Safe and sound

Sound Event Recognition-based Vigilance for Alerting and Localisation.

Where humans are, there is noise. Although rowdiness of humans may be irritating, we use it in our favour. Indeed, poachers too make noises that give them away. The challenge is whether we are able to distinguish between ‘normal’ human sounds and those of poachers at work. And, of course, whether we can pick up these sounds in remote and desolate areas.

To determine the meaning of a sound, one has to interpret it. The sound of a car, for example, may be very common in areas where people go on safari, but much less so at night. This is true for most types of sounds. Their meaning varies with its context.

To develop a smart sensor which can interpret contextualised sounds, SensingClues sought the help of M+P (see earlier post) and Sensory Cognition Group of the Rijksuniversiteit Groningen. Both are reputable for their acoustic expertise. Both accepted the challenge.

The working name for our sound sensor is SERVAL, an acronym for Sound Event Recognition-based Vigilance for Alerting and Localisation. Once we finished the design we will search for a funds to develop it, and add it to our anti-poaching toolkit. 

We help rangers to turn wild spaces into safe havens!

Kasigau: poachers detected!

Tsavo East, Tsavo West, and the Kasigau Wildlife Corridor which connects them form the largest natural reserve in Kenya. Elephants, lions, giraffe, and many other species roam the area freely. It’s a lust for the eye. A treasure to cherish!

A real threat to this paradise is poaching. Not only to the animals, but also to the (unarmed) rangers that protect them. Two years ago two of them were shot by poachers. One ranger was killed. The other survived, but has a prosthetic shoulder since. This hero is still a ranger.

To help the rangers to stop poaching, SensingClues joined Tech-For-Tusks, a  project in which Wildlife Works, ShadowView, ProTrek, Microflown, NCSR, and SensingClues collaborate to protect elephants. After months of hard work, this was a very exciting opportunity to test our newly developed early warning system.

With baboons looking over our shoulders, long days and nights were spend to test, improve, and re-test our sensors. Until we were satisfied. Since this weekend our camouflaged sensors are on the alert for poaching activities. Tireless. Relentless. Once a poacher is detected, an alert is send to the patrolling rangers, directing them straight away to the actual threat.

Last Wednesday, only days after the sensors had been placed, the first alert came through. Human behaviour had been detected near a pond which elephants regularly visit. Being a restricted area, no people are allowed there. Verification learned that it was not a ranger. Hence, this was an all hands on deck situation!

A short while later we learned our poacher was actually a lost soul, taking her father on a safari. Although we were a bit disappointed with this outcome, the system had proven its strength. We’ve been able to detect human activity and alert the rangers in real-time!

It’s a really nice feeling to put the rangers on the right track. It’s good to know they are on the alert and ready to respond..

SensingClues is going to protect Nguvu!

Last week was a fantastic week!

  • for Nguvu, because our early warning and fast-response systems are going to help rangers to keep him and his family safe;

  • for SensingClues, because we gained our first contract!

SensingClues is partnering with Wildlife Works, ShadowView, Protrek and NSCR in the Tech for Tusks project. The project aims at developing and testing hi-tech systems to protect elephants in the Kasigau Wildlife Corridor, Kenya. We bring in our expertise and early warning and fast-response technologies.

The last few months we have been working hard to make our technologies ready. We’ve tested and demonstrated them in the Netherlands. Our potential customers and techno team were satisfied. So, now it’s time to test them in the harsh conditions of the African bush! Recent incidents, including ones that made the world-wide press, urge us to wast no more time.

The project is starting mid October. The tests will take about one month. We’ll keep you posted through  Twitter and this blog!


Detecting presence of bush crafter

In a southern nature reserve in the Netherlands, the State Forestry Department has spotted a tree hut, made by a bush crafter. In its neighbourhood rangers have found snares, suggesting that the bush crafter is also a poacher.

Because the rangers are too few in number, the forests too extensive, and the visits of the bush crafter irregular, hiding and waiting until the bush crafter arrives is not an option.

