Python package and commandline tool to evaluate the similarity between two images with eight evaluation metrics:
- Root mean square error (RMSE)
- Peak signal-to-noise ratio (PSNR)
- Structural Similarity Index (SSIM)
- Feature-based similarity index (FSIM)
- Information theoretic-based Statistic Similarity Measure (ISSM)
- Signal to reconstruction error ratio (SRE)
- Spectral angle mapper (SAM)
- Universal image quality index (UIQ)
Supports Python >=3.8.
pip install image-similarity-measures
Optional: For faster evaluation of the FSIM metric, the pyfftw
package is required, install via:
pip install image-similarity-measures[speedups]
Optional: For reading TIFF images with rasterio
instead of OpenCV
, install:
pip install image-similarity-measures[rasterio]
To evaluate the similarity beteween two images, run on the commandline:
image-similarity-measures --org_img_path=a.tif --pred_img_path=b.tif
Note that images that are used for evaluation should be channel last. The results are printed in machine-readable JSON, so you can redirect the output of the command into a file.
--org_img_path FILE Path to original input image
--pred_img_path FILE Path to predicted image
--metric METRIC select an evaluation metric (fsim, issm, psnr, rmse,
sam, sre, ssim, uiq, all) (can be repeated)
from image_similarity_measures.evaluate import evaluation
evaluation(org_img_path="example/lafayette_org.tif",
pred_img_path="example/lafayette_pred.tif",
metrics=["rmse", "psnr"])
from image_similarity_measures.quality_metrics import rmse
rmse(org_img=np.random.rand(3,2,1), pred_img=np.random.rand(3,2,1))
Contributions are welcome! Please see README-dev.md for instructions.
Please use the following for citation purposes of this codebase:
Müller, M. U., Ekhtiari, N., Almeida, R. M., and Rieke, C.: SUPER-RESOLUTION OF MULTISPECTRAL SATELLITE IMAGES USING CONVOLUTIONAL NEURAL NETWORKS, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-1-2020, 33–40, https://doi.org/10.5194/isprs-annals-V-1-2020-33-2020, 2020.