See all results >
Table of contents

Release Notes v3.x - DynamsoftCaptureVisionRouter Module

3.6.2000 (09/14/2026)

Highlights

Multi-Threaded Barcode Decoding

  • Get results sooner with parallel processing - Barcode decoding now uses a breadth-first strategy that decomposes a single DBR Task into one Localization Work and one or more Decoding Works. This improves thread utilization and reduces the chance that a slow DeblurMode attempt blocks other faster decoding attempts, helping valid results come back sooner.

DataMatrix Color Inversion Detection

  • Handle normal and inverted DataMatrix more efficiently - Added AutoDetectColorInversion to automatically handle both normal and inverted DataMatrix barcodes. Instead of processing the whole image twice, the SDK applies dual-polarity handling only to localized DataMatrix regions, which makes processing faster in dual-polarity scenarios.

Cross-Version License Support

  • Use a single license across SDK versions - Full License 1.0 keys (starting with “f”) that are non-perpetual are no longer version-checked, so the same key can be used across SDK versions without reactivation.

Text Line Orientation Detection

  • Improved handling of rotated text lines - Added OrientationDetectionModes to detect text line orientation and correct it for further processing. Two modes are available: ODM_SPATIAL_REFERENCES and ODM_CHARS_ORIENTATION_NEURAL_NETWORK.
  • Current limitations - This capability is currently effective only in MRZ scenarios and supports only upright and upside-down text lines (0 degrees and 180 degrees). It does not yet handle 90 degrees, 270 degrees, or arbitrary rotation angles.

New

  • Added support for Micro PDF417-specific decoding mode.

  • Added GetCentrePoint method to Quadrilateral to get the centre point of the quadrilateral.

  • Added ConvertRectToViewCoordinates and ConvertPointToViewCoordinates methods to CameraEnhancer to convert video coordinates into MAUI camera view coordinates (measured in dp).

  • Added AutoDetectColorInversion parameter for BarcodeFormatSpecification to support automatic color-inversion detection for DataMatrix barcodes.

  • Added OrientationDetectionModes parameter for the SST_LOCALIZE_TEXT_LINES stage with two supported modes: ODM_SPATIAL_REFERENCES and ODM_CHARS_ORIENTATION_NEURAL_NETWORK.

  • Added TextLineOrientationCls.data model file for text line orientation classification.

Changed

  • MaxParallelTasks now controls the total number of Work-level threads in the CVR thread pool. For DBR tasks, each Localization Work and Decoding Work occupies one thread slot. DLR and DDN tasks continue to occupy one thread per task.

  • SetDeviceFriendlyName now enforces parameter constraints: maximum 64 characters, allowed characters are letters (a-z, A-Z), digits (0-9), hyphen (-), underscore (_), and period (.), and must start and end with a letter or digit. Returns EC_PARAMETER_VALUE_INVALID if constraints are not met.

  • Improved the default display behavior of corner adjustment points in ImageEditorView. Previously, users had to tap the view before the corner adjustment points became visible.

Fixed

  • Fixed an issue in GS1-Databar AI 17 (YYMMDD) results where the month field could be missing a leading zero.
  • Fixed several known crash issues.

3.4.3000 (07/10/2026)

Security Updates

  • Updated third-party libraries to incorporate the latest security fixes.

3.4.1300 (05/20/2026)

Fixed

  • Fixed potential issues that could occur on devices running a kernel with a 16KB page size.
  • Fixed crash and hang issues that could occur in certain scenarios.

3.4.1200 (04/09/2026)

Highlights

AI-Powered Barcode Detection and Decoding

  • PDF417 Localization Model – Introduces the PDF417Localization neural network model for improved detection of PDF417 barcodes, especially under challenging conditions.
  • Code39/ITF Decoding Model – Adds the Code39ITFDecoder model for enhanced decoding of Code 39 and ITF barcodes under blurred or low-resolution conditions.
  • Deblur Models for 2D Barcodes – Adds the DataMatrixQRCodeDeblur and PDF417Deblur models provide more effective recovery from motion and focus blur for DataMatrix, QR Code, and PDF417 barcodes.

ECI (Extended Channel Interpretation) Support

  • ECI Information Return – Adds support for retrieving Extended Channel Interpretation (ECI) data from barcodes. The new ECISegment class, along with ECISegments property in BarcodeResultItem class, enables access to character encoding information embedded in barcodes.
  • ECI-Based Text Interpretation – Adds support for interpreting ECI segments during barcode decoding, improving compatibility with international character sets.

Performance Improvements

  • On-Demand Model Loading – Implements lazy loading for AI models, reducing initialization time by loading models only when first needed.
  • Smart Model Selection – Models are now loaded based on configured barcode formats, minimizing memory usage by excluding unused models.
  • Improved Confidence Scoring – Enhances confidence score calculation for results from neural network models, providing more accurate quality indicators.
  • DPM Barcode Optimization – Improves recognition rate for Direct Part Marking (DPM) barcodes commonly used in industrial and manufacturing environments.

