Summary of Enhanced Infield Agriculture with Interpretable Machine Learning Approaches For Crop Classification, by Sudi Murindanyi et al.
Enhanced Infield Agriculture with Interpretable Machine Learning Approaches for Crop Classification
by Sudi Murindanyi, Joyce Nakatumba-Nabende, Rahman Sanya, Rose Nakibuule, Andrew Katumba
First submitted to arxiv on: 22 Aug 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 The abstract discusses the advancements in image classification for agricultural applications using Artificial Intelligence (AI) and Machine Learning (ML). Researchers evaluated four approaches: traditional ML with handcrafted feature extraction, Custom Designed CNNs, transfer learning on established DL architectures, and cutting-edge foundation models. The study highlights the limitations of existing techniques and emphasizes the importance of model explainability. Specifically, it presents the explainability of the Xception model using LIME, SHAP, and GradCAM. The results indicate that Xception achieved 98% accuracy with a model size of 80.03 MB and prediction time of 0.0633 seconds. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Image classification is helping farmers make better crop management decisions using AI and Machine Learning. Researchers tested four ways to classify crops: old methods, new computer-designed models, pre-trained models, and super-advanced foundation models. They found that all worked well, but one model called Xception was the best at predicting what crop it was looking at. It’s fast too! The team also showed how they could explain why this model made certain predictions, which is important for people to trust AI in farming. |
Keywords
* Artificial intelligence * Feature extraction * Image classification * Machine learning * Transfer learning
