Agricultural Landscape Understanding At Country-Scale
Authors:
Radhika Dua,
Nikita Saxena,
Aditi Agarwal,
Alex Wilson,
Gaurav Singh,
Hoang Tran,
Ishan Deshpande,
Amandeep Kaur,
Gaurav Aggarwal,
Chandan Nath,
Arnab Basu,
Vishal Batchu,
Sharath Holla,
Bindiya Kurle,
Olana Missura,
Rahul Aggarwal,
Shubhika Garg,
Nishi Shah,
Avneet Singh,
Dinesh Tewari,
Agata Dondzik,
Bharat Adsul,
Milind Sohoni,
Asim Rama Praveen,
Aaryan Dangi
, et al. (10 additional authors not shown)
Abstract:
Agricultural landscapes are quite complex, especially in the Global South where fields are smaller, and agricultural practices are more varied. In this paper we report on our progress in digitizing the agricultural landscape (natural and man-made) in our study region of India. We use high resolution imagery and a UNet style segmentation model to generate the first of its kind national-scale multi-…
▽ More
Agricultural landscapes are quite complex, especially in the Global South where fields are smaller, and agricultural practices are more varied. In this paper we report on our progress in digitizing the agricultural landscape (natural and man-made) in our study region of India. We use high resolution imagery and a UNet style segmentation model to generate the first of its kind national-scale multi-class panoptic segmentation output. Through this work we have been able to identify individual fields across 151.7M hectares, and delineating key features such as water resources and vegetation. We share how this output was validated by our team and externally by downstream users, including some sample use cases that can lead to targeted data driven decision making. We believe this dataset will contribute towards digitizing agriculture by generating the foundational baselayer.
△ Less
Submitted 8 November, 2024;
originally announced November 2024.
Applied Deep Learning to Identify and Localize Polyps from Endoscopic Images
Authors:
Chandana Raju,
Sumedh Vilas Datar,
Kushala Hari,
Kavin Vijay,
Suma Ningappa
Abstract:
Deep learning based neural networks have gained popularity for a variety of biomedical imaging applications. In the last few years several works have shown the use of these methods for colon cancer detection and the early results have been promising. These methods can potentially be utilized to assist doctor's and may help in identifying the number of lesions or abnormalities in a diagnosis sessio…
▽ More
Deep learning based neural networks have gained popularity for a variety of biomedical imaging applications. In the last few years several works have shown the use of these methods for colon cancer detection and the early results have been promising. These methods can potentially be utilized to assist doctor's and may help in identifying the number of lesions or abnormalities in a diagnosis session. From our literature survey we found out that there is a lack of publicly available labeled data. Thus, as part of this work, we have aimed at open sourcing a dataset which contains annotations of polyps and ulcers. This is the first dataset that's coming from India containing polyp and ulcer images. The dataset can be used for detection and classification tasks. We also evaluated our dataset with several popular deep learning object detection models that's trained on large publicly available datasets and found out empirically that the model trained on one dataset works well on our dataset that has data being captured in a different acquisition device.
△ Less
Submitted 22 January, 2023;
originally announced January 2023.
