Smart Natural Assets

We're using a ground sensors, drones, mobile data, and AI to understand tree health and socio-ecological parameters in urban green spaces.

Smart Natural Assets

Things used in this project

Ā 

Hardware components

HARDWARE LIST
1 Raspberry Pi 4 Model B
1 DFRobot FireBeetle ESP32 IOT Microcontroller (Supports Wi-Fi & Bluetooth)
1 Inertial Measurement Unit (IMU) (6 deg of freedom)
1 terros 12 [soil health sensor- soil volumetric water content, temp, dielectric constant]
1 Adafruit Waterproof DS18B20 Digital temperature sensor
1 DFRobot Gravity: Analog Capacitive Soil Moisture Sensor- Corrosion Resistant
1 Seeed Studio Soil Moisture & Temperature Sensor
1 dji drone

Software apps and online services

Ā 

Arduino IDE

Ā 

python

Ā 

Microsoft Azure IoT Hub

Ā 

influxDB

Ā 

Grafana

Ā 

Microsoft Custom Vision

Story

Ā 

We're a research team at the University of British Columbia from Electrical and Computer Engineering and Forestry. We're studying tree health and the associated social and ecological values within urban green spaces.

We're using a combination of technologies in our study, including IoT sensors, drones, mobile data, and AI to build our IoT/AI cloud-cellular solution architecture.

Ā 

The data is sent to Azure IoT Hub and to open-source tools for data storage and visualization, including InfluxDB and Grafana.

Ā 

More info here:

Ā 

http://brianchami.com/index.php/natural-assets/

Ā 

We're also teaching a work session series on Drones + Custom Vision [Funded by NRCan], where we teach participants:

Ā 

Introduction to drones and Drone data typesĀ Drones in Sustainability and Urban Green SpacesĀ Intro to Custom VisionĀ Training a Custom Vision Model to count and classify tree typesĀ Deploy the model on a user-friendly webapp to potentially be used by practitioners, city planners and mangers, community leaders.Ā 

The idea is use tech to engage cities and communities, especially remote communities that lack the necessary tools and expertise in data analysis.

Ā 

Work Session Link:

Ā 

https://aggregate-intellect.notion.site/Drone-Image-Classification-of-Urban-Green-spaces-214499d7dc4840e0a69ea556963421b8

Ā 

We're also thinking about integrating the system with a Custom Vision-based tool for wildlife monitoring. [if there's enough time]:

Ā 

http://brianchami.com/index.php/cameratrap/


Ā 

The article was first published in hackster, June 9, 2022

cr: https://www.hackster.io/ejri2/smart-natural-assets-8f8fdf

author: ejri

License
All Rights
Reserved
licensBg
0