Released: Sep 28, 2024

Actualizaciones en V24.9

Características

• Added the ability to reduce PDF file size at the expense of lower background image quality.
• Implemented next-get text-in-wild OCR model with improved recognition accuracy and multi-language support.
• Improved image orientation detection to prevent certain images from being incorrectly rotated upside-down.
• Refined content area detection algorithms for TABLE and PHOTO modes.

Released: Aug 24, 2024

Actualizaciones en V24.8.1

Características

• Improved the recognition accuracy for headers and footers.

Correcciones

• Fixed the issue where some of the image content was ignored by the OCR engine.
• Removed an empty page that occasionally appears when converting a TIFF image to a searchable PDF document.
• Fixed the issue with incorrect detection of image orientation that caused certain images to be rotated upside-down.

Released: Aug 6, 2024

Actualizaciones en V24.8

Características

• Significantly improved automatic correction of rotated and upside-down images.
• Enhanced the handling of background images in searchable PDFs.
• ONNX Runtime dependency has been updated to version 1.18.0.

Released: Jul 31, 2024

Actualizaciones en V24.7

Características

• Improved the stability of image recognition on x86 platforms.

Released: Jun 14, 2024

Actualizaciones en V24.6

Características

• Added the ability to extract essential details (such as a number, name, date of birth, and so on) from US passport images.
• Enabled embedding of user-specified fonts in recognition results saved as PDFs.
• Aspose.OCR.Country.UNIVERSAL recognition parameter that disables extraction of key details from passport images has been replaced with Aspose.OCR.Country.NONE.

Released: May 29, 2024

Actualizaciones en V24.5

Características

• Added support for the automatic detection of problematic areas of an image that can significantly impact the accuracy of OCR.
• Significantly improved the recognition of languages based on the Latin alphabet.

Correcciones

• Fixed character bounding boxes detection.

Released: May 1, 2024

Actualizaciones en V24.4.2

Características

• Added Persian (Farsi) language recognition and recognition of texts in mixed Persian/English.
• Added Urdu language recognition and recognition of texts in mixed Persian/English.
• Added Uyghur language recognition and recognition of texts in mixed Persian/English.
• Improved Arabic text recognition.

Released: Apr 24, 2024

Actualizaciones en V24.4.1

Características

• Improved saving of recognition results as searchable PDFs.
• Improved DetectAreasMode.PHOTO document areas detection mode.

Released: Apr 15, 2024

Actualizaciones en V24.4

Características

• Enabled recognition process cancelation.
• Added support for recognizing Arabic text.

Released: Mar 22, 2024

Actualizaciones en V24.3.1

Características

• Added support for TIFF images with 16 bits per pixel bit depth.