imagehash

ImageHash is a Zig package for generating robust image fingerprints using four popular algorithms: average (ahash), difference (dhash), perceptual (phash), and wavelet (whash) hashing.

✨ Features

• Multiple Hash Algorithms:

  • ahash (Average Hash)
  • dhash (Difference Hash)
  • phash (Perceptual Hash)
  • whash (Wavelet Hash)

🚀 Getting Started

Fetch via zig fetch

You can use the built‑in Zig fetcher to download and pin a tarball:

zig fetch --save git+https://github.com/galactixx/imagehash#v0.2.0

This adds an imagehash entry under .dependencies in your build.zig.zon.

Then in your build.zig:

const imagehash_mod = b.dependency("imagehash", .{
    .target = target,
    .optimize = optimize,
}).module("imagehash");

// add to library
lib_mod.addImport("imagehash", imagehash_mod);

// add to executable
exe.root_module.addImport("imagehash", imagehash_mod);

This lets you const ih = @import("imagehash"); in your Zig code.

📚 Usage

const std = @import("std");
const ih = @import("imagehash");

pub fn main() !void {
    const file = "testdata/checkerboard.png";

    // Compute all four hashes
    const ahash = try ih.averageHash(file);
    const dhash = try ih.differenceHash(file);
    const phash = try ih.perceptualHash(file);
    const whash = try ih.waveletHash(file);

    // Print hex digests
    var buf: [16]u8 = undefined;
    std.debug.print("ahash: {s}\n", .{ahash.hexDigest(buf[0..])});
    std.debug.print("phash: {s}\n", .{phash.hexDigest(buf[0..])});

    // Compare two hashes
    const dist = ahash.distance(dhash);
    std.debug.print("Hamming distance: {}\n", .{dist});
}

🔍 API

pub fn averageHash(filename: []const u8) Error!ImageHash

Computes an 8×8 average-based hash (ahash).

pub fn differenceHash(filename: []const u8) Error!ImageHash

Computes a 9×8 horizontal difference-based hash (dhash). s

pub fn perceptualHash(filename: []const u8) Error!ImageHash

Computes a 32×32 perceptual (DCT-based) hash and reduces to an 8×8 block (phash).

pub fn waveletHash(filename: []const u8) Error!ImageHash

Computes a 64×64 wavelet-based hash with 4-level decomposition (whash).

pub fn (self: ImageHash) distance(other: ImageHash) u64

Returns the Hamming distance between two hashes.

pub fn (self: ImageHash) toJSON(alloc: *std.mem.Allocator) ![]u8

Serialize an ImageHash to a JSON byte slice.

pub fn fromJSON(json: []const u8, alloc: std.mem.Allocator) ParseError!ImageHash

Parse an ImageHash from JSON.

🤝 License

This project is licensed under the MIT License. See the LICENSE file for details.

📞 Contact

Have questions or need help? Open an issue on the GitHub repository.