The point transform is the number of least significant bits truncated from each input sample prior to compression. A point transform is another way to use the JPEG-LS algorithm in a near-lossless mode.

Non-zero values denote near-lossless compression; the source image will no longer be reconstructed exactly from the compressed image. The allowed range for Point is zero (default value) or 2 to 16.

• The JPEG-LS standard allows input samples to range from 2 to 16 bits (called the Precision), hence the allowed range for Point.

Precision is the number of bits used for each sample in an image. For example, a 24-bit image uses 8 bits for each red, green and blue sample in each pixel.

JLSPoint should be no greater than the actual Precision of the input samples. This affects the bits per channel of an image. So if you had a 24-bit RGB image, it has 8-bits per R, G, and B channel. For example, if Point = 2 then the image will effectively have 6-bits per channel (8 - 2 = 6). This will affect the calculations for JLSMaxValue property and therefore the maximum value for JLSNear property.

Note: Available in Professional Edition.

ImagXpress Object  | ImagXpress Members  | Saving