Anomaly Detection System Development

By: Vikas Chinchansur (Vicky)

Objective

The objective of the project is to modify/develop capability in the CSIRO data system to invoke the ML model on real-time data streams.

Vision

To improve trust of real time water quality data by having an artificial expert classify the data as good or bad.

Technology

The classification system developed will be deployed on the cloud (AWS) and run periodically on the incoming data streams.

Who are we

view-Great-Barrier-Reef-Australia-coast.

The Problem

Nitrogen losses from intensive crop production into wet tropical catchments of north Queensland are a significant threat to the health of the Great Barrier Reef. However, lack of information on the losses and the extreme variability in climate and rainfall over small distances, it's been difficult for growers to link nitrogen that appears in the environment and nitrogen management on their farm

Problem

The Solution

1622 Suite of Apps

CSIRO is developing a suite of apps, called 1622™, to deliver information services to farmers that address their concerns, make for faster change in farm management practices and help growers reduce impacts of cropping on the Great Barrier Reef. The 1622™ apps are at different stages of product development, with the earliest app, 1622WQ™ ('water quality'), ready for public release very soon.

1622WQ App

1622WQ™ provides real-time water quality information. It allows farmers to see, for example, the influence of recent rainfall on water quality or whether management actions such as recent fertilizing has affected nitrogen losses. Farmers can also see the seasonal climate outlook to help plan prior.

the solution

The Result

The final expected outcome is that 1622™ meets industry needs. For the first time, farmers will have real-time information on critical factors for growing sugarcane. When complete, 1622™ brings together information on sugarcane production and environmental performance to help farmers: evaluate their crop management, facilitate better decisions, and help them protect the Great Barrier Reef. The below slideshow shows the results generated.