Wavelet Based Image Compression Results

Here are the reconstructed outputs for our Wavelet Based Image Compression Algorithm. Error measures for different bit rates & wavelets are given in the table below.
 
 
Type
Zeros at z=-1
Rc/Entropy/LZ coded bits per pixel (bpp)
PSNR
dB
Comments
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
2.0/0.9/1.25
(x 6.4)
38.2
Perceptually 
Loss-less
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
1.0/0.45/0.67
(x 11.9)
35.5 
Perceptually 
Loss-less
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
0.5/0.23/0.326
(x 24.5)
32.5 
Loss just apparent
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
0.25/0.126/0.202
(x 39.7)
30.15
Lossy
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
0.125/0.062/0.089
(x 90)
27.4
Lossy
P(z) = 15 tap Max-Flat.
Factorization: 9/7
8
0.08/0.042/0.067
(x 119)
26.0
Lossy
P(z) = 15 tap windowed FIR.
Factorization: 9/7
0
0.5/0.266/-
15.2
Simple subband decomposition with short filters
P(z) = 15 tap Max-Flat.
Factorization: 7/9
8
0.5/0.236/-
31.0
Lossy
P(z) = 15 tap Max-Flat.
Factorization: 6/10
8
0.5/0.222/-
30.5
Lossy
P(z) = 24 tap Max-Flat.
Factorization: 7/17
12
0.5/0.234/-
32.2
Loss just apparent