Sri Lankan NIC toolkit

JavaScript helpers for working with Sri Lankan National Identity Cards (NIC). The package provides:

• NIC parser with gender, DOB, sequence and age extraction for both legacy and 2016+ formats.
• OCR-driven extraction helpers backed by Tesseract.js with Sinhala/Tamil/English language packs.
• PDF417 barcode decoding with rotation fallbacks and metadata comparison utilities.
• A high-level verifyNicFromImages orchestrator for combining front/back/selfie checks.

Installation

npm install @osca16/slnic

Usage

import { parseSriLankaNIC, ocrNicFromImage, decodeBackBarcode, verifyNicFromImages } from "@osca16/slnic";

const parsed = parseSriLankaNIC("####");

const ocr = await ocrNicFromImage(frontCanvas);
const barcode = await decodeBackBarcode(backCanvas);

const verification = await verifyNicFromImages({
	frontImage: frontCanvas,
	backImage: backCanvas,
	selfieImage: selfieCanvas,
	languages: ["en-LK", "si-LK"]
});

React + Vite quick start

1. Install the dependencies:

   npm install @osca16/slnic tesseract.js @techstark/opencv-js

2. Teach Vite to keep the OpenCV WebAssembly asset and avoid pre-bundling the large module (optional but reduces build warnings):

   // vite.config.js
   import { defineConfig } from "vite";
   import react from "@vitejs/plugin-react";
   
   export default defineConfig({
       plugins: [react()],
       assetsInclude: ["**/*.wasm"],
       optimizeDeps: {
      	 exclude: ["@osca16/slnic", "@techstark/opencv-js"]
       }
   });

3. Use the helpers from a React component. The verifyNicFromImages helper takes care of OCR, barcode scan, and OpenCV-based selfie comparison:

   import { useState } from "react";
   import { verifyNicFromImages } from "@osca16/slnic";
   
   export function NicVerifier() {
       const [result, setResult] = useState();
   
       async function handleSubmit(event) {
      	 event.preventDefault();
      	 const form = event.currentTarget;
      	 const front = form.front.files[0];
      	 const back = form.back.files[0];
      	 const selfie = form.selfie.files[0];
   
      	 if (!front) return;
   
      	 const verification = await verifyNicFromImages({
      		 frontImage: front,
      		 backImage: back,
      		 selfieImage: selfie,
      		 ocrOptions: {
      			 workerPath: new URL("tesseract.worker.min.js", import.meta.url).href,
      			 langPath: "/tesseract-langs/"
      		 },
      		 faceOptions: {
      			 targetSize: 192,
      			 blurThreshold: 14
      		 }
      	 });
   
      	 setResult(verification);
       }
   
       return (
      	 <form onSubmit={handleSubmit}>
      		 <input name="front" type="file" accept="image/*" required />
      		 <input name="back" type="file" accept="image/*" />
      		 <input name="selfie" type="file" accept="image/*" />
      		 <button type="submit">Verify</button>
      		 <pre>{result ? JSON.stringify(result, null, 2) : ""}</pre>
      	 </form>
       );
   }

   Place tesseract language data (e.g. eng.traineddata, sin.traineddata, tam.traineddata) under public/tesseract-langs/ so the worker can load them when offline.

Environment notes

• The library targets modern browsers. For Node.js usage, provide HTMLCanvasElement, ImageData, or run under a DOM-enabled environment (e.g. canvas + jsdom).
• Ensure the Tesseract language data for eng, sin, and tam is available; tesseract.js downloads files on first use.
• When working with large images, pre-scale to improve recognition speed.
• You can control OCR worker/core/lang paths by passing an options object as the second argument to ocrNicFromImage or via verifyNicFromImages({ ocrOptions: {...} }).
• compareFaces now uses a lightweight OpenCV-based blur + similarity heuristic. Pass { faceOptions: { blurThreshold, targetSize } } to tune thresholds for your camera setup.
• When bundling, make sure opencv_js.wasm from @techstark/opencv-js is served (copy it to public/ or expose via Module.locateFile) so that face comparison can initialise correctly.

Feedback

If you encounter issues or have feature requests, please open an issue or share improvements.