Changelog

Plate ID OCR Filter release notes

v0.1.7 - 2025-09-29

Changed

• Updated documentation

v0.1.6 - 2025-09-27

Added

• Multi-topic processing with main-first output ordering
• forward_upstream_data config to pass-through non-image frames
• End-to-end usage script
• Tests: tests/test_integration_config_normalization.py, tests/test_smoke_simple.py

Improved

• Config normalization: string-to-type coercion and FILTER_* env support
• Overlay robustness (font scaling guard, bounds checks)

v0.1.5 - 2025-08-07

Changed

• Updated dependencies
• Added Python 3.13 support

v0.1.4 - 2025-08-07

Changed

• Updated dependencies

v0.1.3 - 2025-08-01

Changed

• Updated dependencies

v0.1.2 - 2025-07-16

Changed

• Updated dependencies

v0.1.1 - 2025-05-22

Changed

• Updated dependencies

v0.1.0 - 2025-05-22

Added

• Initial release of the Plate ID OCR Filter for visualizing OCR license plate results on video frames.
• Overlays OCR-detected license plate text onto the main frame using:
  • Automatic font scaling to fit within a padded text box.
  • Configurable font scale and thickness.
• Imposes the corresponding cropped license plate image (if available) in the top-left corner of the main frame:
  • Uses inset_size and inset_margin to control placement and size.
• OCR filtering:
  • Validates and filters OCR strings using a regex pattern for standard license plate format (ABC1234).
  • Caches and reuses the last seen valid plate when no new valid text is found.
• Customizable behavior via config:
  • cropped_topic_suffix to specify which frame topic contains cropped license plates.
  • font_scale, font_thickness, inset_size, inset_margin tunable via config or environment variables.
• Frame integrity:
  • Ensures main frame is only updated when both image and overlay bounds are valid.
  • Gracefully skips overlays when dimensions do not fit.
• Standalone CLI support via FilterLicenseAnnotationDemo.run().

Changed

• Improved config normalization with .env support for all key fields.
• Centralized inset and font sizing logic to reduce duplication and improve robustness.
• Enhanced debug logging to show setup state and filtered OCR text results.

Fixed

• Fixed parsing of inset_size and inset_margin from string format (e.g., "200x60") to tuple.
• Resolved potential errors from overlaying text or images outside the frame bounds.
• Ensured fallback to previous license plate text avoids unnecessary frame updates.

Internal

• Added detailed docstrings for config class fields and core methods.
• Included validation for environment-derived config values.
• Improved log output clarity during setup and processing phases.