I am reading ISO/IEC 14495-1 : 1999 (E). Under section C. JPEG-LS preset coding parameters, one can find the valid values for T1, T2, T3 and RESET. I am interested in selecting "optimized" T1,T2,T3 value for lossless compression (NEAR=0).

The table Table C.1 – LSE marker segment parameter sizes and values for JPEG-LS preset coding parameters simply gives lower and upper bounds:

Parameter Values
T1 0, or NEAR + 1 ≤ T1 ≤ MAXVAL
T2 0, or T1 ≤ T2 ≤ MAXVAL
T3 0, or T2 ≤ T3 ≤ MAXVAL
RESET 0, or 3 ≤ RESET ≤ max(255, MAXVAL)

I fail to understand how those presets actually works/drive the compression process. How can pick T1, T2 or T3 to achieve a better compression ratios? If that simplify the problem: MAXVAL=65535 for me, and since I am only interested in lossless compression: NEAR=0.




This site is temporarily in read-only mode and not accepting new answers.

Browse other questions tagged .