by

High Performance Toolkit for Post-Scan Image Enhancement

    PixTools IP DataSheet.pdf

With the growing popularity of production and personal scanners, users are integrating paper-based information into their applications more than ever before. And, with the proliferation of Intranet/ Internet and fax communications, businesses need to be able to view information at their desktops in real time. Users are seeking new ways to enhance document images and automatically capture embedded bar code information.

PixTools/Image Processing is a full featured toolkit that provides twelve different operations which can be applied to an image, alone or in combination. These operations, or filters, can be used to improve quality and readability, reduce file size, or both.

The toolkit can be used alone, or in conjunction with another PixTool product (PixTools/Scan, PixTools/View, or PixTools/EZ).

Production and personal scanners are growing in popularity as organizations/ users are integrating paper-based information into their applications more than ever before. With the proliferation of Intranet/Internet and fax communications, businesses need to be able to view information at their desktops in real time. Users are seeking new ways to enhance or clean-up document images and automatically capture embedded barcode information.

PixTools/Image Processing is a full featured toolkit that provides seventeen different operations which can be applied to black and white images. These operations, or filters, can be used to improve quality and readability, reduce file size, or both. These quality improvements can substantially enhance the accuracy of optical character recognition (OCR) and minimize the labor-intensive Reject and Repair operations in many applications.

Demonstration applications are available to evaluate this exciting technology on Windows and Windows NT.

The IP toolkit can be used alone, or in conjunction with another PixTools product such as PixTools/Scan, Pix- Tools/View, or PixTools/EZ to obtain a more complete development environment.

Fast

PixTools/Image Processing is optimized for rapid analysis and modification (where appropriate) of images while using a minimal amount of memory. That means image enhancement and barcode recognition can be added to applications for realtime processing with most desktop scanners at speeds of up to 400 pages per minute.

Easy to Use

PixTools/Image Processing comes with complete documentation, online help, numerous code samples in C, C++, and Visual Basic, and the full backing of Pixel Translations' renowned technical support department. Easy to Integrate PixTools/Image Processing components are provided in the form of ISIS drivers, which utilizes the well defined ISIS Pipes methodology for interconnecting imaging components with minimal code and effort. Existing ISISbased applications can typically add image processing functions in less than a day! Easy to Configure Each of the filters in PixTools/Image Processing comes with configuration dialogs, plus an API for complete programmatic control. The developer has complete control over how the capabilities are presented to the user, but can use standard dialogs where desired.

Comprehensive Set of Filters

PixTools/Image Processing provides twelve different operations which can be applied to an image, including:

• Deskew
• Border removal
• Margin cropping
• Halftone removal
• Noise removal
• Line removal
• Hole removal
• Smoothing
• Character erosion
• Character dilation
• Character skeletonization
• Barcode recognition
• Blank Page recognition
• Patchcode recognition
• Rotation
• Invert image
• Scale image to desired size

Filter Operation Overview

Deskew - This filter detects the skew of the page by analyzing the text or image elements and modifies the image to remove the skew. Parameters include minimum angle to detect, max angle to correct, and what percentage of the image to use in detection processing. Deskewed images typically compress better, resulting in smaller file sizes.

Border removal - The border removal filter crops black edges around the page which are produced by some scanners.

Margin cropping - The margin cropping filter crops the image edges down to a definable margin around the text or image on the page.

Halftone removal - The halftone filter removes halftone backgrounds from an image, while leaving the text alone. When processing a number of pages with halftones in the same area, the filter can be configured to operate on the specified area.

Noise removal - "Noise" is small black marks on the image that can come from fax transmission errors, dirty scanners or copiers, etc. This filter removes the noise from images to enhance recognition and reduce file size.

Line removal - The line removal filter finds and removes horizontal and vertical lines and repair overlapping text and graphics. This decreases the file size, and will often help OCR produce better results. The line removal filter can also be configured to reconstruct horizontal and vertical lines, "redrawing" them straight with smooth edges.

Hole removal - The hole removal filter is specifically designed to find and remove the marks left on an image when scanning a document which had binder holes punched in it (for example, three hole punching in the left margin).

Smoothing - The smoothing filter removes bumps and spurs on characters and fills in small holes. This process improves the appearance of the image, and helps reduce compressed file size without losing image detail.

Character erosion - This filter thins characters, reducing file size, but with a loss of image detail.

Character dilation - This filter grows characters, which can increase file size, but provide a better looking image, especially with dot matrix or very light printers.

Character skeletonization - This filter reduces characters to a one pixel thick skeleton. It should only be used in conjunction with certain OCR engines which operate well on this type of data.

Barcode recognition - The barcode filter detects and reads bar codes anywhere in the image. Up to 30 bar codes can be read on each page. The filter can be configured to "look" in a specific location in applications where the barcode location is predefined. The following types of bar codes can be recognized: Code 39, UPC-A, EAN-13, Code 28, UPC-E, ISBN Addon 2, Code 93, Codabar, ISBN Addon 5, Interleaved 2/5, EAN-8

Patchcode recognition - The patch code filter detects and reads patchcode types 2, 3, and T.

Blank Page recognition – The blank page recognition filter allows users 4 levels of blank page sensitivity, Pristine, Dirty White, Very Dirty White, or One Line Acceptable.

Image Rotation – This filter rotates image clockwise, 180 degrees, and counter clockwise. Enabling the Mirror option flips the image across its vertical axis.

Image Scaling – Using this filter resizes the image to a specified Width and Height.

Image Inversion – This filter inverts white pixels to black pixels and vice versa.