Imagine two fish cavorting 50 feet below our boat. Our echoscopic "fish-finder" allows us to watch them via echoes of sonic clicks that have known and reliably detectable features.
The animation here illustrates two things:
1) a possible early version of "the world's best fish-finder," and
2) the simulations of the "Experiment Machine" by means of which the mystery of the imaging sonar of dolphins is being solved.
The animation, from 2005, is composed of a series of individual pictures. Each was computed from simulated echoes of just one click - like a picture might be taken using a flash. But, our imaging system does not use an acoustic lens. It computes from times-of-arrival of echoes at a multitude of microphones. It uses "active" sonar - in which both the time and place of origin of a click are known.
Echoes from various parts of the two fish arrive, in possibly different orders, at different acoustic pickups. To a great extent, they destructively overlap one another on most of the sensors.
Yet pictures can, demonstrably, be computed. You see two simulated fish.
In addition to the main point-of-view, there is, at the bottom of the window, a side view illustrating the three-dimensional nature of the model that results from echoscopic computing in "active mode" -- in which both the time and the place of emission of a click are known.
The color of a dot indicates the loudness of the echo which produced the dot, in decreasing order: red, green, blue, grey.
The grid lines are about 6 inches apart in both the Top and the Side views. They represent distances at the central range of the fishs' motions, about 50 feet from boat to fish.
So, the total length of a fish seems to be about 20 inches?
Each fish is composed of grey skeletal sections along with a tail and a dorsal fin. Using only parts of a fish decreased time-of-computation in running experiments.
Each fish, though skeletal, is simulated by about 5000 points.
Each acoustic sensor reveives the sum of 5000 little echoes.
Each image was computed from echoes of a single click.
The pictures, that are the individual frames of the animation, were produced during design/testing/re-design etc and
were not improved much for impressing a viewing public.
The picture might seem a bit jerky. The number of frames in the movie
was kept low for practical reasons.
This animation shows that two fish can be separated by the dolphin-inspired methods. An earlier animation shows more (and not all) that could be done, with the methods of 2004, in imaging a single fish. Both the 2004 and 2005 animations were made via active sonar, hence, very good range-information allowing computation of side-views.
The animation here came from simulation of "active sonar."
Almost certainly, dolphins do not use active sonar. More probably, they use FBP, feature-based, passive sonar, introduced in this website.
A first animation showing results of FBP
More informatin at:
Overview of this website.
Introduction: Home of Dolphin Inspired Sonar