Hi Gregory,

That is quite shocking to hear that a road could be so detrimental to wildlife. If you were able to place a speed hump or rumble strip where you place the camera you could even embed the metal detecting sensor in them to ensure accurate detection of every vehicle when they are going slower.

Our camera takes a burst of images and this can be set to be quite a quick burst or a slower burst so with a little bit of testing I am sure we could get the numberplates.

I am assuming that a lot of time these collisions happen at night. Would there be a way to have a fixed light at the camera locations as otherwise the cameras can be blinded by headlights and taillights making reading the VRN difficult.

At the moment our system is satellite connected. This does have a relatively high monthly data cost. We are also planning to develop a cellular version but this will require more funding and development time that we have not got yet.

The system is in the final stages of development and we plan to launch it in mid-2020 once it has undergone some prolonged field trials at ZSL sites, so it is not currently available. Final costings are still under review and will be released early next year.

I am sorry that we are not able to provide this capability to you right now. 

Have you found any commercially available camera systems? In the UK I believe the automatic number plate recognition cameras placed on gantrys over motorways are called SPECS and a quick google said they are sold by a company called Jenoptik. If you find out how much they are and how they work, perhaps you could update this thread?

Thanks,

Sam 

Thanks Sam, presumably you have lots more power than would be available from the battery in a tracking collar.

Peter

Thanks Sam

I am looking at LoRa for high resolution GPS tracking, and the project also involves camera trapping at scent marking sites. Power is already limiting collar lifetime (most of it goes to the GPS rather than the LoRa module), but transmitting real time pictures from marking sites while the marking is going on would be neat.

A few years ago this solution would have been thought of as nuts.

Have another cheap camera pointing at the expensive camera.

How it would work is like this.

The expensive camera or its housing detects it is being tampered with (tilt detector attached to a microcontroller).  The microcontroller is also attached to a small short range radio (NRF24L01).  This thing will be asleep all the time and use almost no power, until it is tilted. BOM (bill of materials) less than $10.

The cheap camera is so small it can mostly be buried underground. That is also controlled from a microcontroller, and that too has a small radio. However this radio needs to be listening all the time - BOM will go up with a bigger battery.  I've had luck with these cheap 'key fob' 808 cameras, they record video onto their own sd card. BOM probably about $30-40.

Now this wont stop someone stealing the cams, but might lead to an arrest. The argument about sticking a GPS in the housing is not convincing for me, a GPS is a very specific type of pcb usually featuring an easily identifyable antenna - so can just be ripped out.  If it were me it would be the first thing I'd look for.

I've used cheap chinese arduino pro-minis and 808 style cams for a while, and while its nice to stick PIRs etc on them to detect intruders etc, the result is usually an empty battery and an sd card full of spider or ant head shots.

If you wanted to scare the wrong-doers off at the moment of theft, then it would be possible to play an audio warning, or shout 'you are now being filmed' or replay the crackle of a walkie-talkie (search : arduino mp3 player). The expensive camera's transmitting radio signal could be picked up by more than one device you see? The noise could be coming from the left, the cam in the ground could be on the right.  These simple radios can have a range of 50+ metres, if you pick the right ones. There are other RF options, as well as GSM dialers of course.

If all this sounds a bit complex, with some work and thought these counter measures could be set and forget "black box devices".

For the 808s I normally go through links on the chucklohr website, the absolute oracle on 808s. My last purchase were Mate 808 1080P "spy" cameras, with a long camera cable, they do away with the whole plastic housing and just give you the board with components.  I ordered from their website in '17, but it seems they are all over the place on ebay ATM.

I tried the SQ11s, but they run awful hot - but for say, 2 minutes of video they work just fine - but minute, I had one working in a hat - thats how I know how hot they can get.

As for "Interface with the 808" I take this to mean 2 things :

a) how do you intercept the video signal to, say, send something to an email address

 - you don't, well at least I couldn't, you have to physically go and get the sd card out

b) how do you turn it on an off?

 - I unashamedly used this idea from Gordon Williams of Espruino fame :

https://www.espruino.com/Time+Lapse+Camera

You'll notice the clever hack to fire up the cam using 2 i/o pins simultaneously.

Now the cam I mentioned above, like them all, has a config text file, so here you set the image size, whether it starts filming when turned on - with some other basic settings such as date/time etc.

There are lots of gotchas with this, not least of which is when reset that that text file is overwritten with some (sane) defaults, which may not be the default you want.

I found the Mate 808 config did most things, all I had to do was turn the cam on and off - and I got this working with an Arduino pro mini 3.3v.

So the "smarts" can be on an Arduino, read sensors, if/else, film for 30 seconds, 60 seconds, dont film and so on - send an alert etc.

I toyed with the idea of "potting" the cam and Arduino in resin to suit the environment, but as I said, the greater part of it could be buried - just means breaking out the microphone, to say, the same length as the extended camera cable. 

