Summary of A New Dataset and Comparative Study For Aphid Cluster Detection and Segmentation in Sorghum Fields, by Raiyan Rahman et al.
A New Dataset and Comparative Study for Aphid Cluster Detection and Segmentation in Sorghum Fields
by Raiyan Rahman, Christopher Indris, Goetz Bramesfeld, Tianxiao Zhang, Kaidong Li, Xiangyu Chen, Ivan Grijalva, Brian McCornack, Daniel Flippo, Ajay Sharda, Guanghui Wang
First submitted to arxiv on: 7 May 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 A novel approach has been developed to address the issue of aphid infestations in wheat and sorghum fields, which result in significant agricultural yield losses. The current method of employing chemical pesticides is inefficient and has negative health and environmental impacts. To create an intelligent autonomous system that can locate and spray large infestations selectively within complex crop canopies, a large multi-scale dataset for aphid cluster detection and segmentation was collected from actual sorghum fields and meticulously annotated to include clusters of aphids. Four real-time semantic segmentation models and three object detection models were trained and evaluated specifically for aphid cluster segmentation and detection. The results showed that Fast-SCNN delivered the most effective segmentation results, achieving 80.46% mean precision, 81.21% mean recall, and 91.66 frames per second (FPS). For object detection, RT-DETR exhibited the best overall performance with a 61.63% mean average precision (mAP), 92.6% mean recall, and 72.55 on an NVIDIA V100 GPU. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Aphid infestations in wheat and sorghum fields are a big problem that causes a lot of damage and loss. Farmers usually use chemicals to get rid of them, but this isn’t very effective and can be bad for people and the environment. Scientists have made a new system that can help find and spray aphids more efficiently. They collected a lot of pictures from real fields and labeled where the aphids were. Then they tested different computer models on these pictures to see which ones worked best. One model, called Fast-SCNN, did very well at finding and separating the aphids. Another model, called RT-DETR, was good at detecting objects like aphids. |
Keywords
» Artificial intelligence » Mean average precision » Object detection » Precision » Recall » Semantic segmentation
