Classification of measured unsafe liquids using microwave spectroscopy system by multivariate data analysis...

Area of Science:

• Applied Physics and Engineering
• Data Science and Machine Learning

Background:

• Aviation security requires effective methods for identifying hazardous materials.
• Current screening methods face challenges in distinguishing diverse liquid substances.

Purpose of the Study:

• To evaluate microwave measurement techniques for detecting illegal and explosive materials.
• To classify passenger-preferred liquids using multivariate data analysis for aviation security.

Main Methods:

• Utilized microwave measurement in a free space reflection configuration across various frequency ranges.
• Applied multivariate data analysis, specifically the self-organizing maps (SOM) algorithm, for liquid classification.
• Focused on distinguishing properties of liquids commonly carried by air travelers.

Main Results:

• Microwave measurement proved effective in identifying hazardous materials.
• The self-organizing maps algorithm successfully clustered and distinguished measured liquids.
• The chosen measurement method demonstrated the ability to reflect unique liquid properties.

Conclusions:

• Microwave measurement combined with self-organizing maps offers a responsive and effective solution for hazardous material detection.
• This approach enhances the ease and accuracy of classifying liquids in aviation security screening.
• The study successfully identified a convenient measurement method for unique liquid property reflection.