Can AI see what we can’t?: Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction

Gautam Siddharth Kashyap; Harsh Joshi; Manaswi Kulahara; Rajkumar Dhakar; Atul Sajjanhar; Jiechao Gao; Sarthak Jain; Shahab Saquib Sohail

doi:10.1109/ICASSP49660.2025.10890344

Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction

Gautam Siddharth Kashyap, Harsh Joshi, Manaswi Kulahara, Rajkumar Dhakar, Atul Sajjanhar, Jiechao Gao^*, Sarthak Jain, Shahab Saquib Sohail

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

10 Citations (Scopus)

Abstract

Precise mapping of crop types and estimating yields are important in gauging agricultural diversity and yield potential, especially in regions dominated by small-scale farming. Nevertheless, these tasks are challenging due to factors such as small field sizes, inter-cropping, and a lack of sufficient ground truth labels for certain regions. In this paper, we propose an approach that combines advanced deep learning algorithms with Sentinel-2 and MODIS satellite data for improving the accuracy of crop type mapping and yield prediction. We used datasets from the main growing season of 2017 in Kenya (Bungoma, Busia and Siaya) coupled with county level yield data from US, Argentina and Brazil spanning from 2005 to 2016. Our models (CNN, SegNet, MaskRCNN, ResNet, UNet) were evaluated on both tasks i.e., classification of crop types and predicting yields.

Original language	English
Title of host publication	2025 IEEE International Conference on Acoustics, Speech, and Signal Processing
Subtitle of host publication	Conference Proceedings
Place of Publication	India
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Number of pages	5
ISBN (Electronic)	9798350368741
DOIs	https://doi.org/10.1109/ICASSP49660.2025.10890344
Publication status	Published - 7 Mar 2025
Externally published	Yes
Event	2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025 - Hyderabad, India Duration: 6 Apr 2025 → 11 Apr 2025

Conference

Conference	2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025
Country/Territory	India
City	Hyderabad
Period	6/04/25 → 11/04/25

Bibliographical note

Publisher Copyright:
©2025 IEEE.

Keywords

Agricultural
Crop Types
Deep Learning
Machine Learning
Yield

Access to Document

10.1109/ICASSP49660.2025.10890344

Cite this

Kashyap, G. S., Joshi, H., Kulahara, M., Dhakar, R., Sajjanhar, A., Gao, J., Jain, S., & Sohail, S. S. (2025). Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction. In 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing: Conference Proceedings Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ICASSP49660.2025.10890344

Kashyap, Gautam Siddharth ; Joshi, Harsh ; Kulahara, Manaswi et al. / Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction. 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing: Conference Proceedings. India : Institute of Electrical and Electronics Engineers (IEEE), 2025.

@inproceedings{72d9698d6bf64c2db7f7dd60b1b13a77,

title = "Can AI see what we can{\textquoteright}t?: Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction",

abstract = "Precise mapping of crop types and estimating yields are important in gauging agricultural diversity and yield potential, especially in regions dominated by small-scale farming. Nevertheless, these tasks are challenging due to factors such as small field sizes, inter-cropping, and a lack of sufficient ground truth labels for certain regions. In this paper, we propose an approach that combines advanced deep learning algorithms with Sentinel-2 and MODIS satellite data for improving the accuracy of crop type mapping and yield prediction. We used datasets from the main growing season of 2017 in Kenya (Bungoma, Busia and Siaya) coupled with county level yield data from US, Argentina and Brazil spanning from 2005 to 2016. Our models (CNN, SegNet, MaskRCNN, ResNet, UNet) were evaluated on both tasks i.e., classification of crop types and predicting yields.",

keywords = "Agricultural, Crop Types, Deep Learning, Machine Learning, Yield",

author = "Kashyap, \{Gautam Siddharth\} and Harsh Joshi and Manaswi Kulahara and Rajkumar Dhakar and Atul Sajjanhar and Jiechao Gao and Sarthak Jain and Sohail, \{Shahab Saquib\}",

note = "Publisher Copyright: {\textcopyright}2025 IEEE.; 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025 ; Conference date: 06-04-2025 Through 11-04-2025",

year = "2025",

month = mar,

day = "7",

doi = "10.1109/ICASSP49660.2025.10890344",

language = "English",

booktitle = "2025 IEEE International Conference on Acoustics, Speech, and Signal Processing",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

