Add OCR quality toggle with ML Kit and Tesseract options #1
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
@@ -11,13 +11,16 @@ import androidx.compose.runtime.LaunchedEffect | |||||||
| import androidx.compose.runtime.getValue | ||||||||
| import androidx.compose.runtime.mutableStateOf | ||||||||
| import androidx.compose.runtime.remember | ||||||||
| import androidx.compose.runtime.rememberCoroutineScope | ||||||||
| import androidx.compose.runtime.setValue | ||||||||
| import androidx.compose.ui.platform.LocalContext | ||||||||
| import androidx.navigation.NavHostController | ||||||||
| import androidx.navigation.compose.NavHost | ||||||||
| import androidx.navigation.compose.composable | ||||||||
| import androidx.navigation.compose.rememberNavController | ||||||||
| import com.skeler.scanely.ocr.OcrEngine | ||||||||
| import com.skeler.scanely.ocr.OcrHelper | ||||||||
| import com.skeler.scanely.ocr.OcrQuality | ||||||||
| import com.skeler.scanely.ocr.OcrResult | ||||||||
| import com.skeler.scanely.ui.components.GalleryPicker | ||||||||
| import com.skeler.scanely.ui.screens.CameraScreen | ||||||||
|
|
@@ -43,10 +46,13 @@ fun ScanelyNavigation( | |||||||
| navController: NavHostController = rememberNavController(), | ||||||||
| currentTheme: ThemeMode = ThemeMode.System, | ||||||||
| onThemeChanged: (ThemeMode) -> Unit = {}, | ||||||||
| ocrQuality: OcrQuality = OcrQuality.FAST, | ||||||||
| onOcrQualityChanged: (OcrQuality) -> Unit = {}, | ||||||||
| ocrLanguages: Set<String> = setOf("eng", "ara"), | ||||||||
| onOcrLanguagesChanged: (Set<String>) -> Unit = {} | ||||||||
| ) { | ||||||||
| val context = LocalContext.current | ||||||||
| val scope = rememberCoroutineScope() | ||||||||
|
|
||||||||
| // Shared OCR state | ||||||||
| var selectedImageUri by remember { mutableStateOf<Uri?>(null) } | ||||||||
|
|
@@ -56,14 +62,14 @@ fun ScanelyNavigation( | |||||||
| // Flag to control if OCR should run automatically when selectedImageUri changes | ||||||||
| var shouldAutoScan by remember { mutableStateOf(true) } | ||||||||
|
|
||||||||
| val ocrEngine = remember { OcrEngine(context) } | ||||||||
|
There was a problem hiding this comment. The |
||||||||
| val historyManager = remember { com.skeler.scanely.data.HistoryManager(context) } | ||||||||
|
|
||||||||
| // Auto-initialize on Quality or Language change | ||||||||
| LaunchedEffect(ocrQuality, ocrLanguages) { | ||||||||
| if (ocrLanguages.isNotEmpty()) { | ||||||||
| kotlinx.coroutines.withContext(kotlinx.coroutines.Dispatchers.IO) { | ||||||||
| ocrEngine.reinitialize(ocrQuality, ocrLanguages.toList()) | ||||||||
| } | ||||||||
| } | ||||||||
| } | ||||||||
|
|
@@ -85,12 +91,15 @@ fun ScanelyNavigation( | |||||||
|
|
||||||||
| navController.navigate(Routes.RESULTS) | ||||||||
|
|
||||||||
| scope.launch(kotlinx.coroutines.Dispatchers.IO) { | ||||||||
| try { | ||||||||
| // Get Tesseract helper for PDF processing (always use Tesseract for multi-language PDFs) | ||||||||
| val tesseractHelper = ocrEngine.getTesseractHelper() | ||||||||
|
Comment on lines
+96
to
+97
There was a problem hiding this comment. The comment explains that PDF processing uses Tesseract for multi-language support, but this creates an inconsistency in user experience. If a user selects "Best (ML Kit)" quality and then processes a Latin-script PDF, they would expect higher quality OCR, but they'll get Tesseract instead. This should either be documented in the UI (e.g., "Note: PDFs always use Tesseract engine") or the logic should respect the user's quality preference for single-language PDFs. |
