OCR API Setup & Deploy with PHP

OCR.Space is a free OCR API which allow parsing images and PDF documents to extract the text in JSON format. Since, it's an API, it could be apply the OCR functions by using GET and POST parameter with any languages in development.

Let's get started by signup an FREE tier OCR.Space API key, this acts as a token for service authentication and initialisation, if no access key is attached in a GET or POST parameter, the return results are usually NULL. Kindly, fill in the signup form at here. Once the request is successful, the API key will be sent to your email inbox.

To use the OCR API, the endpoint for POST had been provided as below and support HTTPS and HTTP protocols.

https://api.ocr.space/parse/image To apply the very basic image parsing via URL and reveal the OCR results, the image below is attached in this demonstration.

(IMAGE URL: COPY HERE )

Simply append the parameter of "apikey" and "url" at the end of the API address:

https://api.ocr.space/parse/image?apikey=YOUR-API-KEY&url=IMAGE-URL

The result of OCR parsed will be as below:

As you can see, the parsing will results a JSON output and from what we can observed, we could extract or export the value we required where param at "ParsedText".

To decode the JSON or extract the value out of this structure, we will suggest a PHP scripting as below:

<?php
$gateway = 'https://api.ocr.space/parse/imageurl';
$key = 'helloworld'; //input your API KEY, this is test key
$img = 'https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiZN1vlbHxOM6J01rScGR_4J-s3MwLbseE-qEiH8KxarHe0M06RjugEknQKpyKMJBDzqPnT77D1sGi20SAbOl7MJLN2OeeSCJh5VcgEXKSC7iAHMxyzpuJY01Ko7Pz4yHSeCtBfwNrPv80/s1600/TEXT01.png';

$url = $gateway."?"."apikey=".$key."&url=".$img; //establish the url
$data = file_get_contents($url); // fetch content to variable array
$txt = json_decode($data, true); // decode the JSON feed

echo "<h1> OCR.SPACE API DEMONSTRATION </h1>";
echo "<h2> LOADED IMAGE </h2>"; //show fetch image

echo "<img border=\"0\" data-original-height=\"739\" data-original-width=\"962\" height=\"306\" width=\"400\" src=\"".$img."\"/>";
echo "<h2> OUTPUT </h2>"; //display image text
echo $txt["ParsedResults"][0]["ParsedText"]; //output the ocr text
?>

Simply deploy the codes and edit the highlighted source with any web server, I'm using APACHE and save it to FILENAME.php 

The result will show as below:

The above demonstration is by default OCR options which the interpretation is in ENGLISH OCR. If you wish to perform an OCR applied with different language, simply modify the parameter of API parsing as below:

Simply append the parameter of "apikey" + "url" + "language" at the end of the API address:

https://api.ocr.space/parse/image?apikey=YOUR-API-KEY&url=IMAGE-URL&language=LANG-PREFIX

Let's try with a image with Chinese characters.

(IMAGE URL: COPY HERE)

Edit the PHP coding and append with the parameter.

<?php

$gateway = 'https://api.ocr.space/parse/imageurl';
$key = 'helloworld'; //input your API KEY, this is test key
$img='https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEj5trq6LDiAKgR_FavIeEI5y4iPlxdvYDKW58iZFowgm8JHmSkTsPsqOme4aqsIkxtLuBVYNCMxR_kxjqRmENPiuQusQ5mm-106ezXTr7AzMmOCdPP4oc-VbBuIlxQ4ZVs4WGMjDVsotZ4/s400/cccc.png';

$lang = 'chs';

$url = $gateway."?"."apikey=".$key."&url=".$img."&language=".$lang; //establish the url

?>

The results are perfect!

Note that the OCR language parameter's parsing option as below, simply change the $lang parameter with any of the below validated with the text appeared in the parsed image.

Arabic=ara
Bulgarian=bul
Chinese(Simplified)=chs
Chinese(Traditional)=cht
Croatian = hrv
Czech = cze
Danish = dan
Dutch = dut
English = eng
Finnish = fin
French = fre
German = ger
Greek = gre
Hungarian = hun
Korean = kor
Italian = ita
Japanese = jpn
Norwegian = nor
Polish = pol
Portuguese = por
Russian = rus
Slovenian = slv
Spanish = spa
Swedish = swe
Turkish = tur

The application of OCR is widely applied in automated processing invalided images to text or to obtain valid values from pictures. The OCR is working in AI, integrated environment or to be a part of a processing node in a system to perform text extraction from provided inputs. Happy Explore & Have a nice day, do leave us any comments and we will reply you shortly.

Feel free to share this article around and do like & follow at our Facebook page. We would like to talk about IoT applications in the next post, stay tune.