Initial commit

Author: wiazur

nodejs/Vision/RecognizeText.js

@@ -0,0 +1,111 @@
+'use strict';
+
+const fetch = require('node-fetch');
+const fs = require('fs');
+const util = require('util');
+const delay = require('delay');
+
+const readFile = util.promisify(fs.readFile);
+
+// This sample uses the Recognize Text API call from Azure Computer Vision
+// This call can recognize printed or handwritten text, just change the image and mode to specify.
+// Computer Vision API - v2.0:
+// https://westus.dev.cognitive.microsoft.com/docs/services/5adf991815e1060e6355ad44/operations/587f2c6a154055056008f200
+
+let key = process.env['COMPUTER_VISION_SUBSCRIPTION_KEY'];
+let baseUrl= process.env['COMPUTER_VISION_ENDPOINT']
+let mode = 'Printed'; // Printed or Handwritten depending on what we want to detect
+
+// Use a local file downloaded from here:
+// https://github.com/Azure-Samples/cognitive-services-sample-data-files/tree/master/ComputerVision/Images
+// Or add your own image to your project folder
+let file = 'printed_text.jpg' // local
+
+// Get the binary data of an image
+async function readImageByPath(imagePath) {
+    const imageBinary = await readFile(imagePath);
+    const fileData = imageBinary.toString('hex');
+    const result = [];
+    for (let i = 0; i < fileData.length; i += 2) {
+        result.push(parseInt(`${fileData[i]}${fileData[i + 1]}`, 16))
+    }
+    return Buffer.from(result);
+}
+
+async function checkAnalyzeImageJob(operationLocationUrl) {
+    return new Promise(async (resolve, reject) => {
+        try {
+            const response = await fetch(operationLocationUrl, {
+                method: 'GET',
+                headers: {
+                "Content-Type": "application/json",
+                "Ocp-Apim-Subscription-Key": key
+                }
+            })
+
+            const json = await response.json();
+            return resolve(json);
+        } catch (e) {
+            return reject(e);
+        }
+    })
+}
+
+async function submitAnalyzeImageJob(imageBinary) {
+    return new Promise(async (resolve, reject) => {
+        const url = baseUrl + "/vision/v2.0/recognizeText" + "?" + `mode=${mode}`;
+
+        try {
+            const response = await fetch(url, {
+                method: 'POST',
+                headers: {
+                    "Content-Type": "application/octet-stream",
+                    "Ocp-Apim-Subscription-Key": key
+                },
+                body: imageBinary
+            })
+
+            if (response.headers.has('operation-location')) {
+                return resolve(response.headers.get('operation-location'));
+            } else {
+                return reject(response.headers);
+            }
+        } catch (e) { 
+            return reject(e);
+        }
+    });
+}
+  
+async function analyzeImage(imageBinary) {
+    return new Promise(async (resolve, reject) => {
+        const jobLocation = await submitAnalyzeImageJob(imageBinary);
+        let isDone = false;
+
+        while (!isDone) {
+            const result = await checkAnalyzeImageJob(jobLocation);
+            if (result.status == 'NotStarted') {
+                console.log('Recognition is not started yet');
+                isDone = false;
+                await delay(1000);
+            } else if (result.status == 'Succeeded') {
+                return resolve(result.recognitionResult);
+            } else if (result.status == 'Running') {
+                console.log('Recognition is running, awaiting result');
+                isDone = false;
+                await delay(1000);
+            } else {
+                return reject('Unknown status: ' + JSON.stringify(result));
+            }
+        }
+
+        return reject('SHOULD NOT REACH');
+    });
+}
+
+const start = async () => {
+    const imageData = await readImageByPath(file);
+    const result = await analyzeImage(imageData);
+    console.log(result);
+};
+
+start();