Wild Elephant Approaching Alert

Photo: Deccan Chronicle

Wild elephants and villagers near nature reserves are often caught up in the human-wildlife conflict. Even in countries where the elephant is sacred, this may lead to elephants being killed by villagers. Out of protection, by accident (as on the photo), or out of revanche.

Elephant rumbles may be detected at distances from 5 to 10km, depending amongst other things on vegetation. Being able to detect them long before the enter the village gives both the villagers and the elephants a better chance of survival.

To this end we are developing an acoustic elephant detection and monitoring system. It requires knowledge of elephant behaviour, knowledge of acoustic signal processing and analysis, and tough engineering skills to get a fully functional system that survives the forces of nature. To complement our team we sought collaboration with various esteemed researchers, specialised in elephant vocalisations and machine listening:

The Wild Elephant Approaching Alerting system is based or our acoustic SERVAL sensor. The SERVAL sensor is connected to IoT sensor platform SCCSS, allowing us to further analyse the signals and send out alerts to all rangers and/or community members involved in a project.

Current status (30 June 2017)

We managed to develop the SERVAL-sensor. That is, a sensor system which can be trained to distinguish different types of sounds. To this end we use a combination of known physical sound characteristics and deep learning. The trained neural networks are being deployed on the sensor, which sends out alerts to the SCCSS-platform, from where it can be dispatched to early responders.

We are at the brink of training the sensor with elephant sounds samples to prepare field sensors tailored for detecting approaching elephants and the detection of trapped elephants. To this end, the following specialists have joined our efforts:

  • Peter Wrege, Director of the Elephant Listening Project at the Cornell Lab of Ornithology
  • Blaise Droz, Documentaliste indépendant et Journaliste chez Journal du Jura

With the help of Karol Piczak we managed to deploy the trained convolutional (neural) network on a Raspberry Pi and are now working on power optimisation.

On 7 July we will provide the students of the Jheronimus Academy for Data Science with the hackathon challenge to outperform the work of Hugo, our chief data science.

As soon as funds are secured for field testings, we will conduct the first field trials.

Can’t wait? Contact us to discuss how we can help each other to speed it up!