Advanced OCR Recognition Settings Window - Elements Tab

Use this tab to specify advanced OCR recognition settings for elements such as tables, pictures, and so on.

Related Tabs

Text block

Use this setting to force recognition of text from left to right, and top to bottom. When selected, the recognized zone is considered as a text block, and the other settings on the Elements tab are unavailable and ignored.

One word per line

Use this setting to ensure the integrity of values in index fields. When selected, the recognized value is treated as a single word because spaces or unwanted characters are removed during the OCR process. For example, spaces would be removed from the value 1 2 3 4 5 6 and the results would be 123456.

Detect tables

This setting, which detects tables during the recognition process, is selected by default. If documents do not include tables, turning off Detect tables may improve recognition performance. When you turn off Detect tables, the following settings are not available:

  • Single line of text per cell: Assumes one line of text per each cell. Selecting this setting may improve recognition performance and/or accuracy. If a single cell in a table contains multiple lines of text, each line is recognized as a single cell. In some cases, the engine may not be able to split up cells by line of text, and the original organization of the cell is retained.

  • No hidden separators: Assumes no hidden separators in the table. The width of the cells in a text table is defined by the position of the separator between neighboring cells. If cells of the table are merged, this separator may not be removed, but it is hidden instead. Selecting this setting may improve recognition performance and/or accuracy.

  • Aggressive table detection mode: Assumes that a document contains many tables. Use this setting to ensure detection of every table on a page.

Detect pictures

This setting, which tells the recognition engine to detect pictures during the recognition process, is selected by default. If documents do not include pictures, turning off Detect pictures may improve recognition performance.

Detect bar codes

This setting, which tells the recognition engine to detect bar codes during the recognition process, is selected by default. If documents do not include bar codes, turning off Detect bar codes may improve recognition performance.