It's probably much easier to find a company selling acoustic modems than to try and create your own...very challenging environment to work within. I have about 10 years working with underwater acoustics (as a non-engineer). 

To give you the best advice, I need a bit more information about your specific application. Underwater communication typically relies on acoustic modems because radio waves don't travel well through water. For longer distances, a hybrid solution is often used. This could involve an acoustic modem transmitting data from your underwater sensor to a surface buoy or platform. The buoy could then relay the data to shore using traditional radio communication, such as 900MHz radios (even commercially available options like Ubiquiti radios, depending on distance and line-of-sight). To help me understand your needs and recommend the right approach, could you tell me more about:

Data Rate Requirements: How much data do you need to transmit, and how often?

Range: What is the vertical distance between your sensor and the surface receiver?

Environment: Are you working in shallow or deep water? What is the water temperature, salinity, and expected noise level?

Power Budget: How much power is available for the underwater sensor and modem?

Cost: What is your budget for the acoustic modem system?

Integration: How will you integrate the modem with your sensor and surface receiving system?

Real-time or Delayed: Do you need the data in real-time, or can it be stored and transmitted later?

I just noticed that TrapTagger has integrated AI reading of timestamps for videos. I haven't had a chance to try it out yet, but it sounds promising. 

Hi Lucy,



I now realised it is an old thread and you most likely have already found a solution long ago but this might be of interest to others.

As mentioned previously, it is definitely much better to take moon phase from the date and location. While moon phase in general is not a good proxy for illumination, that moon phase symbol on the video is even worse as it generalises the moon cycle into a few discreet categories. For calculating moon phase you can use suncalc package in R but if you want a deeper look and more detailed proxy for moonlight intensity, I wrote a paper on it 

Read "Biologically meaningful moonlight measures and their application in ecological research | Behavioral Ecology and Sociobiology " on SpringerLink

Biologically meaningful moonlight measures and their application in ecological research | Behavioral Ecology and Sociobiology

Light availability is one of the key drivers of animal activity, and moonlight is the brightest source of natural light at night. Moon phase is commonly us

SpringerLink

 with accompanying R package called moonlit 

Read "GitHub - msmielak/moonlit: moonlit - R package to estimate moonlight intensity for any given place and time" on GitHub

GitHub - msmielak/moonlit: moonlit - R package to estimate moonlight intensity for any given place and time

moonlit - R package to estimate moonlight intensity for any given place and time - msmielak/moonlit

GitHub

When it comes to temperature I also agree that what is recorded in the camera is often very inconsistent so unless you have multiple cameras to average your measurements you are probably better off using something like NCEP/NCAR Reanalysis (again, there is an R package for that) but if you insist on extracting temperature from the picture, I tried it using tesseract and wrote a description here: 

Read "Michał Śmielak - Extracting date and time from camera trap photos using R and tesseract" on

Michał Śmielak - Extracting date and time from camera trap photos using R and tesseract

Good luck!

Thank you for sharing! 

Hi Brett,

I am interested in developing really low cost multi channel underwater acoustic recorders.  Can you tell me a bit more about the board and stuff you were using for yours? You can get me at chris@wilcoanalytics.org.

Thanks,

Chris

Jared's answer is excellent IMHO. One level down, you have a couple of options. One is to charge for the convenience of buying a finished, assembled, and tested product, that's for the folks that have money and no time. Those that have time but no money can build it themselves from the open source. One danger to this approach is if low cost chinese clones come on the market, whether they do probably depends a lot on the size of the market.

Another option is to withhold some 'pro/plus' feature from the open source, that depends on whether you can identify such features. A similar option is to offer features that big outfits with good funding need. For example, you could offer calibrated versions at a premium, if that's applicable and needed by such organizations. In many cases companies sell devices cheap that are usable by someone who one has one or a couple and then charge for a management system for those that have dozens or hundreds of devices, dunno whether this applies to your case. All this comes down to whether you can segment your customer base so you can create premium features for the wealthy segment.

Hi Ben!

Great initiative! 

A review of deep learning techniques for detecting animals in aerial and satellite images

https://www.sciencedirect.com/science/article/pii/S1569843224000864#b0475 

lists a number of data sets (incl. one published by you it seems...) 

Also, @dmorris keeps a list of Terrestrial wild animal images (aerial/drone):

https://lila.science/otherdatasets#images-terrestrial-animals-drone

Read "List of other conservation data sets - LILA BC" on LILA BC

List of other conservation data sets - LILA BC

LILA BC - List of other conservation-related data sets

LILA BC • 16 Mar 2017

which seem like it might be useful for you.

Hi @benweinstein !

1. Surely a general detector might be very useful for detecting objects in aerial imagery! Maybe something similar to what MegaDetector does in camera trap images, generally detecting person, animal, and vehicle (and thus also empty photos). This could greatly improve semi-automated procedures. It could also serve as a first step for context-specific detectors or classifiers to be developed on top of this general one.
2. There is also the WAID dataset that is readily available. Our research group could also provide images containing cows, sheep, and deer from South America.

Best