Skip to main content

image processing

IMG PROCESSING Class​

⚠️ only support for format image jpg or png

IMAGE OCR - Tesseract​

IMG class adalah class khusus untuk optical character recognition, dibangun menggunakan teserract ocr php, tesseract OCR Enggine sendiri merupakan software Open source yang dikembangkan oleh google team dengan bahasa c++ . reference :

Note​

disini saya menggunakan windows sebagai OS, jadi saya termasuk windows users, untuk temen temen lain yang ingin menggunakan OCR wajib menginstall capture2text version 3.9 recommend menggunakan chocolately packages for windows. selengkapnya bisa read di sini

full doc about tesseract ocr for php

for install choco

install used administrative powershell

choco install capture2text --version 3.9

file test untuk file test bisa di check di halaman ini

Image Processing​

  1. Rescaling | done
  2. binarisation | done
  3. noise removal | done
  4. dilation / erotion | beta
  5. rotation / deskewing | not ready
  6. borders | not ready
  7. transparancy / alpha channel | not ready

usage​

basic usage image processing my librar and you need install imagick php and gdimage2

basic usage​

paramter optional for image processing :

  • rescaling default

    • 1200 DPI

  • Binarisation default :

    • threshold = 0.44
    • channel = 1

  • NoiseRemoval

    • quality = 5
include "vendor/autoload.php";

use Nagara\Src\Img\ImgMagic;
use Nagara\Src\Img\ImageProcessing;

// convert image to black and white step 1
$image = new ImgMagic;
$path =  "kk2.png";
$image->filter($path, "grayscale");

// image processing step 2 # example 1
$imgProcess = new ImageProcessing($path);
$imgProcess->Rescaling(1500)->Binarisation()->NoiseRemoval();
$imgProcess->showImage(); # return image


// image processing step 2 # example 2
$imgProcess = new ImageProcessing($path);
$imgProcess->Rescaling(1500); # only rescaling
$imgProcess->ShowImage(); # return image

get info​

basic get information

$path =  "kk2.png";
$imgProcess = new ImageProcessing($path);
$imgProcess->GetDimension(); # return only string with and height

MIX with Tesseract OCR​

include "vendor/autoload.php";
use Nagara\Src\Img\ImgParser;
use Nagara\Src\Img\ImgMagic;
use Nagara\Src\Img\ImageProcessing;

// convert image to black and white
$image = new ImgMagic;
$path =  "indonesia.png";
$image->filter($path, "grayscale");

// image processing
$imgProcess = new ImageProcessing($path);
$imgProcess->Rescaling(1500)->Binarisation()->NoiseRemoval();
// $imgProcess->showImage();

// ocr
$img = new ImgParser('C:\Program Files\Tesseract-OCR\tesseract.exe'); # example path exe'C:\Program Files\Tesseract-OCR\tesseract.exe'
$img->parseFile($path, "recognition");


https://mlocati.github.io/articles/php-windows-imagick.html https://windows.php.net/downloads/pecl/releases/imagick/ https://www.php.net/manual/en/imagick.requirements.php https://windows.php.net/downloads/pecl/

tesseract setup

install capture2text for windows

choco install capture2text --version 3.9

install tesseract lang

choco install tesseract --force `tesseract-lang

Maintenner

this code maintenace by me miyuki nagara