Javascript BCR Library 1.0.12

Authors: Gaspare Ferraro, Renzo Sala, Simone Ponte, Paolo Macco

BCR Library is a javascript library, using the OCR engine Tesseract.JS, that extracts name, company name, job, address, phone numbers, email and web address out of a business card picture.

The library is written in Javascript and can be used in any Javascript project (included in projects using frameworks for hybrid mobile applications, like Apache Cordova, Phonegap or Ionic).

The library can be used offline, no online dependencies are required.

Installation

Copy the content of the repository and reference bcr via script tag in your HTML project:

<script type="text/javascript" src="src/bcr.js"></script>

Sample

The sample application in the repository must be executed on a web server.

If you have python it's enough to run python -m http.server 8000 in the project folder.

If you use cordova, you can add the browser platform and run it (it works on other platforms like Android or iOS too).

Reference

Methods

Init methods

bcr.initialize(ocrEngine, crop, language, width, height, QRScanner, dynamicInclude);

Initialize the bcr reader. If ocrEngine is set to ocrEngines.TESSERACT, initialize the tesseract engine. If ocrEngine is set to ocrEngines.GOOGLEVISION, initialize bcr reader given the ocr from google mobile vision text recognition API (cordova-plugin-mobile-ocr).

Where:

• STRING ocrEngine the selected engine (see ocrEngines), default ocrEngines.TESSERACT.
• STRING crop: the crop strategy (see languages), default languages.GERMAN.
• STRING language: the language trained data (see cropStrategy), default cropStrategy.SMART.
• NUMBER width: max internal width, default 2160.
• NUMBER height: max internal height, default 1440.
• BOOLEAN QRScanner: check first for VCard QR Code in image, default true.
• BOOLEAN dynamicInclude: if the references are not included externally, default true.
• Return Promise about JS loading.

Recognize business card

bcr.recognize(base64image, displayResultCallback, displayProgressCallback, ocr);

Where:

• STRING base64image: base64 string of the image to analyze.
• FUNCTION displayResultCallback(result_data) function called when the analysis of the business card is completed.
• FUNCTION displayProgressCallback(progress_data) function called after each progress in the analysis.
• OBJECT ocr: object containing ocr results data from google mobile vision (optional, default ``).

Getter methods

bcr.cropStrategy()

• Return the strategy label internally set.

bcr.maxWidth()

• Return the value of the max width used internally to normalize the resolution.

bcr.maxHeight()

• Return the value of the max height used internally to normalize the resolution.

bcr.language()

• Return the value of the language trained data.

bcr.tesseract()

• Return the initialized tesseract worker.

bcr.ocr()

• Return the ocr passed.

bcr.ocrEngine()

• Return the ocr engine selected.

bcr.qrScanner()

-if VCard QRScanner read is enabled.

Object

result_data

JSON object in the format:

{
  "Company": "STRING",
  "Email": "STRING",
  "Address": {
      "StreetAddress": "STRING",
      "ZipCode": "STRING",
      "Country": "STRING",
      "Text": "STRING",
      "City": "STRING"
  },
  "Web": "STRING",
  "Phone": "STRING",
  "Text": "STRING",
  "Fax": "STRING",
  "Job": "STRING",
  "Mobile": "STRING",
  "Name": {
      "Text": "STRING",
      "Surname": "STRING",
      "Name": {
          "FirstName": "STRING",
          "Text": "STRING",
          "MiddleName": "STRING",
          "ExtraName": "STRING"
      }
  }
}

progress_data

JSON object in the format:

{
  "section": "STRING",
  "progress": {
    "status": "STRING",
    "progress": "FLOAT"
  }
}

ENUM

languages

• languages.DANISH: Danish language
• languages.GERMAN: German language
• languages.ENGLISH: English language
• languages.FRENCH: French language
• languages.ITALIAN: Italian language
• languages.SPANISH: Spanish language
• languages.SWEDISH: Swedish language

cropStrategy

• cropStrategy.SMART: clean the image

ocrEngines

• ocrEngines.TESSERACT: use the tesseract internal engine
• ocrEngines.GOOGLEVISION: use Google Mobile Vision external engine

JS Libraries used

• Tesseract.JS - 1.0.19
  Tesseract.js wraps an emscripten port of the Tesseract OCR Engine.

Required Cordova Plugins (in case of cordova project)

• cordova-plugin-ionic-webview - 4.0.0
  A Web View plugin for Cordova, focused on providing the highest performance experience for Ionic apps (but can be used with any Cordova app).

Contribution

The current status of the library is alpha. Looking forward for your contribution to make the first release.