NOAA Right Whale Recognition Competition, January 2016

364 teams | $10,000 prize

https://www.kaggle.com/c/noaa-right-whale-recognition 

Competition Details:

With fewer than 500 North Atlantic right whales left in the world's oceans, knowing the health and status of each whale is integral to the efforts of researchers working to protect the species from extinction.

Currently, only a handful of very experienced researchers can identify individual whales on sight while out on the water. For the majority of researchers, identifying individual whales takes time, making it difficult to effectively target whales for biological samples, acoustic recordings, and necessary health assessments.

To track and monitor the population, right whales are photographed during aerial surveys and then manually matched to an online photo-identification catalog. Customized software has been developed to aid in this process (DIGITS), but this still relies on a manual inspection of the potential comparisons, and there is a lag time for those images to be incorporated into the database. The current identification process is extremely time consuming and requires special training. This constrains marine biologists, who work under tight deadlines with limited budgets.

This competition challenges you to automate the right whale recognition process using a dataset of aerial photographs of individual whales. Automating the identification of right whales would allow researchers to better focus on their conservation efforts. Recognizing a whale in real-time would also give researchers on the water access to potentially life-saving historical health and entanglement records as they struggle to free a whale that has been accidentally caught up in fishing gear.

From what I can gather, the winning solution was submitted by deepsense.io. They've written a full blog post about it here: 

http://deepsense.io/deep-learning-right-whale-recognition-kaggle/

Wildbook / IBEIS. Open-source effort to combine web-based mark-recapture database with ML/CV photo detection and identification.  http://wildbook.org
[ Full disclosure: I am a member of the non-profit team working on this project! ]

Hypraptive and Brown Bear Research Network collaboration to develop a deep learning, brown bear face identification system: BearID Project.

[Disclosure: I am a member of hypraptive, and maintain the hypraptive blog]

Hi Will, 

We are collaborating with WCS, San Diago zoo and the Wildtrack team from last year's zoohackathon to build a similar solution to the one you describe. The solution uses Twilio and you get a full overview of the work already built here: https://devpost.com/software/wildtrack

It would be good to share requirements and see if there is a way the Wildtrack solution can work for your needs. We would love to work with you to understand your full set of requirements and see if these can be accommodated. 

If this is of interest please do drop me an email at sophie.maxwell@zsl.org. I will also be out in Kenya from 25th April for two weeks, so if you'd like to meet we can hopefully do this face to face.

Sophie

Hi all

I'd be interested to hear if there were any updates on the discussion above. We @marianneteoh are looking at establishing informant networks for reporting illegal fishing and marine wildlife trade and a means of anonymous reporting would be incredibly useful.

Thanks in advance,

Kate

Hi Kate,

At ZSL, we're continuing to work with the WildTrack team on their solution - now named SmartSkout. We've just had an article published on the Guardian about the project https://www.theguardian.com/vulcan-partner-zone/2017/aug/22/new-app-enable-citizens-report-poaching-real-time 

We're currently focussed on producing the MVP for terrestrial SMART sites but we'd be very interested to hear more about your work establishing informant networks to see what potential there is for broader applications of the technology in the future. Feel free to connect with me at rachael.kemp@zsl.org 

Best

Rachael

Hi Mazidi,

A couple of options;

- Marco Polo.    http://eurekaproducts.com/rc-model-tracking-and-recovery/    

I know of people who have relocated their machine using this device.   

- Conventional VHF for wildlife tracking.  old but solid technology.

Also have a look at commercial / open source devices used for model rockets.

regards

MJ

EDIT: I see that you have already identified Marco Polo.

Hi Mazidi, all the systems that I recommended are separate systems with their own GPS, not connected to the drone.

Just the last one was the only finished product.

Good luck. Please tell us if you found something good and cheap!

Abstract here from the IMC Congress in Perth, they found the higher-res versions are much better than the less expensive (e.g. iphone) versions when it comes to finding animals. 

Now you see me: How thermal imaging increases detection of cryptic nocturnal mammals in tropical regions

Avril Underwood1 , Mia Dehre-Miller2 1. James Cook University, Smithfield, Queensland, Australia 2.

University of Lancaster, Lancaster, United Kingdom Accurate measures of species abundance and distribution underpin effective wildlife management and conservation strategies. Obtaining these data is very challenging when researching the cryptic nocturnal arboreal mammals of in the dense forest of the Wet Tropics. Our research shows that using a hand-held thermal imager significantly increases the detection rate of all mid-sized arboreal mammals when compared to the most commonly used method of spotlighting. We determine the optimal environmental and mechanical parameters for thermal imaging, and we investigate ‘cost-versus-result’ of hand-held imaging devices currently available on the market. We suggest that hand-held thermal imaging is a great emerging technology for cryptic arboreal mammal detection across tropical forested regions and beyond.

My own experience is with the FLIR iphone attachment, just informal testing, but hasn't been that great at finding animals.  I also think it would work better in the early AM.  In the evening there can still be a lot of rocks and trees still warm from the sun.

Hi Gabriela, i`ll check the game. Very interesting!!

Hi guys,

I'll be staffing the Miradi Software booth that Nick mentioned. So in case you don't make to any of the sessions, you can swing by our booth and ask questions, learn more about Miradi.org and MiradiShare.org and/or take part in our raffle to win  free subscription for Miradi.

Looking forward to seeing you all real soon,

Felix

Hi Chloe,

I've had promising results with the Scoutguard MG983G-30M, and I believe a 4G version has just been released. The 3G version has some useful features, like two-way communication to change settings/get images and a audio call feature (which I haven't used). I'd be very pleased to hear how your tests go.

Cheers,

Rob

@@Chloe+Aust -  pls google “ltl acorn uk”， you will get a lot info.  No, no limited, just the company's sale rules. You could use it anywhere, but for example, if you wants to use it in kenya, you need a local sim card, such as safaricom... when you check photos from china, you will find most of them photoed by are ITI acorn. Panda,tiger, snow leopord, cloud leopard, leopard, golden cat.... 

Scoutguard is also a good china camrea traps. oldest company. only company have espano menu. but they are focus on US hunting market. not special for wildlife researching use.

The LOREDA is also good, the only brand focus on wildlife-bio researching camra traps in china. Might be do not have dealers/stock in uk.