Cleaned up and improved OCR operation

This commit is contained in:
n1474335 2019-09-13 14:34:08 +01:00
parent a8ad10757c
commit 7eabaf0de6
3 changed files with 48 additions and 28 deletions

View file

@ -1,4 +1,5 @@
/**
* @author n1474335 [n1474335@gmail.com]
* @author mshwed [m@ttshwed.com]
* @copyright Crown Copyright 2019
* @license Apache-2.0
@ -7,30 +8,36 @@
import Operation from "../Operation.mjs";
import OperationError from "../errors/OperationError.mjs";
import { isImage } from "../lib/FileType.mjs";
import { toBase64 } from "../lib/Base64.mjs";
import { isWorkerEnvironment } from "../Utils.mjs";
import jimp from "jimp";
import Tesseract from "tesseract.js";
const { TesseractWorker } = Tesseract;
/**
* OCR operation
* Optical Character Recognition operation
*/
class OCR extends Operation {
class OpticalCharacterRecognition extends Operation {
/**
* OCR constructor
* OpticalCharacterRecognition constructor
*/
constructor() {
super();
this.name = "OCR";
this.module = "Default";
this.description = "Optical character recognition or optical character reader (OCR) is the mechanical or electronic conversion of images of typed, handwritten or printed text into machine-encoded text.";
this.infoURL = "https://en.wikipedia.org/wiki/Optical_character_recognition";
this.name = "Optical Character Recognition";
this.module = "Image";
this.description = "Optical character recognition or optical character reader (OCR) is the mechanical or electronic conversion of images of typed, handwritten or printed text into machine-encoded text.<br><br>Supported image formats: png, jpg, bmp, pbm.";
this.infoURL = "https://wikipedia.org/wiki/Optical_character_recognition";
this.inputType = "ArrayBuffer";
this.outputType = "string";
this.args = [];
this.args = [
{
name: "Show confidence",
type: "boolean",
value: true
}
];
}
/**
@ -39,34 +46,32 @@ class OCR extends Operation {
* @returns {string}
*/
async run(input, args) {
if (!isImage(input)) {
const [showConfidence] = args;
const type = isImage(input);
if (!type) {
throw new OperationError("Invalid File Type");
}
try {
if (isWorkerEnvironment())
self.sendStatusMessage("Performing OCR on image...");
let image;
try {
image = await jimp.read(input);
image = await image.getBase64Async(jimp.AUTO);
} catch (err) {
throw new OperationError(`Error loading image. (${err})`);
}
const image = `data:${type};base64,${toBase64(input)}`;
const worker = new TesseractWorker();
const result = await worker.recognize(image)
.progress(progress => {
if (isWorkerEnvironment()) self.sendStatusMessage(`${progress.status} - ${(parseFloat(progress.progress)*100).toFixed(2)}%`);
if (isWorkerEnvironment()) {
self.sendStatusMessage(`Status: ${progress.status} - ${(parseFloat(progress.progress)*100).toFixed(2)}%`);
}
});
return result.text;
if (showConfidence) {
return `Confidence: ${result.confidence}%\n\n${result.text}`;
} else {
return result.text;
}
} catch (err) {
throw new OperationError(`Error performing OCR on image. (${err})`);
}
}
}
export default OCR;
export default OpticalCharacterRecognition;