Talks and presentations

Published:

Peña, M. Using histogram equalization to visualize acoustic and ancillary data.

Published in Working Group on Fisheries Acoustics, Science and Technology (WGFAST)., 2022

Echogram visualization and processing is one of the most time-consuming tasks for fisheries acousticians. Analyzing target species or features is often based on experience while visualization settings are inherited from colleagues or established for standardization purposes within international efforts. Acoustic data is often visualized with a standard minimum threshold for Sv that varies with target species. For instance, small pelagic fishes are often visualized in European waters from − 60 dB re 1 m2 m− 3 while echograms of the mesopelagic zone (200–1000 m depth) usually employ a − 90 or − 80 dB re 1 m2 m− 3 minimum threshold. However, numerical volume density changes greatly with depth (particularly beyond the shelf), time of the day and season, and thus setting an incorrect threshold may mask part of the population at some times or areas. Following on Blackwell et al. (2019) that showed the best colormaps to be employed in fisheries acoustics, this presentation focus on further parameters (general thresholds and location of the color limits) of the colorbar. Most colormaps employ linear relationship between data and color, where every color represents a similar range of values. This can hide interesting features in only one or two colors. Histogram equalization is a non-linear interpolation technique that locates thresholds and color limits at the quantiles of the image data. Examples using acoustic and ancillary data will be shown to highlight the benefits of this technique. While no definite conclusion is provided, some food for thought on the influence of the number of colors employed will also be provided. Keywords Echogram visualisation, colormaps, histogram equalizarion

Published:

Published: