psnr-hvsm 0.1.0

Accelerated implementations of the PSNR-HVS, PSNR-HVS-M, PSNR-HA and PSNR-HMA image metrics

psnr_hvsm

Accelerated Python package for computing several image metrics based on human perception.

This is an implementation of the PSNR-HVS, PSNR-HVS-M, PSNR-HA and PSNR-HMA metrics developed by Nikolay Ponomarenko.

The values produced by this library have been cross-checked against the results within the TID2013 dataset. (See the folder tid2013_results.) The only difference is that this library follows the common convention that PSNR for identical signals equals 100.0.

A miniscule discrepancy for PSNR-HMA (<0.01dB on average) is under investigation.

Bibliography

• Egiazarian, Karen, et al. "New full-reference quality metrics based on HVS." Proceedings of the Second International Workshop on Video Processing and Quality Metrics. Vol. 4. 2006.
• Ponomarenko, Nikolay, et al. "On between-coefficient contrast masking of DCT basis functions." Proceedings of the third international workshop on video processing and quality metrics. Vol. 4. 2007.
• Ponomarenko, Nikolay, et al. "Modified image visual quality metrics for contrast change and mean shift accounting." 2011 11th International Conference The Experience of Designing and Application of CAD Systems in Microelectronics (CADSM). IEEE, 2011.

Installation

psnr_hvsm supports Python 3.6-3.9. Packages are distributed on PyPi. Be sure to have an up-to-date pip to be able to install the correct packages on Linux:

python -m pip install --upgrade pip

pip install psnr_hvsm

Usage

Command line

Command line support is an extra that pulls opencv-python-headless:

pip install psnr_hvsm[command_line]

python -m psnr_hvsm original.png distorted.png

As a library

The function psnr_hvs_hvsm accepts images as single-channel floating-point NumPy arrays. The images need to be normalised, i.e. the values need to be in the range [0,1]. This can be achieved by converting the image to float and dividing by the maximum value given the bit depth. For 8 bits per component this is 255.

The images must be padded to a multiple of 8 in each dimension.

from imageio import imread
from psnr_hvsm import psnr_hvs_hvsm

image1 = imread('tests/baboon.png').astype(float) / 255
image2 = imread('tests/baboon_msk.png').astype(float) / 255

psnr_hvs, psnr_hvsm = psnr_hvs_hvsm(image1, image2)

print(psnr_hvs, psnr_hvsm)

34.427054505764424 51.64722121999962

If you need to measure PSNR-HVS and PSNR-HVS-M on an RGB image, you need to convert it to an YUV colorspace and pass in only the luma component.

Computing metrics for the TID2013 dataset

If you have a copy of the TID2013 dataset, you can re-verify the metrics for yourself:

python -m psnr_hvsm.tid2013_metrics D:\tid2013\ .\tid2013_results\

Other exported functions

• psnr_hvs - returns only the PSNR-HVS value
• psnr_hvsm - returns only the PSNR-HVS-M value
• hvs_mse_tile - compute the weighted MSE of two 8x8 tiles
• hvsm_mse_tile - compute the weighted MSE with masking correction of two 8x8 tiles
• hvs_mse - compute HVS scores on all 8x8 tiles in the images, returns an array of numbers
• hvsm_mse - compute HVS-M scores on all 8x8 tiles in the images, returns an array of numbers

Building

Dependencies

psnr_hvsm has several dependencies:

• FFTW3 >= 3.3.9
• pybind11
• xtensor
• xtensor-python

All of the above can be automatically resolved by running deps.ps1, which is a cross-platform PowerShell script (i.e. it can also be run under Linux if you have PowerShell installed).

Development mode

To install in development mode:

pip install --upgrade -r requirements.txt

Creating Python wheel

pip install --upgrade -r requirements-build.txt
python setup.py bdist_wheel

Running tests on different versions of Python using tox

pip install --upgrade -r requirements-tox.txt
tox --parallel auto