Table 1.
Comparison of existing machine vision system for chemical structure recognition. O and X denotes the availability of key features listed in the first column: O = Positive and X = Negative.
| Kekule | IBM OROCS | CLiDE | OSRA | chemOCR | ChemReader | |
| Written language | C++ | C | C++ | C++ | Java | C++ |
| Running Platform | MS windows | IBM OS/2 | MS windows | Linux/MS Windows/OS X | Independent | MS Windows |
| Batch mode | X | X | O | O | O | O |
| Bond streo | O | X | O | O | O | O |
| Abbreviation Interpretation | O | X | O | O (limited)1 | O | O |
| Chemical Knowledge | O | X | X | X | O | O |
| Document analysis | O | X | O | X | X | Under development |
| Automatic Extraction2 | X | O | O | X | X | Under development |
| Open source | X | X | X | O | X | X |
| Customizable extensibility | X | X | X | O | O | O |
1 OSRA has a hardcoded matching table to interpret only a few chemical abbreviations.
2 A functionality for extracting digitized images of chemical structure diagrams from scanned pages