Summary of Deep Learning For Automated Multi-scale Functional Field Boundaries Extraction Using Multi-date Sentinel-2 and Planetscope Imagery: Case Study Of Netherlands and Pakistan, by Saba Zahid et al.
Deep Learning for automated multi-scale functional field boundaries extraction using multi-date Sentinel-2 and PlanetScope imagery: Case Study of Netherlands and Pakistan
by Saba Zahid, Sajid Ghuffar, Obaid-ur-Rehman, Syed Roshaan Ali Shah
First submitted to arxiv on: 24 Nov 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 study explores the use of deep learning semantic segmentation architecture and multi-temporal satellite imagery to improve functional field boundary delineation in two distinct farming systems: Netherlands and Pakistan. Four UNET-based models were evaluated using different combinations of multi-date images and NDVI stacks in the Netherlands, with a focus on IoU scores. The findings were then applied for transfer learning, using pre-trained models from the Netherlands on the Pakistani dataset. Results show that multi-date NDVI stacks provide additional temporal context, reflecting crop growth over different times of the season. The study highlights the importance of diverse geographical areas and fine spatial resolution for field boundary extraction in small-scale farming. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The research looks at how to use satellite images and computer learning to better define farm borders. They compared different ways of using images from different times of year and found that combining them helps a lot. This can be useful for farmers who want to automatically find their fields without having to manually map them out. The study shows that it’s important to have data from many different places and at high resolution to get the best results. |
Keywords
» Artificial intelligence » Deep learning » Semantic segmentation » Transfer learning » Unet
