Home > Products > SmartReco-C > Classifiers | Test Version | Licences & Prices



What is SmartReco-C?


SmartReco-C ...
 > is a high-quality, flexible and fast ICR/OCR software for the recognition of single handwritten or machine-printed characters or symbols in binary images.
 > does not have to be painstakingly trained by the user, but has powerful classifiers which are developed and configured by RecoScript.
 > is not an end user product, but a subprogram library for software developers.
 > requires the presence of segmented single characters or symbols.
 > can be used as a central basis component of a line reader or as an additional recognizer in a voting system.
 > is based on complex mathematic transformations and provides a very high standard of recognition reliability.
 > has a C-API and an ANSI-C core.
 > features three functions which can be very easily integrated, and the use of which resembles file handling with fopen, fread and fclose:
  src_open   - opens a character or symbol classifier,
  src_read    - reads a character or symbol with the aid of the classifier,
  src_close   - closes the classifier and frees the corresponding memory.
 > was designed platform independent and is now available as a static or dynamic library for the X86 platforms Windows 2K/XP/Vista/7/8, Linux and Solaris, whereas the corresponding C/C++ compilers of Microsoft, Borland, GNU, Watcom and Sun are supported.
  Porting for other platforms - also embedded systems without file I/O - should be problem-free due to the universal software concept, in which float arithmetic was fully excluded.
 > can also be used by VisualBasic, Delphi, Java and .NET programs with the corresponding wrapper.
 > has been optimized system-specific in spite of the principle platform independence. On all X86 platforms time-critical C-loops have been replaced by a highly optimized assembler code.



The following diagram illustrates the call of the function src_read in a typical SmartReco-C application (line reader):

The handwritten word "WORLD" contained in a binary image is to be read.

First src_open is used to open the required classifier (e.g. handprint alphanumeric USA). All single characters are then extracted by the application program and processed with src_read, whereas one character and the corresponding classifier handle must be passed per call. src_read generates a result data record for each character, which among others contains a list of possible unicodes with corresponding quality values (probabilities). Using its context knowledge and the quality values from the result data records of the single characters, the program then forms the required word "WORLD".




Are you interested in SmartReco-C but cannot find "your" operating system or "your" compiler in the list above?
Tell us your development platform and we are sure to find a solution!