Group

Camera Traps / Feed

Looking for a place to discuss camera trap troubleshooting, compare models, collaborate with members working with other technologies like machine learning and bioacoustics, or share and exchange data from your camera trap research? Get involved in our Camera Traps group! All are welcome whether you are new to camera trapping, have expertise from the field to share, or are curious about how your skill sets can help those working with camera traps. 

article

Collaboration Spotlight: BoomBox

Ellie Warren
Today we're celebrating the #Tech4Wildlife Photo Challenge by shining a spotlight on one of our favorite WILDLABS collaboration success stories: the BoomBox! This collaboration between Dr. Meredith Palmer, Jacinta ...

0
See full post
discussion

MegaDetector on Edge Devices ??

Hi all.  I'm a relatively new member of the community and have been trying to consume the many excellent videos, discussions and resources before asking questions. ...

2 0

Thanks for your interest in MegaDetector!  You're right that it's not practical to run MegaDetector on edge devices; its architecture is chosen to prioritize accuracy over speed, so it likes a big GPU.  Well, more accurately... one *can* run anything on anything, but you will pay such a price in hassle and time that it's almost certainly not worth it.

Moreover, if we made a "light" version of MegaDetector (or any heavyweight model),  it would still be too heavy for some environments, and too light (i.e., insufficiently accurate) for others.  And you would still be spending lots of the model's bandwidth on animals and ecosystems that may not be relevant for you.

So... a more common approach among MD users who want to run edge models has been to take some sample unlabeled data from your specific ecosystem, or similar data that's publicly available (there's a lot of camera trap data at http://lila.science, for example), run that data through MegaDetector, and use the resulting boxes to train a model that fits your specific compute budget and your a framework that's easy to use in your environment (sometimes TFLite, often YOLO).  This is an inelegant but very effective form of model compression, and it has the benefit of only asking your small model to deal with images that are relevant to your project (as opposed to MegaDetector, which uses up model bandwidth to deal with all kinds of ecosystems you may never see).

Hope that helps... of course if someone wants to take on the task of building a *family* of compressed MegaDetectors to provide a more off-the-shelf approach, we'd like to see that happen too!

-Dan

To follow up with results from my testing, you can run MegaDetector on a Pi, if you're not in a hurry. I followed the instructions on GitHub for running on Linux and the installation of Python packages went smoothly. On a Pi4 with 8GB RAM it took just over 2 min per image (using 3 megapixel images from Reconyx cameras). So if you're capturing less than 700 images per day then the Pi could keep up. It won't keep up with real-time captures though, particularly if you get bursts of images. Even high-end GPU's struggle with processing more than an image per second. It could be quite a useful way to reduce the processing burden at the end of a camera trapping session, or to trigger another event, such as sending only images with animals via email/telemetry etc.

See full post
discussion

Wide angle camera trap

I am looking for a camera trap model with a wide(r) angle lense than the conventional ones offer. The model needs to be able to function properly in a tropical forest environment...

12 0

Important considerations raised by Peter!

In my case it was not important though, as I was using the trap camera in timelapse mode.

 

I agree the PIR sensor and the camera will be "seeing" different pictures, but I believe that is exactly the effect that is sought: now too much of the elephant is out of the frame when the camera is triggered, and the wide angle lens is desired so that more of the elephant would fit in the frame at that same triggering point.

In the TrailCamPro link above (comparing FOV and Detection Angle), I see a few "panorama" camera models.  For example, the Moultrie 150i or Moultrie 180 (https://www.moultriefeeders.com/catalogsearch/result/?q=panoramic) - although they are all listed as discontinued.

It seems this might be a solution for Daniela's scenario.

I'm also interested because it could offer more forgiving setup (if the subject does not travel exactly where expected.)

Has anyone here worked with a panoramic camera? What did you find to be their pros/cons?

See full post
discussion

Recommendations needed: Rechargeable batteries for camera traps

We are exploring rechargeable battery options, particularly for Cuddeback XChange Color (C1) models. These cameras use 8x AA batteries, and ideally need a charge of 1.5V...