Identity Document Processing

  • Enhanced Passport Processing – Improves document edge detection accuracy for passport documents through optimized processing workflows.
  • Portrait Zone Detection – The MRZLocalization model now supports detecting portrait zone on identity documents, enabling automatic extraction of photo regions.

New

Changed

  • Barcode text encoding fallback changed from UTF-8 to ISO-8859-1 when no ECI information is present in the barcode.
  • Improved license binding stability on MacOS devices.
  • Updated default value of compensation parameter in ImageProcessor.ConvertToBinaryLocal from 0 to 10.
  • ConvertToBinaryGlobal and ConvertToBinaryLocal of ImageProcessor class now support color and binary images as input in addition to grayscale images.

Removed

Fixed

  • Fixed incorrect coordinate in barcode result when using neural network models with a specified region.
  • Fixed crash and hang issues that could occur in certain scenarios.
  • Fixed various minor bugs and improved overall stability.

3.2.5000 (12/18/2025)

This release includes security maintenance updates to ensure continued protection of the product.

Security Updates

  • Updated third-party libraries to incorporate the latest security fixes.

3.2.3000 (11/20/2025)

🎉Milestone Release

Version 3.2.3000 introduces a series of AI-driven improvements designed to enhance barcode and MRZ detection accuracy, processing speed, and configuration flexibility.

This release focuses on practical performance gains for production environments across retail, logistics, manufacturing, and identity verification workflows.

✨ Key Highlights

AI-Powered Barcode Detection and Decoding

  • New Localization Models – Introduces OneDLocalization and DataMatrixQRCodeLocalization neural network models for improved detection of blurred / low-resolution 1D codes, or partially damaged DataMatrix/QR codes.
  • Specialized Decoders – Adds EAN13Decoder and Code128Decoder models optimized for long-distance and motion-blurred decoding scenarios.
  • Redesigned Deblur Model – The OneDDeblur model now provides more effective recovery from motion and focus blur.
  • Configurable Model Selection – The new ModelNameArray parameter supports flexible model loading and fine-grained control for specific barcode types.

Precision and Processing Control

  • Enhanced Deblur Methods – DM_DEEP_ANALYSIS now includes sub-level control with OneDGeneral, TwoDGeneral, and EAN13Enhanced options.
  • Barcode Count Expectation – The new ExpectedBarcodesCount parameter enables format-specific quantity control and early termination in fixed-count workflows.
  • Improved Region Detection – The new RPM_GRAY_CONSISTENCY mode provides more precise region extraction based on grayscale uniformity and local consistency for document and label processing.

AI-Powered MRZ Detection

  • Neural MRZ Localization – The new MRZLocalization model improves region detection accuracy and delivers up to 42.7% faster processing for MRZ-based document workflows.
  • Configurable Localization Control – The new LocalizationModes parameter allows configuration for text line detection.

Smart Document Capture

  • Clarity-Based Frame Selection – Automatically selects the sharpest and highest-quality frame in live capture workflows.
  • Cross-Frame Verification – Updated verification algorithms enhance result reliability.

Performance Highlights

Barcode Workflows

  • Up to 26.5% higher read rates under blur conditions with as much as 44% faster processing.
  • Reliable decoding of DataMatrix and QR codes with missing or damaged finder patterns.
  • Extended operational range beyond 75 cm for long-distance barcode scanning.

Document Workflows

  • Improved performance in live video capture environments.
  • Consistent document quality through clarity-based frame evaluation.
  • Faster MRZ processing for high-throughput identity verification

Developer Notes

  • Backward Compatibility – Fully compatible with existing integrations; no code-level changes required for upgrade.
  • Configuration Flexibility – Expanded parameter set allows comprehensive model configuration for scenario-specific tuning.
  • Production Stability – All new models validated in enterprise environments.

New

3.0.5200 (08/18/2025)

Fixed

  • Fixed an xcframework signature issue.

3.0.5100 (08/12/2025)

Fixed

  • Small fixes and tweaks.

3.0.3100 (05/30/2025)

Highlights

  • Workflow Improvements
    • Restructured the parameter control hierarchy at all levels for finer scope definition and more granular process management, with the stage level newly added.
    • Enabled custom combinations and sequences of sections, increasing flexibility and operational customization under specific conditions.
    • Redesigned document normalization sections to better accommodate diverse document processing operations.
  • Deep Learning Integration
    • Improved the reading rate of 1D barcode by introducing a new deblurring deep-learning model.
    • Enhanced text recognition capabilities with deep learning-based text-line recognition.
  • Algorithm Enhancements
    • Enabled deduplication at the Region of Interest (ROI) level to consolidate results from multiple tasks.
    • Enhanced the text recognition workflow by integrating improved multi-step recognition processes and validation methods.
    • Improved the CODE_128 and DataMatrix DeepAnalysis algorithms for better decoding accuracy and performance.
    • Added support for new barcode types: CODE_32, MATRIX_25, KIX, and TELEPEN.
    • Added GS1 Application Identifiers (AI) support for improved code parsing capabilities.
  • Engineering Optimizations
    • Unified template-loading logic to reduce I/O overhead.
    • Implemented conversion functionality between ImageData and image files, including both on-disk and in-memory files.

This page is compatible for: