Standard OCR
Standard OCR reads characters by determining which of a set of possible characters is in each field, or character position, of a string. Once the tool has read the string, it returns the string along with additional data generated during the decoding operation.
When it decodes a string, the tool performs the following steps:
- After an image is acquired, it can use one or more filters to preprocess the region of interest.
- It locates the string and its individual fields.
- It compares the character in each field to each character in the font model, computes a score for each comparison, and selects the highest‑scoring character.
- For each field, it compares the best character score to the character acceptance threshold for Standard OCR and passes that character if its score is greater than or equal to the threshold. A field whose score falls below the character acceptance threshold is a failed field. and the character is not found.
- It computes the string score and builds the result string. It marks failed fields with either:
- A “*” to indicate that the score for the character is below the character acceptance threshold, but above the minimum character acceptance threshold
- A “?” to indicate that the score for the character is below the minimum character acceptance threshold
- If the checksum is SEMI, BC412, or IBM412, it computes the checksum.
- It reports the following results:
- Character scores
- Character locations
- Number of fields
- String score
- Result string