DSSDK logo Document Scanner SDK

Version history Pixelnetica™ Document Scanning SDK for Apple iOS

All notable changes to the Document Scanning SDK for iOS are recorded here, so you can see what changed before updating.

3.0.1

Fixed

  • The camera shutter no longer stalls on iPads when the live document detector is on. On some iPad models the still capture could hang for a long time after you tapped the shutter, because continuous document detection kept the camera’s processing queue busy and starved the capture. Detection now pauses while a photo is being taken, so the shutter responds promptly.
  • The camera screen’s torch button now reflects the real state of the torch. Previously the torch icon could show as on while the light stayed off (for example, when the system briefly refused the torch), and it did not update when iOS turned the torch off on its own. The icon now follows the actual torch state and self-corrects after a system-initiated shutoff, so the control and the light always agree.

Changed

  • On large iPads the live camera preview now streams at a moderate resolution instead of the sensor’s full video resolution, and document detection runs on a downscaled frame. This lowers CPU use, heat, and battery drain while scanning. The resolution of the captured photo is unchanged.
  • The camera screen’s torch button now appears only on devices that have a torch. On hardware without a rear torch (for example, some iPads) the button is hidden instead of being shown as a control that does nothing.
  • The bundled EasyScanner sample app is now localized into German, Spanish, French, Italian, Simplified Chinese, and Japanese — including the ready-made SDK-UI screens (camera, page editor, language picker), which it localizes from its own app bundle to show how you can localize those screens in your app. The frameworks themselves continue to ship their strings in English.

3.0.0

A full rewrite of the Document Scanning SDK for iOS. If you are upgrading from 2.x, read the migration guide first — the API surface has changed.

Added

  • Swift Package Manager distribution. Add https://github.com/Pixelnetica/DocScanningSDK-iOS as a package dependency; the SDK ships as two binary frameworks: DocScanningSDK (core scanning, image processing, and document export) and DocScanningSDK-UI (ready-made screens).
  • On-device OCR with support for over 100 languages. Recognition runs entirely on the device — no cloud processing — and handles multi-language documents in a single pass, including right-to-left (RTL) scripts. Recognised text exports as plain text or as a searchable PDF.
  • Offline text-orientation detection (OSD). Automatically corrects page orientation from document content, and works on a fresh install with no network download.
  • PDF composing engine with advanced image compression — up to 90% smaller files for colour and greyscale pages and 50% for black-and-white, with selectable compression levels from “Lossless” to “Extreme”. It produces layered (“sandwiched”) PDFs, embedding a searchable text layer above the page image for indexing, search, and copy.
  • Batteries-included UI screens in DocScanningSDK-UI, each adoptable in a single line: a smart document camera with live border detection, automatic capture, and interactive quality guidance (trapezoid-distortion and effective-document-area indicators); an OCR results editor with manual correction; an OCR language picker with search; and a page/border editor.
  • A bundled, browsable API reference and a refreshed documentation set, including a 2.x → 3.x migration guide.
  • Third-party attribution notices bundled with the framework.
  • An Apple privacy manifest (PrivacyInfo.xcprivacy) in each framework, declaring on-device processing with no tracking and no data collection. This lets your app’s App Store submission account for the SDK automatically.

Changed

  • Rebuilt on a new Objective-C API surface. Existing 2.x integrations need code changes; see the migration guide.
  • The minimum supported iOS version is now 16.3.

2.5.2

  • Added compatibility with iOS 15 and Xcode 13.
  • Fixed Camera UI on iOS 15.
  • Minor bug fixes and demo improvements.

2.5.1

  • Added compatibility with Xcode 12 and above.
  • Minor bug fixes and demo improvements.

2.4.1

  • Fixed: Pure black or white images processing.

2.4

  • Added compatibility with Apple iOS 13.

2.3.4

  • Added additional license key initialization method - from license.txt file.
  • Application Bundle ID changed to conform Apple App Store requirements.
  • Demo application published at App Store as EasyScan: PDF Doc Scanner.
  • Demo application source code available on GitHub.
  • Added license debug info.
  • Demo application UI updates.

2.3.3

  • Added build-phase script to safely and conveniently strip unnecessary architecture slices from framework.
  • Demo application UI compatibility with iPhoneX series improved.
  • Minor demo application UI updates and fixes.

2.3.2

  • Added demo application camera compatibility with iPhoneX series models.
  • Fixed problem with invalid document corners error in case of “stretched parallelogram” document borders shape.
  • Fixed several memory leaks issues.
  • Changed PxMetaImage::image property from retain to assign.
  • Changed method names from createXXX to newXXX for the ARC compatibilty. Specifically, this change affected the following methods:
    • newWithType: (PxImageWriter)
    • new: (PxMetaImage)
    • new: (PxMetaImage) withMetadata:
    • new: (PxMetaImage) withPath:
    • new: (PxMetaImage) withURL:
  • Minor bug fixes and performance improvements.

2.3.1

  • Added compatibility with Apple iOS 12.

2.3

  • DSSDK now available as an Embedded Framework.
  • Changed PxDocCutout class.
  • Added PxAutoShotDetector class - a more convenient helper to detect when document corners are stable and it is safe to make a camera shot.

2.2

  • New license protection and key format now supporting subscription based licensing.
  • Minor bug fixing and improvements in demo application and documentation.

2.1

  • Added Effective Document area (the area document occupies in the viewfinder) and Trapezoid Distortion control into demo application camera.
  • Minor bug fixing and improvements in demo application.
  • Minor Documentation update.

2.0

  • Added new protocol PxImageWriter (and related enum) for document writing in various formants.
  • Added GLImageSDK::detectDocumentCorners2:correctDocument:corners, with more precise document corner detection algorithm.
  • GLImageSDK::correctDocument:corners now can corrects user changed document corners.
  • Added class DocumentCutout for document corners check and correction.
  • Added class-factory PxImageWriterFactory.
  • Added new protocol PxMetaImage (and related enum), representing Image and related EXIF metadata.
  • Added class-factory PxMetaImageFactory.
  • Added class PxException for developer mistakes indication (e.g. wrong enum value or unsupported method call, etc.).
  • Added VertexDrawingFilter.
  • Documentation update:
    • New documentation format - now in convenient html.
    • Content update.
  • Demo Application update:
    • Resulted document could be written as: PDF, PNG, TIFF (G4 compressed) and Jpg files.
    • Simulate multi-page output (writes same page three time in row) to demonstrate how to create multi-page documents.
    • Resulted document now could be shared through standard iOS dialogue.

1.3.1

  • Fixed bug with torch in demo application.

1.3

  • Demo application Smart Camera and AutoShot support.
  • Demo application camera UI update.
  • Demo application image rotation support.
  • SDK lib now supports multiple bundle ids.

1.2.3

  • Fixed bug in imageWithoutRotation method.

1.2.2

  • Increased compatibility with iOS 8x versions.

1.2

  • *Added Bitcode support.
  • Library became universal and includes support (slices) for devices as well as simulator in single file.
  • Live document border detection support added.
  • Demo application updated:
    • Added ability to scan documents from camera.
    • Smart Camera with live document border detection in viewfinder.
    • Added “Settings” section with ability to change color processing profile as well SmartCrop feature switching.

See also

Top