||||||||
|
|
||||||||
| // Ensure Tesseract is initialized | ||||||||
| if (!tesseractHelper.isReady()) { | ||||||||
| progressMessage = "Initializing OCR..." | ||||||||
| val initSuccess = tesseractHelper.initialize(ocrLanguages.toList()) | ||||||||
| if (!initSuccess) { | ||||||||
| isProcessing = false | ||||||||
| isPdfProcessing = false | ||||||||
|
|
@@ -103,7 +112,7 @@ fun ScanelyNavigation( | |||||||
| val pdfResult = com.skeler.scanely.ocr.PdfProcessor.extractTextFromPdf( | ||||||||
| context = context, | ||||||||
| pdfUri = uri, | ||||||||
| ocrHelper = tesseractHelper, | ||||||||
| enabledLanguages = ocrLanguages.toList(), | ||||||||
| onProgress = { update -> | ||||||||
| progressMessage = if (update.statusMessage.isNotEmpty()) { | ||||||||
|
|
@@ -154,6 +163,7 @@ fun ScanelyNavigation( | |||||||
| } | ||||||||
|
|
||||||||
| // Process image when selected (skip if PDF is being processed OR if shouldAutoScan is false) | ||||||||
| // Note: ocrQuality is intentionally NOT in dependencies - quality changes only affect new scans | ||||||||
| LaunchedEffect(selectedImageUri, isPdfProcessing, shouldAutoScan) { | ||||||||
|
Comment on lines
+166
to
167
There was a problem hiding this comment. The comment states "ocrQuality is intentionally NOT in dependencies" but this could lead to unexpected behavior. When a user changes OCR quality in settings and returns to the main screen with a previously selected image, the quality change won't take effect until they select a new image. This behavior should be clearly documented in the UI or the dependency should be added to reinitialize when quality changes.
Suggested change
|
||||||||
| if (isPdfProcessing) return@LaunchedEffect | ||||||||
|
|
||||||||
|
|
@@ -167,9 +177,9 @@ fun ScanelyNavigation( | |||||||
| isProcessing = true | ||||||||
| ocrResult = null | ||||||||
|
|
||||||||
| val initialized = ocrEngine.initialize(ocrQuality, ocrLanguages.toList()) | ||||||||
| if (initialized) { | ||||||||
| val result = ocrEngine.recognizeText(uri) | ||||||||
| if (result != null) { | ||||||||
| ocrResult = result | ||||||||
| // Save to history automatically if successful | ||||||||
|
|
@@ -232,6 +242,8 @@ fun ScanelyNavigation( | |||||||
| SettingsScreen( | ||||||||
| currentTheme = currentTheme, | ||||||||
| onThemeChange = onThemeChanged, | ||||||||
| ocrQuality = ocrQuality, | ||||||||
| onOcrQualityChanged = onOcrQualityChanged, | ||||||||
| ocrLanguages = ocrLanguages, | ||||||||
| onOcrLanguagesChanged = onOcrLanguagesChanged, | ||||||||
| onNavigateBack = { navController.popBackStack() } | ||||||||
|
|
||||||||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,170 @@ | ||
| package com.skeler.scanely.ocr | ||
|
|
||
| import android.content.Context | ||
| import android.graphics.Bitmap | ||
| import android.net.Uri | ||
| import android.util.Log | ||
| import com.google.mlkit.vision.common.InputImage | ||
| import com.google.mlkit.vision.text.TextRecognition | ||
| import com.google.mlkit.vision.text.TextRecognizer | ||
| import com.google.mlkit.vision.text.latin.TextRecognizerOptions | ||
| import kotlinx.coroutines.Dispatchers | ||
| import kotlinx.coroutines.suspendCancellableCoroutine | ||
| import kotlinx.coroutines.sync.Mutex | ||
| import kotlinx.coroutines.sync.withLock | ||
