Summary of Pahd: Perception-action Based Human Decision Making Using Explainable Graph Neural Networks on Sar Images, by Sasindu Wijeratne et al.
PAHD: Perception-Action based Human Decision Making using Explainable Graph Neural Networks on SAR Images
by Sasindu Wijeratne, Bingyi Zhang, Rajgopal Kannan, Viktor Prasanna, Carl Busart
First submitted to arxiv on: 5 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG); Image and Video Processing (eess.IV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Machine learning models like CNNs and GNNs are used in military applications to identify ground-based objects from Synthetic Aperture Radar (SAR) images. The goal is to determine the vehicle class, such as tanks or missile launchers, which can help commanders make decisions about whether a target object is an ally or enemy. Current ML algorithms provide feedback on recognized targets but leave final decisions to commanding officers. To improve decision-making, our framework provides detailed information alongside identified targets, including SAR image features that contributed to classification, classification confidence, and probability of misclassification. Our Graph Neural Network (GNN)-based ATR framework achieves an overall accuracy of 99.2% on the MSTAR dataset, outperforming previous state-of-the-art GNN methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers use special computers to help military commanders identify vehicles from pictures taken by radar satellites. They teach these computers to recognize different types of tanks and trucks by showing them lots of examples. The computers are very good at this job, but they don’t make the final decision about what to do with the information. Instead, the commander makes that decision based on how confident the computer is in its identification. To help the commander make a better decision, the researchers provide more details about the picture and why the computer thought it was one type of vehicle rather than another. |
Keywords
* Artificial intelligence * Classification * Gnn * Graph neural network * Machine learning * Probability
