Lib.rs

A Rust library for calculating the similarity between two images.

List of metrics

• Sum of Absolute Differences (SAD)
• Mean of Absolute Differences (MAD)
• Sum of Squared Differences (SSD)
• Mean of Squared Error (MSE)
• Normalized Cross Correlation (NCC)
• Peak Signal to Noise Ratio (PSNR)
• Structural Similarity (SSIM)
• Multi-Scale Structural Similarity (MS-SSIM)
• Gradient Magnitude Similarity (GMSD)
• ...

Usage

[dependencies]
image = "0.23.4"

let image_a = image::open("image_a.png").unwrap();
let image_b = image::open("image_b.png").unwrap();
let difference = similarity_sad(&image_a, &image_b);

Examples of results

Image A	Image B
SAD¹:	0
MAD¹:	0.0
SSD¹:	0
MSE¹:	0.0

SAD¹:	13500834
MAD¹:	51.501595
SSD¹:	1179189332
MSE¹:	4498.2505

SAD¹:	16662881
MAD¹:	63.563847
SSD¹:	1680929909
MSE¹:	6412.239

SAD¹:	19903895
MAD¹:	75.92734
SSD¹:	1681425109
MSE¹:	6414.128

SAD¹:	46021705
MAD¹:	175.55887
SSD¹:	9046268163
MSE¹:	34508.773

SAD¹:	69718295
MAD¹:	265.9542
SSD¹:	20780685733
MSE¹:	79272.02

¹: The smaller the value, the more similar the images are.