Maximize your document processing rates with our Document Scanner SDK’s new image filters

app store

We’re proud to announce that we’ve significantly improved the image filter options of our Document Scanner SDK. Built using computer vision algorithms, these enhanced filters increase the processing speed on the device and help make scanned documents even more human- and machine-readable. On average, applying a filter is now 50% faster.

Low-contrast documents (e.g., with colored paper or gray text) and images captured under unfavorable conditions (such as dark or very bright environments) now pose even less of a problem to your backend software.

For most use cases, the following two filters are a good starting point.  

The Document Scanner SDK’s color filter is well-suited for documents containing more than just text. It brightens up the image and enhances its contrast.

Conversely, our ScanbotBinarization filter is a good fit for most document scanning use cases involving text on paper. It reduces the image’s color depth to black-and-white and makes the text more machine-readable, thus optimizing it for OCR.

To further tailor the binarization filter to your needs, you can use one of four presets. These presets increase performance in specific areas (e.g., processing speed, shadow handling). Try them all to see which one best fits your use case.

Example of our new image filters in action
left: original, right: binarized with anti-aliasing

How to use the image filters effectively

You can choose whether to apply filters to scanned documents automatically or let users decide for themselves. The filters can be applied both to images captured by the SDK and to those imported from the device’s gallery.

Depending on whether the binarized documents will be processed by employees or by your backend software, there are matching additional settings to further enhance readability:

  • Anti-aliasing smooths out jagged edges in text characters. This gives them a rounder look and makes them more pleasing to the human eye.
  • Denoising removes tiny image artifacts from binarized scans that can impede OCR. Since too much denoising can make parts of the text disappear, you can configure the level to fit your backend requirements.

The Scanbot SDK team will gladly help you find the best filter configuration for your use case. For example, our customers in the transportation industry typically choose a preset configured for scanning colored documents with gray text, since this is a common combination in the sector.

How you can get started using our new filters

The new image filters are part of the new Document Scanner SDK v.5.0.0 for Android and iOS. After updating your SDK version, you can immediately start applying the filters to both scanned and imported documents. For more information, please refer to our documentation or reach out to the Customer Success team via csm@scanbot.io.

Developers, ready to get started?

Adding our free trial to your app is easy. Download the Scanbot SDK now and discover the power of mobile data capture