Brute stack from original seismic data

(Courtesy of Amoco Production Co.)

Click here to close the image frame
Geophysical - Original
Original
Click here to close the image frame
Geophysical - Compression - 40:1
40:1 Compression
Click here to close the image frame
Geophysical - Compression - 60:1
2048:1 Compression
Click here to close the image frame
Geophysical - Compression - 100:1
100:1 Compression
Click here to close the image frame
Geophysical - Residual - 40:1
40:1 Residual
Click here to close the image frame
Geophysical - Residual - 60:1
60:1 Residual
Click here to close the image frame
Geophysical - Residual - 100:1
100:1 Residual
 

Legend
 40:1
Brute stack from compression - decompression of prestack seismic data at a compression ratio of 40:1.

Mean Squared Error (MSE) = 34 db

Residual stack equal to the difference of the original stack and the one resulting after compression-decompression at 40:1
 60:1
Brute stack from compression - decompression of prestack seismic data at a compression ratio of 60:1.

Mean Squared Error (MSE) = 31 db

Residual stack equal to the difference of the original stack and the one resulting after compression - decompression at 60:1
 100:1
Brute stack from compression - decompression of prestack seismic data at a compression ratio of 100:1.

Mean Squared Error (MSE) = 25 db

Residual stack equal to the difference of the original stack and the one resulting after compression - decompression at 100:1

Paper onWavelet transform coding image compression

Wavelet transform coding image compression is applied to seismic data. The filter lengths, depth of decompression and quantization methods are varied throughout the compression experiment to decide which is the best parameter combination. Based on the compression results, it was decided that long filters, moderate decomposition depths and uniform quantization yield the best results.