasible. We present a method and open-source software called...
Authors
Peter C. Esselman, Shadi Moradi, Joseph K. Geisz, Christopher Roussi
Related
A transferable approach for quantifying benthic fish sizes and densities in annotated underwater images
1. Benthic fishes are a common target of scientific monitoring but are difficult to quantify because of their close association to bottom habitats that are hard to access. Advances in image-acquisition technologies, machine vision, and deep learning have made capturing and quantifying fishes with cameras increasingly feasible. We present a method and open-source software called...
Authors
Peter C. Esselman, Shadi Moradi, Joseph K. Geisz, Christopher Roussi