The drawback with PIRs I found was all the false positives, had better luck with these small microwave "radar sensors" - Andreas did a fine recap on what was available about 2 years ago.

https://www.youtube.com/watch?v=9WiJJgIi3W0

Given that the rechargable batteries are starting out with lower voltage, I would suggest trying to run a test wherein you use standard alkaline or high powered Lithium as a test to see if that resolves your issue.   If it does, and you want to make sure you can still use rechargeable batteries, maybe consider something like:

https://www.amazon.com/AA-Batteries-Rechargeable-ECO-Friendly-Recyclable/dp/B079JFK22D/ref=sr_1_3?crid=3UY7WGNZXC31Q&keywords=lithium+rechargeable+aa+batteries&qid=1561989505&s=gateway&sprefix=lithium+rec%2Caps%2C173&sr=8-3

as those will maintain a full 1.5 v level until they totally run out of power.  Both NiCD and NiMH batteries exhibit a voltage drop as power depletes, which could be what is causing decreased detections.

There was a talk recently at ZSL where a similar thing was found using camera traps studying wild pigs in the Osa Penisula. Tracking and other audio evidence showed that the pigs populations were circumnavigating camera traps which appeared to be giving a reduced level of detection.  

 @ollie.wearn made an interesting point as he has found in his research that wild pigs are extremely sensitive to human or unfamiliar smells and will avoid anthing that holds the scent. Once the pigs know the cameras are there they are likely to avoid them. 

Something to think about as I would doubt it would be anything to do with those cameras. 

Ollie I have all of my camera traps in Pelican Boxes but I buy camping foam and line the inside, then I buy 50mm foam and layer the cameras so they dont rest on each other and then foam on top, it sems to work well and my cameras get driven thousands of km over bad roads.

Paul

For transport of equipment on flights I suggest Pelican's line of cases called "Pelican Air".  They have slightly lighter shells so you don't run in to weight overage issues as much...

Eric

A cheaper alternative to Pelis is MAX cases. They are just as strong:

https://www.trifibre.co.uk/product-category/products-by-type/waterproof-cases/max-cases/?utm_source=google&utm_medium=ppc&utm_campaign=Search_MAX-Cases_Exact&gclid=Cj0KCQjw3uboBRDCARIsAO2XcYBSp1XbwS_KSQC-Y9brRjwKhJmjZjK6_vxT37LHrtLlr-uEavWuD0gaAgehEALw_wcB

Hi Toby, hi all,

I've been keeping very quiet about Instant Detect 2.0 on WildLabs as we wanted to ensure we didn't generate any hype or make any promises that we cannot fulfil....but I think we will have something that fits this brief almost exactly relatively soon and it would be a shame for the same work to be replicated. 

I have just returned from Kenya where I was testing our Instant Detect 2.0 prototypes and I am very pleased to report that the system is now working. We still have some optimisation work to do and we plan to run a number of longer trial deployments over a number of months later in the year to completely identify, and correct the sort of issues that only crop up over time and in the field. We plan to record and publish all the results from these trials.

Once we believe that the system is ready we hope to be able to supply it to the conservation community at a very low rate at the start of next year. Importantly it will be fully certified and legal to use, it is not 'hacky'. 

I am in the process of writing a blog about the testing but if you would like to know more perhaps we could have a chat sometime.

Best wishes,

Sam Seccombe

Hi Greg,

You are getting some great technical advice from folks a lot more qualified than me.   But as I read through everything, I see you are potentially wanting to use your solution for citing violations.  Whatever solution you install needs to be "future proof" for this from the start.  Not sure how that would work where you are (Zimbabwe, I believe?), but usually the accuracy of the speed reading and the vehicle identification has to be beyond reproach.  Calibration and accuracy, data integrity, etc., all need to be considered.

I would at least look into a buy vs. make decision (or a hybrid of the two) before committing to a completely homegrown solution.  It sounds like you only want to install one system.  You can find small but acurate radar systems and cameras pretty readily on amazon if you wanted a hybrid solution, or you could look into any one of the dozens of companies who make traffic speed monitoring equipment.  You already have a place to install, so that helps.  Some systems would be too expensive for your budgets or your needs, but some would be pretty reasonable.  Furthermore, those companies could potentially be enticed to provide a system at a substantial discount if they can consider it a charitable contribution.    But I'm just thinking out loud here, and maybe it would be difficult for these companies to ship a system to Zimbabwe.  

Just my 2 cents.

Drue

Hi Greg,

Indeed, the German solution you identified might be a better solution than what I had found on Amazon, which were mostly Doppler solutions, and thinking more about it, would not give you the ability to do long term data analytics on the traffic patterns, which might be important to achieving your objectives.  A few alternatives to the German solution would be:

https://enforcementlogix.com (Canadian)

https://www.atsol.com/solutions/speed-safety/ (US)

Good luck,

Drue