A Decision Support Tool for Selection of Data Labelers to support Aided Target Recognition for the Department of the Army Science and Technology

Aided Target Recognition (AiTR) is a critical technology for U.S. Army systems. Aided Target Recognition relies upon machine learning/artificial intelligence-enabled Automated Target Recognition (ATR) algorithms, which require vast amounts of labeled data to detect, classify, and identify potential targets. The performance of the ATR algorithms is directly impacted by labeling quality. Data labeling is time-consuming, costly, administratively burdensome, and tedious (Graceffo, 2021). To reduce the time, cost, and processing burdens, it is important to partner with industry data labeling vendors to ensure high-quality labeled data is generated in a timely manner to enable ATR algorithm development and training. Commercial vendors can label large quantities of data quickly utilizing artificial intelligence tools and robust quality control practices. A Technique for Order of Preference by Similarity to Ideal Solution Multi-Criteria Decision Aid is needed to determine the optimal vendor to partner with for data labeling to support ATR algorithm development.Keywords: Multi-Criteria Decision Aid, Technique for Order of Preference by Similarity to Ideal Solution, Data Labeling, Computer Vision, Automatic Target Recognition

Author

• Fields, Michael Gerard

Language

• en

Keyword

• Multi-Criteria Decision Aid
• Technique for Order of Preference by Similarity to Ideal Solution
• Data Labeling
• Automatic Target Recognition
• Computer Vision

Date created

• 2023

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Widener, Donald

Committee Member(s)

• Young, Stephen
• Etemadi, Amir

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/9s161707c