Registry / testing / image-similarity-measures

image-similarity-measures

library0.3.6pypypi✓ verified 83d ago

A Python library providing evaluation metrics to assess the similarity between two images. Version 0.3.6, released infrequently. Supports Python >=3.8.

testing

pip install image-similarity-measures

INSTALL

IMPORT

SIG · IMAGE-SIMILARITY-M

image-similarity-measures

testingpythonv0.3.6

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

image_similarity_measures

✓ import image_similarity_measures

Compute PSNR and SSIM between two images.

import cv2 from image_similarity_measures import quality_metrics img1 = cv2.imread('image1.png') img2 = cv2.imread('image2.png') # Ensure images are same size img1 = cv2.resize(img1, (256, 256)) img2 = cv2.resize(img2, (256, 256)) # Compute metrics psnr = quality_metrics.psnr(img1, img2) ssim = quality_metrics.ssim(img1, img2) print(f'PSNR: {psnr}, SSIM: {ssim}')

Debug

Known issues

gotchaAll functions expect images in BGR format (OpenCV default). Using RGB images may produce incorrect results.

fix

Convert RGB images to BGR before passing: img_bgr = cv2.cvtColor(img_rgb, cv2.COLOR_RGB2BGR)

affects: all

gotchaImages must have the same size and number of channels. Mismatched dimensions cause silent incorrect results or errors.

fix

Resize or crop images to match dimensions before calling metrics.

affects: all

deprecatedThe 'evaluate' function from early versions is deprecated. Use 'quality_metrics' module directly.

fix

Replace 'from image_similarity_measures import evaluate' with 'from image_similarity_measures import quality_metrics' and call metrics explicitly.

affects: >=0.3.0

Errors

Common errors & fixes

AttributeError: module 'image_similarity_measures' has no attribute 'quality_metrics'

Old version (<0.3.0) or wrong import path.

fix

Upgrade library: pip install --upgrade image-similarity-measures ; then use 'from image_similarity_measures import quality_metrics'

TypeError: expected image data type is uint8, got float32

Images loaded with incorrect dtype (e.g., normalized to 0-1 float).

fix

Convert images to uint8: img_uint8 = (img_float * 255).astype(np.uint8) or load with cv2.imread for uint8.

ValueError: images must have same shape

Input images have different dimensions or number of channels.

fix

Ensure images are resized to same shape: img1 = cv2.resize(img1, (width, height)); img2 = cv2.resize(img2, (width, height))

Upgrade

Version history

0.3.6latest on PyPI · released May 4, 2023

Audit

Dependencies

No dependency data recorded yet.

Agent activity

6 hits · last 30 days

node

Resources

No resource links recorded.