Hence, the local ranger suggested us to test our Trespasser sensor here. The idea was simple: we would hide our sensor near the tree hut. Once someone would stay near that hut for more than 15 minutes, an alert would be send to the rangers, enabling them to catch him red-handedly.

So we did. This week we placed the sensor. However, it did not come online. A small investigation learned us that network roaming was the problem. Indeed, the area was very near the German border. It is a minor problem, but to fix it we had to take the sensor back.

Conclusion: this time the bush crafter was lucky. But not for long. We’ll be back shortly. From then on our sensor will inform the rangers when the bush crafter arrives at the scene…

Sensor-journey day

The challenge:

Measure the presence of a poacher at work and send an alarm to a group of fast-responders.

The solution:

You want to use the technology without worrying about its complexities. Like your computer or phone: it should just work! We took that at heart and have been working on it for several months now. With a little help of our partners we developed a first beta-version of a human-presence detection sensor, a web-based intelligence platform (SCCSS), and Cluey, our fast-response app. Today the sensor measurements completed the full journey – a memorable moment!

Within a few weeks we will be testing our solution in the Dutch forests. Within a few months this technology will have matured enough to be tested in the tough operational field conditions of Kenya.


M+P announces its partnership with SensingClues

M+P announces its partnership with SensingClues:

The wilderness should become a safe place for wildlife again. At the moment it is not because of poachers and other hostile human activities. SensingClues develops solutions for this problem. Recently, we have become a strategic partner for them. Together we develop smart sound- and vibration sensors that can track human threats at an early stage so that authorities can take appropriate action.

Read the full English version.

Lees de volledige Nederlandstalige versie.

We are partnering with MarkLogic!

As of today we are working with MarkLogic to develop our SensingClues Cloud Services Suite (SCCSS). Our first tool in this suite is FOCUS. FOCUS is an interactive real-time big-data analytics tool. It is designed to support heuristic analysis, performed by crisis response decision makers, rather than data scientists. As such it is the tool for both centralised and distributed field operations’ intelligence and command centers.

MarkLogic is the only Enterprise NO-SQL database (Not Only SQL). It is a rocketing new generation storage and retrieval tool that is built for dealing with today’s high variety of data, high volumes, and high velocities. In other words, designed for big data. Think sensor data, geographical data, open source social media data, geo-tagged photo’s, etc. Anything that can be used to inform decision making within fast response teams. And very important for supporting collaboration between stakeholders in the field: data security is built in. We thus can guarantee that information is only shared with those entitled. And we can share information across organisational boundaries without losing that control.

In short, as of today SCCSS is powered by MarkLogic !

Bridge the gap!

Poachers, illegal loggers, and the like weigh their potential gain against the risk and price of being caught. By catching them red-handedly, all evidence is present for conviction. Raising the chance of being caught, discourages poachers. But how can we increase the chance of apprehension?

Approaching the problem from an information, knowledge and expertise coordination angle, we identified 6 gaps that have to be bridged in order to be successful. Through advice and hands-on help,  SensingClues helps you to bridge these gaps.

1: the alerting gap

When did you learn about the threat? And when did it actually start? Knowing what is happening and when it is happening is crucial for catching poachers in the act. Our human senses are very localised and restricted to seeing, hearing, feeling,  tasting, and smelling. Technological sensors can do all that. In addition, they perform 24/7 and extend senses to ranges that cannot be heard, seen or perceived by our human senses.

Sensor-technology challenges are manifold, but solvable. First, you have to know what discriminates poaching behaviour from other forms of human behaviour. Second, you have to think about clever ways to detect them: what sensor can be used, where should they be positioned. And third, you have to make the sensor-data available in real-time. In remote areas this can be quite difficult. The sensors have to be connected to the web, self-supporting in energy, and robuust enough to withstand its environment. More and more promising examples become available (e.g. chainsaw and tresh-passing detection).

SensingClues believes that no-one can stop the ongoing slaughter of precious species alone. If we want to be effective, we must collaborate. It’s time to join forces! Hence, we work with a wide range of partners to make technologies available and affordable to protectors all over the world. To stimulate collaboration we are developing the SensingClues Cloud Services Suite – SCCSS (pronounced as ‘success’), thus avoiding that users have to acquire, install, learn, and maintain a multitude of systems.