| import kotlinx.coroutines.withContext | ||
| import kotlin.coroutines.resume | ||
|
|
||
| private const val TAG = "MlKitOcrHelper" | ||
|
|
||
| /** | ||
| * ML Kit Text Recognition OCR helper. | ||
| * Provides high-quality text recognition for Latin script. | ||
| * | ||
| * Uses Google's ML Kit which runs on-device for privacy. | ||
|
Comment on lines
+20
to
+24
There was a problem hiding this comment. The documentation states that both OCR engines run locally/offline and maintain privacy-first design, but there's no mention that ML Kit may download models on first use. While ML Kit does support on-device processing, it typically needs to download the text recognition model initially, which requires an internet connection. This should be documented to set proper user expectations about the "offline" capability. |
||
| */ | ||
| class MlKitOcrHelper(private val context: Context) { | ||
|
|
||
| private val mutex = Mutex() | ||
| private var recognizer: TextRecognizer? = null | ||
| private var isInitialized = false | ||
|
|
||
| /** | ||
| * Initialize ML Kit Text Recognizer. | ||
| * ML Kit automatically handles model downloading and initialization. | ||
| * | ||
| * @param languages Ignored for ML Kit (uses Latin recognizer) | ||
| * @return true if initialization succeeded | ||
| */ | ||
| suspend fun initialize(languages: List<String> = emptyList()): Boolean = withContext(Dispatchers.IO) { | ||
| mutex.withLock { | ||
| try { | ||
| if (isInitialized && recognizer != null) { | ||
| return@withContext true | ||
| } | ||
|
|
||
| // Create text recognizer with Latin script options | ||
| recognizer = TextRecognition.getClient(TextRecognizerOptions.DEFAULT_OPTIONS) | ||
| isInitialized = true | ||
|
|
||
| Log.d(TAG, "ML Kit Text Recognizer initialized successfully") | ||
| true | ||
| } catch (e: Exception) { | ||
| Log.e(TAG, "ML Kit initialization failed", e) | ||
| false | ||
| } | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Recognize text from an image URI. | ||
| */ | ||
| suspend fun recognizeText(imageUri: Uri): OcrResult? = withContext(Dispatchers.IO) { | ||
| if (!isInitialized || recognizer == null) { | ||
| Log.e(TAG, "MlKitOcrHelper not initialized") | ||
| return@withContext null | ||
| } | ||
|
|
||
| try { | ||
| val inputImage = InputImage.fromFilePath(context, imageUri) | ||
| recognizeFromInputImage(inputImage) | ||
| } catch (e: Exception) { | ||
| Log.e(TAG, "Failed to create InputImage from URI", e) | ||
| null | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Recognize text from a bitmap. | ||
| * Note: ML Kit handles thread safety internally, so no mutex is needed. | ||
|
There was a problem hiding this comment. This comment is misleading. According to the official ML Kit documentation, |
||
| */ | ||
| suspend fun recognizeText(bitmap: Bitmap): OcrResult? = withContext(Dispatchers.IO) { | ||
| if (!isInitialized || recognizer == null) { | ||
| Log.e(TAG, "MlKitOcrHelper not initialized") | ||
| return@withContext null | ||
| } | ||
|
|
||
| try { | ||
| val inputImage = InputImage.fromBitmap(bitmap, 0) | ||
| recognizeFromInputImage(inputImage) | ||
| } catch (e: Exception) { | ||
| Log.e(TAG, "Failed to create InputImage from Bitmap", e) | ||
| null | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Perform text recognition from InputImage. | ||
| * Uses mutex to ensure thread-safety since TextRecognizer is not thread-safe. | ||
| */ | ||