9 0

I have only used Tenergy NiMH rechargeables, these put out 1.2 V. We've used it on Bushnell Trophycam, which have 2x4 battery sets, i.e. you only need 4 batteries for it to have enough voltage, the additional batteries don't increase the voltage, they only prolong working time. So 4x1.2=4.8 V instead of the expected 6 V from alkaline batteries. I would think that Cuddeback has a similar circuit setup, so an external 12 V battery might actually be too much, as some have already pointed out. With regular NiMH batteries what's most important is to get batteries with high mAh. The ones sold commercially usually have very low mAh, so they won't last very long. We have 2400 mAh, I think, and it works reasonably well, they can last for about a month in the field, IR flash works.

While the manufacturer might claim that the camera requires 8xAA at 1.5v each, most likely it will work just fine with NiMH batteries that have nominal voltage of 1.2V.
I have used eneloops with Reconnyx cameras for a long time, as well as with handheld GPSes and a myriad of electronic devices and not once run into trouble because of the lower voltage. Your camera should have a setting in the menu to select NiMH batteries, that will prevent it from shutting down too soon.
I suggest you do your own testing - run 2 cameras one next to another on a 1.5V alkaline battery and a 1.2V NiMH rechargeable one until it switches off and check the voltage on the "empty" cell.

Your issue will likely not be the voltage of the cell, but the current the battery can deliver, as it has to recharge a capacitor in the incadescent flash light. I see that the manufacturer declares up to 20s recharge time for night photos, which is a lot. That is a downside of a colour camera.

I don't know what the best source is for batteries in SA, but if possible, get rechargeable batteries from IKEA (Ladda NiMH batteries). They are rebranded eneloops pro for around 30% less and I am yet to find a better battery for a camera trap. Otherwise, like mentioned before, eneloop pro will be hard to beat for performance and reliability.

See full post
discussion

Best Camera Trap Models Database: Input Needed

Hi Wildlabbers, We received a question from the National Geographic Exploration Technology Lab about the most popular low to mid-range camera trap models within our...

40 0

Gracias, Joaquín, for the useful Coto Doñana camera evaluation information in https://zslpublications.onlinelibrary.wiley.com/doi/full/10.1111/jzo.12945 .  This initiative will presumably eventually appear in the European Observatory of Wildlife map.  Currently it only displays three entries for Iberia two of which, disappointingly, appear to serve as guides to where "Big Animals" can be killed. 

Many thanks, "mactadpole" for the promising remarks concerning the Browning Dark Ops Pro XD dual-lens BTC-6PXD:

"...we are extremely pleased with the BTC-6PXD. We went with these because they only use 6 aa batteries and they were smaller/lighter than the BTC-8A."

Given the similarity between the western Ecuador conditions you describe and those we face in Costa Rica the Browning - 180$ from Amazon where 37 reviews are predominantly favourable - sounds like the camera for us.  Your 12.2.2021 report is now over a year old, however.  Please, has anything changed since then?  Any other candidate we should consider?

Hi Shawn,

I am looking into camera traps to use for an arboreal project in Panama, I am really interested in your experience of mounting camera traps up trees. The photo shows an interesting mount, did you make it yourselves?

How were the seals on the Brownings? I have been tempted to go for reconyx cause they have really good o-ring seals but they may just be too pricy so looking for a reliable alternative.

Anything you can share will be useful.

Cheers

Lucy

See full post
discussion

I made an open-source tool to help you sort camera trap images

Hi all. For the past couple of months I have been volunteering with the New Zealand Department...

14 0

Hi seems great and easy to use! Just a question, can the software success to identify the species or "only" categorize animal/ vehicle/human? Can we "trained" the software to detect a specific species?

 

thank you 

Right now the only classifications are animal/vehicle/person/empty. It cannot discern between different species.

There is no support for training at the moment -- I am envisioning something down the line but I wouldn't say that's coming any time soon.

Hope that helps!

Hi

I just tried and works great, I will include it in my workflow. 

Thanks for your work!

Juan

 

See full post
article

Protecting Wildlife with Machine Learning

Hack the Poacher
Last year, Tim van Deursen and Thijs Suijten shared their new "Hack the Poacher" system with us, presenting a unique way to detect poachers in real-time within protected national parks. Read on to learn about their next...

