I started building Rasterly as a side-project to be able to dynamically visualize drone and satellite imagery of any size. You can calculate and view spectral indexes too!
The goal is to make it really easy for drone surveyors, conservationists, researchers, and so on to share aerial imagery without anyone needing to download anything.
No customers yet. Focusing on improving the performance of the tile server and preprocessing of images, and of course trying to connect with potential users! Struggling to transition into a salesman though.
Grapefruit juice can also increase the amount of a drug absorbed into the bloodstream. It blocks an enzyme in the small intestine that breaks down many types of drugs.
Yes definitely. I’m also curious to know, what are Ukrainian software contractors doing now? And assuming this war ends, what, if anything, will they need to rengage with foreign companies?
Quite likely! I'm planning on making the gorillas identification project open source as there's actually quite a lot to it (the core AI functionality, management of lots of images and gorillas data, making sure it all works fine on a crappy connection with old Androids).
LoRaWAN is used for an alert system in an area where local communities are threatened by an armed group called the ADF (aka ISIS-DRC). In the future we hope to roll out a large-scale LoRaWAN to monitor wildlife (elephants, lions) as well as vehicles and rangers on patrol.
Gorilla tracking devices are likely to be intrusive and cumbersome for them. They are also known to help each other out and remove devices. For now, gorillas are tracked on foot by a team of rangers.
For now we are using Azure Custom Vision and have a working demo that achieves solid results. This seems sufficient for now. This will fit into a web app that uses React and Django.
I can really only speak to conservation to Congo. One thing I've noticed is that conservation organizations in DRC take on a lot more responsibilities than just wildlife conservation. In Virunga National Park, for example, the park has built and operates a power utility to provide an alternative to charcoal. This is made up of four run-of-the-river hydroelectric plants, hundreds of kilometers of distribution lines (high and medium tension), and smart meters to connect customers. You can check out virunga.org to learn more!
In fact, we are basically doing facial detection based on cropped photos of gorillas around the nose. Thanks for the links! I'll be sure to check it out.
You're definitely right that DRC wants to move up the value chain. Most raw materials are shipped out of the country for processing or smuggled to neighboring states (gold). However, the impact on wildlife is most certainly negative. The economic development impact is pretty unclear; DRC has shown time and again that mineral wealth is equitably distributed. Despite the jobs some new processing plants might afford, the profits from these operations will likely just line the pockets of those in charge.
I spent two years working in DRC and have now moved to a remote position (based in the US). There are a few organizations I've worked with that do great work: Allen Institute for AI (EarthRanger), SmartParks.org and Conservation X Labs. There are surely others, but these offer products directly to national parks to improve their capabilities.
I work for a national park in the Democratic Republic on the Congo as a tech lead. Some of the things we are doing: LoRaWAN for tracking and emergency response, ML to identify gorillas by their unique nose prints, and long-range drones for mapping and surveillance, and a management web app.
If you’re interested in conservation / sustainable development and associated technologies let me know! Always looking to collaborate and bounce ideas off others.
I ran a curl test and noticed that the TCP connect time is where things are getting hung up, taking over 80 seconds! Does anyone know if this is indicative of deliberate rate limiting or just a bad peering connection?
I started building Rasterly as a side-project to be able to dynamically visualize drone and satellite imagery of any size. You can calculate and view spectral indexes too!
The goal is to make it really easy for drone surveyors, conservationists, researchers, and so on to share aerial imagery without anyone needing to download anything.
No customers yet. Focusing on improving the performance of the tile server and preprocessing of images, and of course trying to connect with potential users! Struggling to transition into a salesman though.