System and methods for processing and the visualization of bioaccoustical information
The present invention is directed to acoustics and signal processing. More specifically, the present invention is directed to a system and methods by which whether and the time during which one or more subjects are present within a given area of a marine or terrestrial ecosystem, and other information regarding the subjects can be efficiently determined by the sounds that the subjects produce. Advantageously, certain embodiments of the present invention permit a user to select and rapidly produce one or more visual presentations of the signals of interest (SOI) recorded from the passive acoustic monitoring of a given area of an ecosystem in order to obtain additional information about the subjects.
1. A method for visually presenting acoustical information comprising the steps of:
scanning digital acoustical data recordings from one or more subjects in an environment;
detecting from the digital acoustical data recordings signals within pre-selected frequency bands;
computing a histogram of feature values of the detected signals;
presenting on a display a first presentation comprising the histogram;
choosing feature values of the detected signals by choosing regions on the histogram;
automatically selecting signals based on the choosing step; and
illustrating on the display a second presentation comprising the automatically selected signals, wherein the first presentation and the second presentation are shown on the display simultaneously.
2. The method according to claim 1 , wherein the feature values are selected from the group consisting of: time of arrival (“TOA”), duration, minimum frequency, peak frequency, signal-to-noise ratio (SNR), maximum bandwidth, instantaneous bandwidth, intensity, signal class.
3. The method according to claim 1 , wherein the histogram is a one-dimensional histogram.
4. The method according to claim 1 , wherein the histogram is a two-dimensional histogram.
5. The method according to claim 1 , wherein the histogram comprises a y-axis directed to sound intensities distributed over frequencies and an x-axis directed to time in seconds.
6. The method according to claim 1 , wherein the computing step includes the step of extracting features using a function ƒ n (t k ), n=1 . . . N, k=1 . . . K representing a n-th feature value of each detected signal at time t k , wherein N is a total number of feature values associated with each detected signal and K is a number of detected signals.
7. The method according to claim 1 , wherein the presenting step further comprises the steps of:
representing the feature value as ƒ n ; and
defining the histogram as H(g n ,T), where g n ∈{g n1 <g n2 < . . . <g nI } is a grid of equally or non-equally spaced values of a n-th feature value, T∈{T 1 <T 2 < . . . <T j } is a time grid, and H(g ni ,T j ) is a number of signals whose n-th feature value falls within an interval [g ni ,g ni+1 [ and a time of arrival falls within an interval [T j ,T j+1 [.
8. The method according to claim 1 , wherein the presenting step further comprises the steps of:
representing the feature value as ƒ n , ƒ k ; and
defining the histogram as H(g n ,g k ), where g n ∈{g n1 <g n2 < . . . <g nI } is a grid of equally or non-equally spaced values of a n-th feature value.
9. The method according to claim 1 , wherein the second presentation is a spectrogram.
10. The method according to claim 1 , wherein the one or more subjects is an animal.