Ability: Our OCR Omnifont Engine is independent of font type and size to comprehend and read almost every text printed. It reads:
Uppercase letters
Lowercase letters
Accented letters
Symbols
Punctuation
Digits
It can recognize words, characters, text lines, paragraphs, or pages that are full. It is equipped to divide and comprehend glued characters to paste and comprehend broken characters, also.
Accuracy: that the OCR engine precision is 100 percent once the print and picture quality is great. Even if the picture quality is moderate, the error rate is normally meager.
Technology: Our OCR recognizer relies on attributes extraction using multi-agent technologies: for every symbol to comprehend it's been categorized a representative trained to extract the characteristics of the particular personality, able to document whether the input image is just like the trained personality along with also the similarity grade (assurance ).
Rate: 200 / 600 CPS, based upon CPU speed. Using Pentium IV chips, the operation is generally over 600 CPS.
Input signal: Color, gray-scale o bitonal images using a 200 DPI or higher resolution may be utilized as input. Better performances are accessed using a gray-scale picture with 300 DPI resolution because the OCR engine functions natively in gray-scale. We propose using our pre-processing SDKs such as deskew, despeckle, and black edge removal to achieve the best results.
Output: the output comprises the ASCII or UNICODE code, the assurance level, the position, and the dimensions on the picture for every recognized character. Exactly the very same information are also available for all of the recognition options for every personality.
Platform: the ocr support all Windows platforms.
Packaging: the ocr motor is dispersed as a single 32 bit DLL (roughly a 1 MB size).
