This dataset includes input chips of 224x224x18, where 224 is the height and width and 18 is combined with 6 bands of 3 time-steps. The bands are:
Blue
Green
Red
Narrow NIR
SWIR 1
SWIR 2
Labels are from CDL(Crop Data Layer) and classified into 13 classes.
The Prithvi-100m model was initially pretrained using a sequence length of 3 timesteps. For this task, we leverage the capacity for multi-temporal data input, which has been integrated from the foundational pretrained model. This adaptation allows us to achieve more generalized finetuning outcomes.
Configuration used for finetuning is available through config.
Results
The experiment by running the mmseg stack for 80 epochs using the above config led to the following result:
Classes
IoU
Acc
Natural Vegetation
0.4038
46.89%
Forest
0.4747
66.38%
Corn
0.5491
65.47%
Soybeans
0.5297
67.46%
Wetlands
0.402
58.91%
Developed/Barren
0.3611
56.49%
Open Water
0.6804
90.37%
Winter Wheat
0.4967
67.16%
Alfalfa
0.3084
66.75%
Fallow/Idle Cropland
0.3493
59.23%
Cotton
0.3237
66.94%
Sorghum
0.3283
73.56%
Other
0.3427
47.12%
aAcc
mIoU
mAcc
60.64%
0.4269
64.06%
It is important to acknowledge that the CDL (Crop Data Layer) labels employed in this process are known to contain noise and are not entirely precise, thereby influencing the model's performance. Fine-tuning the model with more accurate labels is expected to further enhance its overall effectiveness, leading to improved results.
Baseline
The baseline model along with its results can be accessed here.
Inference
The github repo includes an inference script that allows to run the hls-cdl crop classification model for inference on HLS images. These input have to be geotiff format, including 18 bands for 3 time-step, and each time-step includes the channels described above (Blue, Green, Red, Narrow NIR, SWIR, SWIR 2) in order. There is also a demo that leverages the same code here.
Feedback
Your feedback is invaluable to us. If you have any feedback about the model, please feel free to share it with us. You can do this by submitting issues on our open-source repository, hls-foundation-os, on GitHub.
Citation
If this model helped your research, please cite HLS Multi Temporal Crop Classification Model in your publications. Here is an example BibTeX entry:
@misc{hls-multi-temporal-crop-classification-model,
author = {Li, Hanxi (Steve) and Khallaghi, Sam and Cecil, Michael and Kordi, Fatemeh and Fraccaro, Paolo and Alemohammad, Hamed and Ramachandran, Rahul},
doi = { 10.57967/hf/0954 },
month = aug,
title = {{HLS Multi Temporal Crop Classification Model}},
url = {https://huggingface.co/ibm-nasa-geospatial/Prithvi-100M-multi-temporal-crop-classification},
year = {2023}
}