2: the intelligence gap

Intelligence is actionable information that gives you the edge over your opponent. Intelligence enables you to out-smart poachers. The question is: how good are you at collecting data? in using it to create intelligence? and in leveraging it to the rangers and other protectors in the field? With divers data sources, formats, forms of ownerships, etc., this question is not easily solved.

To integrate, handle, analyse, and leverage the information we are developing an interactive real-time intelligence toolFOCUS. By structuring 5 key-elements of data that inform action, the tool helps to focus on what is important to start and coordinate a fast response.

FOCUS is especially designed for fast-response professionals and does not require statistical or technological skills, as is often the case in business intelligence software. It also allows for distributed access and collaboration, as may be expected from a net-centric solution.

3: the action repertoire gap

Responding to a threat may be very complicated when the poachers are larger in numbers, armed, and dangerous. Or if they are simply out of reach due to the distance that has to be bridged. Working together with non-professional first-responders is delicate, but possible. This may start by asking people in the area to function as “eyes and ears” of law enforcers. But other tasks such as scaring away animals or creating blockages are within reach.

SensingClues has developed the SCUM-method to explore effective strategies involving multiple partners.

4: the response-time gap

Poachers preferably work when protectors are far away or off-duty. The time needed to formulate an appropriate response and mount an operation, is used by poachers to escape. Thus, a short response-time is critical.

Based on the principles of net-centric organising, SensingClues provides hand-on advise to decrease the response-time gap.

5: the evidence gap

Often poachers are gone by the time rangers arrive at the scene. But they seldom left without a trace. Knowing how to approach a poaching site safely, and to gather and preserve forensic evidence that may later be used in court, is critical.

Through our network of law enforcement professionals we can organise “art of inspection” workshops and provide field training to rangers.

 6: the technology gap

Technology is a sine-que-none factor in developing effective poaching solutions. Without proper technology, such as sensors, drones, information and communication technology, and coordination technologies, protectors make no chance to beat poachers. The challenge is the last mile. That is, from the internet and office environment to the user in the field: point-to-point real-time interaction, while all points are moving from one place to another.

It’s our strength and pride to help our partners to bridge that last mile.

Kickoff meeting SensingClues

Last Friday we organized our first SensingClues kickoff session! Nine highly qualified and enterprising specialists were invited to hear about our ambitions. And to explore opportunities to join or collaborate. Indeed, our mission to turn wild spaces into safe havens requires a cutting edge and daring team plus a network of skillful partners!

Of course we ran out of time. But it was enough to convince everyone present that SensingClues’ mission and business model are inspiring, solid, and manageable. On the 5th of June we will continu our talks.

Hello world, SensingClues is live!

Friends!

SensingClues has been launched. As of today we start developing our services to turn wild areas into safe havens for wildlife. What a happy day this is!

Through this medium we will keep you posted of our projects. Take your time to check-out our website and let us know what you think.

We believe that if all involved in protecting wildlife alliance against poaching, we should be able to accomplish our mission. We have no time to loose, let’s get going!

Multi-sensor project

People make sounds that differentiate them from other species. SensingClues is working on a smart sensor that can record sounds on large distances, can differentiate between types of sounds, and which is accurate. The sensor needs to be self-sufficient on energy and should last unattendedly for at least 1 year. Moreover, the sensor is being designed to withstand heat, cold, water and dust. The sensor will be able to connect with the internet – where-ever it is deployed – to transmit its alerts and to enable distant upgrades. And last but not least, it is going to be affordable!

In this project we work together with scientists, (electrical) engineers, and programmers.

Do you want to bring in your expertise to make the sensor even smarter, cheaper, or more robust? Contact us to discuss opportunities.

Are you a nature reserve manager or anti-poaching program manager and do you have suggestions or questions regarding the capabilities of this sensor? Please let us know. We would love to talk with you about the specific requirements you have!