|Year||2017 to 2020|
Steven SuProject LeaderHigh-level design of electronics circuit and data analysis algorithm solutions for Data Arena
Renping LiuProject LeaderHigh-level design of remote server and cybersecurity solutions for Data Arena
Shari L. ForbesProject LeaderOdour analysis expert in the chemical analysis area. Test protocol design and test results analysis
Maiken UelandChemical analysis and odour analysis expertUsing GCGC-TOFMS to double-check the test results and odour data analysis
Wentian ZhangNOS.E system and hardware developerRemote server and equipment development
Taoping LiuNOS.E algorithms developerData analysis algorithms development
Amber BrownChemical analysis expertUsing GCGC-TOFMS to double-check the test results and odour data analysis
Ben SimonsData Arena Technical DirectorTech Lead Support
Darren LeeData Arena Software DeveloperInitial Software Support
The Australian Seafood Quality Index highlights the importance of understanding and monitoring fish freshness with chilled seafood.
There are opportunities to improve on this Quality Index (QI); a scoring system based on demerit points. The index might be based on a visualization of fish odor at different freshness levels, measured with an electronic nose.
The UTS NOS.E team is an interdisciplinary electronic nose research & development team. The members range from chemical science, computer science, electrical engineering, mechanical engineering, and statistical analysis backgrounds, who work collaboratively to achieve the common goal: developing NOS.E.
Our research experience is not based solely on academics but extends to real industrial research and development procedures. The UTS NOS.E team has built an electronic nose system, which has been tested for monitoring the freshness of beef (2016).
The test result shows the beef freshness index over ten days as a log likelihood form by using one of the Data Arena visualization methods, which can provide valuable information to beef quality assessment.
The electronic nose (E-nose) is a device that mimics the mammalian olfactory system and is designed to detect and classify different aroma mixtures. Supported by the powerful and immersive UTS Data Arena, in this project the most recent E-nose prototype constructed by UTS NOS.E team was further developed through "dynamical feature visualization" and feature optimization for the application of fish freshness monitoring and assessment.
UTS NOS.E system can also identify six different kinds of perfumes with a higher accuracy rate (72%~100%). This research indicates that the electronic nose techniques developed within UTS are a practical solution for quality evaluation of beef freshness and has great potential for measuring fish freshness.
In 2018 we therefore proposed to implement our system and UTS Data Arena to fish freshness assessment usage scenario as a rapid monitoring method for the fish market.
The desired aims of this project are as follows:
Improve the working efficiency of the inspector from the fish market and save the labor cost.
Improve the safety and quality of the fish by monitoring the freshness index during the whole blockchain.
Build a good reputation for UTS and the Sydney Fish Market.
The new (2018) E-nose prototype was equipped with a well-constructed gas sensor array. Based on its sensitivity to typical volatile compounds emitted during fish spoilage, together with a machine-learning-based pattern-recognition module, the prototype aims to graphically display the "freshness" of the fish during the storage and transportation.
NOS.E has potential market value for use in the Sydney Fish Market
A Visualisation of the responses of the multi-sensors over time was implemented in the UTS DATA ARENA. Various combinations of sensors were visualized in order to find the signatures of gases in different dimensions.
The DATA ARENA was used to view the data in parallel coordinates by using the parallel coordinates builder in the [Data Arena Virtual Machine (DAVM)].(https://dataarena.net/dive-in/wiki/davm)
According to the research outcomes of the Data Arena Research Exhibition Grant in 2016, we found UTS Data Arena is useful analysis tool for the NOS.E researchers and a potential attractive classification solution for the customers.
In the 2018 Data Arena Research Exhibition Grant, we took a further step. With the power of Data Arena to visualize the variations in odor of different stages of fish during storage and transport, we developed a Data Arena based fish freshness index calculator.
The Sydney Fish Market is interested in this E-nose technology. The purpose of this project was to develop Data Arena based visualization tools to vividly display the drift of odor features in multiple angles and multiple dimensions. Such a visualization can optimally demonstrate the effectiveness of our technologies, and support our R&D investment proposition.
Based on some preliminary test results and the research outcomes from our Data Arena Research Exhibition Grant 2016, we found the UTS Data Arena is a useful analysis tool for the NOS.E researchers.
We improved the data analysis efficiency and created an attractive classification solution for the customers
In 2016 we fully visualized the dynamic change of unique features in beef odor using Data Arena data visualization tools.
Our follow-up work in 2018 visualized and captured feature drift associated with the changes in fish odor and its chemical component variation with a visual confidence afforded by the power of the Data Arena.
In 2016, we reduced the number of gas sensors for the detection of ivory and rhinoceros horn samples by using Data Arena visualization tools.
Likewise in 2018 we also reduced the number of gas sensors required for fish freshness assessment.
Sydney Fish Market has shown a great interest regarding the use of the Data Arena to visualize the fish freshness. It provides a more accessible and efficient interface for the fish freshness index.
Instead of a boring data analysis report, the UTS Data Arena's 360-degree interactive data visualization facility helped us to attract more attention and support