0
See full post
discussion

Tech Tutors: How do I start a Tech4Wildlife collaboration?

Hi Wildlabbers, Ed Miller and Melanie Clapham tackled the...

2 0

One barrier that was discussed in the presentation was the "risk" or uncertainty of a payoff when talking to potential funders. It seems to me that the "risk" factor of collaborations like this, for funders, would be analogous to the investment and payoff when developing a genomics program. Relative to traditional field research, it takes a much greater investment in time, with different expertise, to create a reference genome for a species of interest and then pull SNPs, etc. Nevertheless, funders seem to recognize the potential for genomic research and genomic tools. Do you think that greater openness to “risk” in conservation genomics results from closer intellectual proximity (i.e., molecular biology vs organismal biology), or to conservation genomics being a more mature field (i.e., more examples of success), or to other factors? If intellectual proximity is a key factor, then conservation tech collaborations will continue to face this barrier, but if lack of successful examples are a key factor, then perhaps this barrier will become less of an issue for conservation tech in the future. Thoughts?

Hi. Just watched the Youtube version (I'm in the unable to watch live hemisphere) and wanted to say that was a great talk. I wish it could have gone for another hour. I liked hearing about the issues and concerns from Melanie on the biology/ecology side. Also could totally relate to what Ed was talking about from the tech side about how it's difficult to have long term collaborations since there isn't really a conservation technology career path at the moment. 
I think one of the things we're trying to do with courses like Build Your Own Datalogger is to get people involved in wildlife/conservation side more comfortable with the tech. The goal is to hopefully build a shared repository of technical knowledge that people in the wildlife community can contribute to and benefit from. 

Akiba

See full post
discussion

Open Source Camera Trap

Hi all new to your site so far a lot of good info from what I see. I have multiple trail cameras some cellular some standard but what I would like to build is a long range WiFi...

5 0

Akiba I have used ubiquiti loco m5 to broadcast internet with great success when I was setting up internet for a park for a 10k run I was able to get atleast a mile with no issues with internet 

not discorcuaged just looking  to piggy back off of a local wifi and be able to connect to a diy camera trap and be able to either do live video feed or capture a pic and then with a program I create send it to email or txt message 

the place I am looking at I can get almost a mile in distance but the problem is the power it would have to be some kind of solar power don't want to have to change batteries ever other day 

Something I'm working on that might work well for you and is super affordable at about $30 per camera.

ESP32-cam (not super high quality photos though) with a motion sensor and it sends the photos to a PI Zero W that can be accesses at any time via wifi. I'm using the TPlink portable but have found that WemosD1 boards and upgrading the firmware to be able to do wifi repeaters works well.

Here's the camera I'm using:

https://starairvision.com/esp32-cam-motion-camera-with-date-and-time/

I'll make a post on here when the full project is built and I've tested the range. The repeaters use something called mesh networking. You can have a number of them in a row to get the camera out into the forest. Or if you have lign of sight you can use a signal booster and or a directional antenna.

Thanks Ryan would like to hear more about this, the project look awesome what distance are you getting from your wifi router 

See full post
discussion

Wireless Camera for Live Viewing

Hi Everyone, Just thought I'd share a project I'm working on for a cheap remote camera system. The range is short but it connects to a cell phone without needing to...

29 0

Look forward to playing with it when mine arrives I have  a sd card arduino module also not finding much documentation on how to hook it up so was going to play with a server and see what I can do with it and photos 

Nice! I'm building an esp32-cam with a Pi Zero server and I have the photos sending to it now. The last step is to build a housing for it and get it outside to capture some photos of the bears coming out of hibernation. Should be fun! I'm hoping to download photos from the camera over 1km away from my house, while sipping tea haha.

I have a raspberry pi3 setup now as a video streaming server so I can keep a eye on my home while I am at work 

See full post
event

Making the Most of Tech Tutors Season 2!

WILDLABS Team
WILDLABS is celebrating its five year anniversary! Throughout the rest of 2020, we'll be sharing articles, community features, and case studies showcasing the incredible projects, collaborations, and successes that this...

0
See full post