Scanned documents often contain undesirable elements, such as specks, dots, black borders, lines, and hole-punch marks. If your software application is designed to perform recognition on a scanned document, you will get far better results if you "clean-up" documents by detecting and removing these elements prior to the recognition process. Wouldn't it be great if you had a "hook" into the API decision-making process, allowing you to obtain information about every detected element, and could use the information to tailor functions to meet your needs?
Powerful Components are available to do just that.
Whether you need components to perform de-skew, dilation, erosion, or min and max filtering you will want to look for components that offer your application the power and flexibility it needs when working with scanned documents.