ADC - .NET Despeckle, Deskew, Auto Crop, Clean Document Images with .NET
Document and Image Cleanup Algorithms
DotImage Advanced Document Cleanup (ADC) is included with DotImage Document Imaging, providing document cleanup algorithms that can be applied to scanned documents to clean them up for better compression and archival, increased readability, and pre-OCR processing.
DotImage ADC uses proprietary algorithms developed by our research and development team that are designed to automatically select the best parameters to apply for fast and accurate processing. The commands included in ADC extend the command interfaces in DotImage and can easily be applied to an existing application built with DotImage. See the online Document Cleanup Demo that demonstrates ADC with our AJAX-enabled Web Image Viewer.
The following commands are available and most can be applied onto the entire image, or a region of interest:
Uses a choice of three thresholding methods (adaptive, global, or simple) to efficiently and accurately convert a color or grayscale image to binary. Binarize often results in better bi-tonal documents than scanning directly to black and white.
Advanced Border Removal
Automatically removes black borders from a bi-tonal image using a sophisticated object detection algorithm. Scanning documents often results in an unnecessary border around the image.
Removes margins (whitespace) from each side of an image. Useful for decreasing image size, and isolating relevant features on an image for processing.
Auto Border Crop
Similar to Advanced Border Removal, but crops the border as opposed to removing the feature. Uses a higher speed algorithm than Advanced Border Removal.
Removes specks of a specified size from a bi-tonal image. Useful for removing noise from an image introduced by a scanner or when thresholding from a color or grayscale image.
Hole Punch Removal
Detects and removes hole punch artifacts from a bi-tonal image. Will automatically remove round hole punches from bi-tonal images from any of the 4 sides that appear when these documents are scanned.
Removes arbitrary blobs of a specified size from a bi-tonal image.
Blank Page Detection
Automatically detects if a bi-tonal document is blank using a sophisticated object detection algorithm. Useful when scanning duplex (both sides) when the back side of the document is blank.
Automatic Page Inversion
Automatically detects and optionally inverts an image when the background is black and the foreground is white. Useful when a scanner mistakenly inverts image data, or if the image was saved with incorrect tags resulting in inverted data.
Automatic Text Inversion
Detects inverted text regions in a bi-tonal image and automatically inverts them so that OCR engines can recognize the text.
Automatically removes horizontal or vertical lines from a bi-tonal image and reconnects broken characters. Useful for pre-processing a form prior to OCR.
Detects and removes shaded regions in a bitonal image, created when thresholding color or gray images to bitonal with shaded image areas. Often referred to as "Dot Shade Removal".
Detects skew in a bi-tonal document and corrects the image by rotating it. Useful for deskewing scanned documents. Note: this command is included in DotImage Document Imaging without the need to purchase ADC.
All of the commands in the Advanced Document Cleanup module can be used in the same basic manner:
BinarizeCommand myBinarizeCommand = new BinarizeCommand();
ImageResults results = myBinarizeCommand.Apply(myOriginalImage);
AtalaImage myNewImage = results.Image;
The commands work best on images that have a 1-bpp pixel format. See these articles for more ways to use the ADC commands:
Deskewing a Color Image
How to Apply the AutoBorderCropCommand to a non 1-bpp image
Creating a Custom ImageCommand
Cleaning Document Images on the Web with AJAX
"By using Atalasoft's viewer, we are able to provide viewing and annotating without the need for the client to install a plug-in and elevate their security. This is critical for us as our customers are often using different computers and usually don't have IT resources to manage this for them. With Atalasoft's technology our clients get the key functionality they need to perform their fax management tasks. They can digitally sign, annotate, and fill out forms without the need to download, print, and fax back. They can do this from any computer with internet access. Without Atalasoft, we could not provide a solution that would be acceptable to our customers."
- Dennis Nasto, CEO, SecureCare Technologies
"The DotTwain component allowed us to easily create an interface for our users to scan their documents. With the DotImage component library, the easy to use ThumbnailViewer and ImageViewer controls allowed us to create an intuitive display for displaying our digital images. Finally, with the Dotimage annotation component, we could provide our users an easy way to annotate their images. Most exciting to us was the launch of Atalasoft's annotations which support Digital Ink and is perfectly suited for the Tablet PC environment. The Atalasoft DotImage Document Imaging suite was truly a complete imaging solution for our needs."
- Tim Cummings, Development Manager, MDTablet, LLC
"Our solution integrates with an existing legacy database to retrieve patient information and information concerning the patient's activities. We then relate this information to scanned documents. By using the combination of DotImage and DotTwain, we were able to immediately mark off 40% of our functional requirements. The open API architecture of DotImage and DotTwain made the integration and customization a straight-forward intuitive process."
- Todd Oliver, Project Lead, LinkedHealth
"The radiologist can read an image anywhere they have web access... a hospital APCS system, a clinic, at home, and even via PDA. There is nothing the doctor needs to load."
- Scott Shuppert, President, Cad/Cam Services, Inc.
"DotImage is simply the best .NET imaging toolkit on the market. We have used several competing products but none of them are built around the .NET architecture like DotImage - so they are harder to use and more complex to distribute."
- David Cardinal, CEO, ProShooters LLC
"Working with Atalasoft has been 'awesome'. The Atalasoft support team was helpful in getting us familiar with the product. DotImage speaks for itself - the organization, the API, and level of abstration. Kudos to Atalasoft."
- Joe Aparo, Senior Developer, EBSCO
"Using Atalasoft's DotImage toolkit, we are offering a single yet scalable and flexible imaging solution to our customers, enabling them to utilize Image Connect across multiple applications as their needs change."
- Brad Kahnamoui, Product Manager, Thomson Elite
"The state is saving thousands of dollars per year and serving its constituents faster than ever. Our software maintenance cost went from about $7,500 annually to less than $300"
(After working with our customer Envoc)
- Marty Beasley, Manager of Geographic Information Systems, Louisiana Office of State Lands
"The Graduate School Office feels very strongly that this system has gone beyond being merely useful and has entered the realm of being necessary. It has allowed us a safe and long-lasting method for archiving student and other records which may need to be accessed in a timely manner. Our ability to view these imaged documents quickly has saved much in terms of valuable staff time which would otherwise be spent going through paper documents in storage cabinets."
- Sandra M. Powers - Dean of Undergraduate Studies, The College of Charleston