Feature Description

LEADTOOLS provides several different ways to automatically and manually segment images that contain text and image regions for .NET (C# & VB), C/C++, WinRT, iOS, OS X, Java, and Web developers. Image segmentation is important for improving OCR speed as well as increasing compression ratios within complex formats such as LEAD MRC, standard MRC T.44, and PDF. It is also useful in medical imaging for separating and identifying specific tissues, organs, and abnormalities.

Overview of LEADTOOLS Image Segmentation SDK Technology

Image Segmentation for OCR Preprocessing

LEADTOOLS OCR SDK technology automatically (using LEAD's AutoZone technology) detects different zones types such as text, graphic, and table in images. Image segmentation is an important step in OCR preprocessing because it helps improve recognition results and speed. LEADTOOLS exposes its powerful and flexible auto-zoning functionality for developers to use in any application that needs to automatically separate images, tables, and text within mixed-content images.

Image Segmentation for Compression

LEADTOOLS includes mixed raster content (MRC) image segmentation technology to optimize compression for images that consist of both text and color images. Standard compression schemes work well when an image is entirely comprised of either text or color elements. The MRC compression scheme segments images into text and color segment types and compresses each segment independently. This results in both high compression factors while retaining important image detail.

Image Segmentation for Medical Imaging

LEADTOOLS provides many powerful medical image processing functions that can isolate objects within medical images with features such as magic wand (seed-based flood fill) region selection, window level, background removal, tissue equalization, histogram equalization, intensity detection, color threshold, and more. Medical image segmentation helps radiologists, oncologists, dentists, diagnosticians, and other health-care professionals diagnose patient images with more accuracy and efficiency than with traditional film images.

Technology Related to Image Segmentation