I was trying to get the mlkit to work and get the text from an image, but it couldnot read the file path it showed cannot read the property of null. I tried creating an array, results, to store the file path, but it did not work… this was done in react-native expo
import React, { useState } from 'react';
import MlkitOcr from 'react-native-mlkit-ocr';
import { View, Text, Image, Button, TouchableOpacity } from 'react-native';
import styles from './Styles';
import * as ImagePicker from 'expo-image-picker';
function App() {
// The path of the picked image
const [pickedImagePath, setPickedImagePath] = useState('');
const results = []
// This function is triggered when the "Select an image" button pressed
const showImagePicker = async () => {
// Ask the user for the permission to access the media library
const permissionResult = await ImagePicker.requestMediaLibraryPermissionsAsync();
if (permissionResult.granted === false) {
alert("No Gallery Access!");
return;
}
//Allow editing after taking the image from gallery
const result = await ImagePicker.launchImageLibraryAsync({
mediaTypes: ImagePicker.MediaTypeOptions.Images,
allowsEditing: true,
quality: 1,
});
// Explore the result
// console.log(result);
if (!result.canceled) {
setPickedImagePath(result.assets[0].uri);
const getPath = result.assets[0].uri;
results.push(getPath)
console.log(results)
}
}
// This function is triggered when the "Open camera" button pressed
const openCamera = async () => {
// Ask the user for the permission to access the camera
const permissionResult = await ImagePicker.requestCameraPermissionsAsync();
if (permissionResult.granted === false) {
alert("No Camera Access!");
return;
}
//Allow editing after taking the picture
const result = await ImagePicker.launchCameraAsync({
mediaTypes: ImagePicker.MediaTypeOptions.Images,
allowsEditing: true,
quality: 1,
});
// Explore the result
// console.log(result);
if (!result.canceled) {
setPickedImagePath(result.assets[0].uri);
// console.log(result.assets[0].uri);
const getPath = result.assets[0].uri;
results.push(getPath)
console.log(results)
}
}
const mlOCR = async() => {
const resultFromUri = await MlkitOcr.detectFromUri(results[0]);
console.log(resultFromUri)
}
return (
<View style={styles.container}>
<Text style={{color:'black',fontSize:20}}>Preview Image</Text>
<View style={styles.imageContainer}>
{
pickedImagePath !== '' && <Image
source={{ uri: pickedImagePath }}
style={styles.previewImage}
/>
}
</View>
<View style={styles.buttonContainer}>
<TouchableOpacity style={styles.button} onPress={showImagePicker}><Text style={styles.buttonText}>Select from gallery</Text></TouchableOpacity>
<TouchableOpacity style={styles.button} onPress={openCamera}><Text style={styles.buttonText}>Open camera</Text></TouchableOpacity>
</View>
<TouchableOpacity style={{marginTop:15}} onPress={mlOCR}><Text style={{color:'black',fontSize:20}}>Continue</Text></TouchableOpacity>
</View>
);
}
export default App;
I tried tesseract js but i read that it was not supported in expo without ejecting (which i dont know how to do), And If there are any other methods of extracting texts from images please let me know how to do so.
2
Answers
I realized that instead of using an array to push in the URI, I can use an useState:
const [uriValue, setUriValue] = useState(null);
and set the uriValue inside the component where I am launching the camera or the gallery:and this worked for me... now i can just access the uri using uriValue.
you can find more information about mlkit text recognition in https://developers.google.com/ml-kit/vision/text-recognition, there’s also github link of sample app within.
Are you trying to do web development? As far as I know, mlkit is for mobile development only.