2 Matching Annotations
  1. May 2021
  2. Nov 2016
    1. 2. The Quality of the Scan Scanners make a digital representation of visual input. The quality and type of output it is able to produce, will have an effect on the quality of the input provided to the OCR engine to process. Software programs (like the Image Capture dialog in Paperless) make it possible for the user to alter the visual properties (such as brightness and contrast) of the scanner's output. One of the biggest factors is DPI or Dots per Inch. Setting the DPI lower than 200 will yield uninteligible results wheras setting it higher than 600dpi will just increase the size of the stored file without yielding much better results. We tend to recommend a 300dpi for in item. Color vs. Black and White, vs. grayscale depends on your source material. Often times documents and receipts will not need to be stored in Color (as it makes the storage file larger) but the difference between B&W and grayscale can be profound depending on the source. In general, a grayscale image will provide a better result than a black and white image. Internally, the software performs adaptive binarization where it changes a grayscale image to a black and white image if this process is already done, presumably by your scanner software) it's possible some fidelity will be lost and the OCR results may suffer.