| private suspend fun recognizeFromInputImage(inputImage: InputImage): OcrResult? { | ||
| return mutex.withLock { | ||
| suspendCancellableCoroutine { continuation -> | ||
| val startTime = System.currentTimeMillis() | ||
|
|
||
| recognizer?.process(inputImage) | ||
| ?.addOnSuccessListener { visionText -> | ||
| val processingTime = System.currentTimeMillis() - startTime | ||
|
|
||
| val text = visionText.text | ||
|
|
||
| // Calculate confidence from element-level confidence scores | ||
| val confidences = visionText.textBlocks.flatMap { block -> | ||
| block.lines.flatMap { line -> | ||
| line.elements.mapNotNull { it.confidence } | ||
| } | ||
| } | ||
|
|
||
| val avgConfidence = if (confidences.isNotEmpty()) { | ||
| (confidences.average() * 100).toInt() | ||
| } else { | ||
| 0 | ||
| } | ||
|
|
||
| Log.d(TAG, "OCR completed in ${processingTime}ms, confidence: $avgConfidence%") | ||
|
|
||
| continuation.resume( | ||
| OcrResult( | ||
| text = text, | ||
| confidence = avgConfidence, | ||
| languages = listOf("eng"), // ML Kit Latin | ||
| processingTimeMs = processingTime | ||
| ) | ||
| ) | ||
| } | ||
| ?.addOnFailureListener { e -> | ||
| Log.e(TAG, "ML Kit text recognition failed", e) | ||
| continuation.resume(null) | ||
| } | ||
| ?: continuation.resume(null) | ||
| } | ||
| } | ||
| } | ||
|
Comment on lines
+100
to
+142
There was a problem hiding this comment. According to the official ML Kit documentation, To ensure thread safety, you should wrap the logic in return mutex.withLock {
suspendCancellableCoroutine { continuation ->
val startTime = System.currentTimeMillis()
recognizer?.process(inputImage)
?.addOnSuccessListener { visionText ->
val processingTime = System.currentTimeMillis() - startTime
val text = visionText.text
// Calculate confidence from block-level confidence
val confidences = visionText.textBlocks.mapNotNull { block ->
block.lines.flatMap { line ->
line.elements.mapNotNull { it.confidence }
}
}.flatten()
val avgConfidence = if (confidences.isNotEmpty()) {
(confidences.average() * 100).toInt()
} else {
0
}
Log.d(TAG, "OCR completed in ${processingTime}ms, confidence: $avgConfidence%")
continuation.resume(
OcrResult(
text = text,
confidence = avgConfidence,
languages = listOf("eng"), // ML Kit Latin
processingTimeMs = processingTime
)
)
}
?.addOnFailureListener { e ->
Log.e(TAG, "ML Kit text recognition failed", e)
continuation.resume(null)
}
?: continuation.resume(null)
}
} |
||
|
|
||
| /** | ||
| * Check if the helper is ready. | ||
| */ | ||
| fun isReady(): Boolean = isInitialized && recognizer != null | ||
|
|
||
| /** | ||
| * Release ML Kit resources. | ||
| */ | ||
| fun release() { | ||
| try { | ||
| recognizer?.close() | ||
| recognizer = null | ||
| isInitialized = false | ||
| Log.d(TAG, "ML Kit resources released") | ||
| } catch (e: Exception) { | ||
| Log.e(TAG, "Error releasing ML Kit", e) | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Reinitialize (for compatibility with OcrHelper interface). | ||
| */ | ||
| suspend fun reinitialize(languages: List<String>): Boolean { | ||
| release() | ||
| return initialize(languages) | ||
| } | ||
| } | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The logic to load the OCR quality preference from SharedPreferences can be simplified. The
getStringmethod with a default value will never returnnull, making the elvis operator (?:) on line 39 redundant. You can make this logic more concise and idiomatic usingrunCatching.