Performance comparison of Engine combinations

CSDK supports multiple omnifont OCR engines for integration. This chart shows the speed/accuracy trade-off using selected Engine combinations and trade-off settings. Data was measured on a 3GHz Core Duo machine with 2GB RAM, with the following test set:

  • About 5000 pages in single-page TIFF files, with 1.13 million words

  • Mixed resolutions (200-300 dpi), both black-and-white and color

  • Mixed languages (English, French, German, Italian, Hungarian, Polish, Portuguese)

  • Mixed layout (magazines, business letters, faxes, business cards, and so on)

  • Mixed scanners


Speed and accuracy graph

For both speed and accuracy the 100% value is for the Plus2W configuration. That means the fastest configuration is slightly over twice as fast as the Plus2W configuration. Starting with CSDK version 21, the default configuration is Plus3W, that provides the highest accuracy. The percentage differences translate into word accuracy values as follows:

Most accurate (Plus 3W) 97.2%
Plus 2W 96.9%
Fastest (FRX + Legacy) 92.5%