Barcode-Kit

Overview

Barcode-Kit is a flutter package that allows you to read barcodes using the camera. It uses native textures to display the camera feed and the barcode overlay. It is built on top of the Google ML Kit and iOS Vision for barcode scanning.

It ships as a single widget that handles most of the work for you.

Prerequisites

• Flutter SDK installed on your machine.
• Familiarity with Flutter development.

Installation

Add the mtrust_barcode_kit to your Flutter project via the pub add command

flutter pub add mtrust_barcode_kit

or manually add it to your pubspec.yaml

dependencies:
  mtrust_barcode_kit: ^2.1.0

Usage

BarcodeKitView(
    // The barcode formats to scan
    formats: _formats,
    cameraFit: BoxFit.cover,
    // The camera overlay mask
    maskHeight: 200,
    maskWidth: 200,
    widgetAboveMask: Widget // Place something above the camera cutout,
    widgetBelowMask: Widget // Place something below the camera cutout,

    // Whether the camera feed is frozen and barcode scanning is paused
    paused: true/false,

    maskAdditionpauseOpacity: 0.2,

    // Whether the camera is rotated along with the device.
    followRotation: true/false,

    // The callback when a barcode is scanned
    onBarcodeScanned: (barcode) {

    },
    // Custom ui for building requesting permissions and loading screens.
    // Build your own by extending the BarcodeKitUiBuilder
    uiBuilder: BarcodeKitUiBuilder(),
)

Barcode Formats

The formats parameter is a list of barcode formats to scan. The supported formats are:

Format	iOS	Android
AZTEC	✅	✅
CODABAR	❌	✅
CODE39	✅	✅
CODE93	✅	✅
CODE128	✅	✅
DATAM ATRIX	✅	✅
EAN8	✅	✅
EAN13	✅	✅
ITF	✅	✅
PDF417	✅	✅
QR-CODE	✅	✅
UPCA	❌	✅
UPCE	✅	✅

OCR (Optional Text Recognition)

In addition to barcode scanning, BarcodeKitView can optionally run on-device OCR (Google ML Kit on Android, Apple Vision on iOS) over the same masked camera region, useful as a fallback for reading identifiers (e.g. lot/product/ serial numbers) that aren't printed as a barcode.

BarcodeKitView(
    // ...
    enableOCR: true,
    // Discards recognized text lines below this confidence (0..1).
    // Defaults to 0 (no filtering).
    minTextConfidence: 0.6,
    onTextDetected: (DetectedText detectedText) {
      print(detectedText.text);
    },
)

onTextDetected is called once per recognized line of text, with a DetectedText carrying:

• text - the recognized line content.
• confidence - recognition confidence (0..1).
• cornerPoints - the line's bounding quadrilateral, in the same full camera-texture pixel coordinate space as DetectedBarcode.cornerPoints.
• blockIndex/lineIndex - grouping/ordering within the page (ML Kit only groups lines into blocks; blockIndex is always 0 on iOS).

The mask cutout (maskWidth/maskHeight) restricts both barcode detection and OCR to the same visible region.

Identifier Classification

Raw OCR text doesn't tell you which line is actually the identifier you care about (a lot number, product number, etc.) versus surrounding label text or unrelated printed text. IdentifierClassifier scores each detected line by how "identifier-like" it is, using shape-agnostic heuristics (digit/letter mixing, character-class transitions, phonotactic naturalness, separators, length, digit ratio) plus optional proximity to configurable context labels (e.g. LOT, REF, S/N), without needing to know the exact format/length of the identifier in advance.

final classifier = IdentifierClassifier(
  IdentifierClassifierConfig(
    // Labels that identify nearby/inline text as an identifier value.
    contextLabels: ['LOT', 'REF', 'S/N', 'SN', 'PN', 'BATCH'],
    // Known real words that should never be treated as an identifier.
    negativeWordList: {'net', 'weight', 'contents'},
  ),
);

BarcodeKitView(
    // ...
    enableOCR: true,
    onTextDetected: (detectedText) {
      final scored = classifier.classify(detectedText);
      if (!scored.isContextLabel && scored.confidence > 0.6) {
        print('Likely identifier: ${scored.token} (${scored.confidence})');
      }
    },
)

Feed every DetectedText event into classify() - the classifier internally maintains its own sliding window of recently seen labels (by position and time), so no manual batching or frame bookkeeping is required. See the dartdoc on IdentifierClassifierConfig for all available tuning options (feature weights, proximity threshold, context window duration, minimum identifier length, etc.).

Contributing

We welcome contributions! Please fork the repository and submit a pull request with your changes. Ensure that your code adheres to our coding standards and includes appropriate tests.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details.