Added support for text recognition from images provided as an array of bytes.
Added a preprocessing filter that automatically removes dirt, spots, scratches, glare, unwanted gradients, and other noise from photos and scans.
Added a preprocessing filter that automatically straightens skewed images.
Public API changes and backwards compatibility
This section lists all public API changes introduced in Aspose.OCR for C++ 22.8 that may affect the code of existing applications.
Added public APIs:
The following public APIs have been introduced in this release:
A new preprocessing filter that automatically corrects image tilt (deskews an image) before proceeding to recognition. Skew angle detection and image straightening is critical to the OCR process as it directly affects the reliability and efficiency of segmentation and text extraction.
A new preprocessing filter that automatically removes noise from images before proceeding to recognition. Dirt, spots, scratches, glare, unwanted gradients, and other noise are a common problem when scanning low-quality sources such as newspapers or old books, or when taking photographs. These image defects can interfere with recognition, significantly reduce the accuracy of OCR, and may cause spots to be misrecognised as characters.
When this filter is applied, images are preprocessed by a specialized neural network, which consumes additional CPU resources and memory.
Updated public APIs:
The following public APIs have been updated in this release:
The recognized image may be provided as a flat byte array representing the amount of each color per pixel in RGB (red-green-blue) order. The amount can range from 0 to 255, where 0 means no that color and 255 is the maximum amount of that color.
Removed public APIs:
The examples below illustrates the changes introduced in this release: