Giter VIP home page Giter VIP logo

imagehash's Introduction

imagehash

Build Status GoDoc Coverage

Golang implementation of image hashing algorithms.

Install:

go get -u github.com/devedge/imagehash

Usage

There are currently two image hashing algorithms implemented:

  • dhash - difference/gradient hash
  • ahash - average hash

To hash an image, it must be opened using OpenImg, a wrapper around imaging's image decoding function.

src,err := imagehash.OpenImg("./testdata/lena_512.png")

dhash

This is an implementation of the dhash algorithm found here (archived link), and also implemented in Python here.

It generates a unique signature for an image based on the gradient difference between pixels.

// The hashes are returned as byte arrays
//
// Calculate both horizontal & vertical gradients, then return them
// concatenated together as <horizontal><vertical>
hash,err  := imagehash.Dhash(src, hashLen)

// Calculate only the horizontal gradient difference
hashH,err := imagehash.DhashHorizontal(src, hashLen)

// Calculate only the vertical gradient difference
hashV,err := imagehash.DhashVertical(src, hashLen)

Using dhash:

package main

import (
  "fmt"
  "encoding/hex"  // To transform the byte array to hex
  "github.com/devedge/imagehash"
)

func main() {
  src,_ := imagehash.OpenImg("./testdata/lena_512.png")

  // The length of a downscaled side. It must be > 8, and
  // (hashLen * hashLen) must be a multiple of 8
  hashLen := 8
  // A value of 8 will return 64 bits, or 8 bytes / 16 hex characters
  // (64 bits = 8 bits length * 8 bits width)

  hash,_ := imagehash.Dhash(src, hashLen)
  hashH,_ := imagehash.DhashHorizontal(src, hashLen)
  hashV,_ := imagehash.DhashVertical(src, hashLen)

  fmt.Println("dhash:           ", hex.EncodeToString(hash))
  fmt.Println("Horizontal dhash:", hex.EncodeToString(hashH))
  fmt.Println("Vertical dhash:  ", hex.EncodeToString(hashV))
}

Implementation:

First, the image is grayscaled:

grayscale

To calculate the horizontal gradient difference, the image is resized down, using the hashLen variable.

In this example, hashLen = 8, so the image is scaled down to 9x8px. Then, if pixel[x] < pixel[x+1], a 1 is appended to a byte array; otherwise, a 0. This results in 8 bits of data per row, for 8 columns, or 64 bits total:

dhashprocess

This array of 1s and 0s is then flattened, and returned as a byte array:
0111011001110000011110010101101100110011000100110101101000111000

Which can also be represented in hex as 7670795b33135a38 using hex.EncodeToString(result)

Conversely, to obtain a vertical diff, the image would be scaled down to 8x9px, and the diff matrix would be the result of pixel[y] < pixel[y+1].

ahash

This algorithm returns a hash based on the average pixel value.

As with dhash, it also grayscales and resizes the image down, using the 'hashLen' value. Then, it finds the average value of the resultant pixels. Finally, it iterates over the pixels, and if one is greater than the average, a 1 is appended to the returned result; a 0 otherwise.

// The hash is returned as a byte array
hash,err := imagehash.Ahash(src, hashLen)

Examples

The Hamming distance between two byte arrays can be determined using a package like hamming:

package main

import (
  "fmt"
  "encoding/hex"
  "github.com/devedge/imagehash"
  "github.com/steakknife/hamming"
)

func main() {
  src512,_ := imagehash.OpenImg("./testdata/lena_512.png")
  src256,_ := imagehash.OpenImg("./testdata/lena_256.png")
  srcInv,_ := imagehash.OpenImg("./testdata/lena_inverted_512.png")

  hash512,_ := imagehash.Dhash(src512, 8)
  hash256,_ := imagehash.Dhash(src256, 8)
  hashInv,_ := imagehash.Dhash(srcInv, 8)

  // Hamming distance of 0, since the images are simply different sizes
  fmt.Println("'lena_512.png' dhash:", hex.EncodeToString(hash512))
  fmt.Println("'lena_256.png' dhash:", hex.EncodeToString(hash256))
  fmt.Println("The Hamming distance between these:", hamming.Bytes(hash512, hash256))
  fmt.Println()

  // Completely different dhash, since an inverted image has a completely
  // different gradient colorscheme
  fmt.Println("'lena_512.png' dhash:         ", hex.EncodeToString(hash512))
  fmt.Println("'lena_inverted_512.png' dhash:", hex.EncodeToString(hashInv))
  fmt.Println("The Hamming distance between these:", hamming.Bytes(hash512, hashInv))
}

Dependencies:

  • imaging - Simple Go image processing package

imagehash's People

Contributors

alexandrestein avatar devedge avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.