Cognitive Solutions B780, A776 manual Configure Binarization Methods and Attributes

134 Chapter 5: Programming commands

r specifies the number of saved gray levels = scale resolution (default value = 0) as follows:

Value	Definition
0	No change from previous setting
1	Minimal noise cleaning preserving 128 gray levels
2	Low Standard noise cleaning preserving 64 gray levels
3	High Standard noise cleaning preserving 32 gray levels
4	Maximum noise cleaning preserving 16 gray levels
5	No noise cleaning and full 256 gray level as captured is preserved, default

Return: None

Description: The parameters set the image’s internal orientation and number of scanned gray levels. The internal format of the image is always 256 level gray scale, one byte per pixel; when sending to the host a reformatting (if not raw) is then done when needed. Normally 16 – 32 gray levels are suggested for viewing on displays, thus allowing more effective LZW compression while preserving distinguishable artifacts of a gray scale image. A TIFF structure is used to return all formats. This structure holds height & width and other document identifying tags as well as images of both sides.

The newest configured values are retained in non-volatile memory, persisting across power cycles.

Configure Binarization Methods and Attributes

Hexadecimal: 1D C5 nL nH m d1 – dn-1

Parameters: n is the number of bytes that follow. n values of 1 indicate a choice of threshold method only without changing methods properties. m selects the threshold method and optional attributes that the application may choose to set: m= 0 reserved for future; a dynamic function threshold method based on experience lookup tables; m = 1 (default) selects the US banking average area threshold method, m = 2 selects fixed value thresholding, m >2 reserved for future other method(s) that may be standardized by the industry. Any dx value = 0 sets the attribute at its default value. Any dx value not sent will remain unchanged.

Method	M	d1	d2	d3
Dynamic (Future)	0	Black bias cross point, 5	NA	NA
		<=d1<=95; default =50
Averaging	1	% difference from average,	Reflectance below which pix is	Reflectance equal or above
		5<=%<=95; default = 20%	always black; default = 51	which pix is always white;
				default = 178
Fixed Threshold	2	The reflectance value to	NA	NA
		be used, 1 <= d1 <= 254,
		default = 96.

For m = 0, the bias cross point selects where favoring black pixels changes over to favoring white pixel.

The averaging method uses 20% as the default difference for a pixel to be judged different than an 1/8” square background average. Absolute values are also used to decide on a pixel’s color (either black or white).

The fixed threshold method uses a simple comparison test against the value.

Return: None

Description: This allows for selection of threshold method and tweaking of any threshold method variables. These are used when TIFF G4 binary format for the transferred image is selected.

The newest configured values are retained in non-volatile memory (persist across power cycles).

A776 (B780) Programming Guide A776-PG00001 C 12/09