Kashyap, GS, Joshi, H, Kulahara, M, Dhakar, R, Sajjanhar, A, Gao, J, Jain, S & Sohail, SS 2025, Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction. in 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing: Conference Proceedings. Institute of Electrical and Electronics Engineers (IEEE), India, 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025, Hyderabad, India, 6/04/25. https://doi.org/10.1109/ICASSP49660.2025.10890344

Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction. / Kashyap, Gautam Siddharth; Joshi, Harsh; Kulahara, Manaswi et al.
2025 IEEE International Conference on Acoustics, Speech, and Signal Processing: Conference Proceedings. India: Institute of Electrical and Electronics Engineers (IEEE), 2025.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Can AI see what we can’t?

T2 - 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

AU - Kashyap, Gautam Siddharth

AU - Joshi, Harsh

AU - Kulahara, Manaswi

AU - Dhakar, Rajkumar

AU - Sajjanhar, Atul

AU - Gao, Jiechao

AU - Jain, Sarthak

AU - Sohail, Shahab Saquib

PY - 2025/3/7

Y1 - 2025/3/7

N2 - Precise mapping of crop types and estimating yields are important in gauging agricultural diversity and yield potential, especially in regions dominated by small-scale farming. Nevertheless, these tasks are challenging due to factors such as small field sizes, inter-cropping, and a lack of sufficient ground truth labels for certain regions. In this paper, we propose an approach that combines advanced deep learning algorithms with Sentinel-2 and MODIS satellite data for improving the accuracy of crop type mapping and yield prediction. We used datasets from the main growing season of 2017 in Kenya (Bungoma, Busia and Siaya) coupled with county level yield data from US, Argentina and Brazil spanning from 2005 to 2016. Our models (CNN, SegNet, MaskRCNN, ResNet, UNet) were evaluated on both tasks i.e., classification of crop types and predicting yields.

AB - Precise mapping of crop types and estimating yields are important in gauging agricultural diversity and yield potential, especially in regions dominated by small-scale farming. Nevertheless, these tasks are challenging due to factors such as small field sizes, inter-cropping, and a lack of sufficient ground truth labels for certain regions. In this paper, we propose an approach that combines advanced deep learning algorithms with Sentinel-2 and MODIS satellite data for improving the accuracy of crop type mapping and yield prediction. We used datasets from the main growing season of 2017 in Kenya (Bungoma, Busia and Siaya) coupled with county level yield data from US, Argentina and Brazil spanning from 2005 to 2016. Our models (CNN, SegNet, MaskRCNN, ResNet, UNet) were evaluated on both tasks i.e., classification of crop types and predicting yields.

KW - Agricultural

KW - Crop Types

KW - Deep Learning

KW - Machine Learning

KW - Yield

UR - https://www.scopus.com/pages/publications/105009601675

U2 - 10.1109/ICASSP49660.2025.10890344

DO - 10.1109/ICASSP49660.2025.10890344

M3 - Conference proceeding contribution

AN - SCOPUS:105009601675

BT - 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - India

Y2 - 6 April 2025 through 11 April 2025

ER -

Kashyap GS, Joshi H, Kulahara M, Dhakar R, Sajjanhar A, Gao J et al. Can AI see what we can’t? Leveraging deep learning and multi-temporal satellite data to revolutionize crop type mapping and yield prediction. In 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing: Conference Proceedings. India: Institute of Electrical and Electronics Engineers (IEEE). 2025 doi: 10.1109/ICASSP49660.2025.10890344