CN111510152A - Compression and decompression method for acoustic logging data - Google Patents

Abstract

The invention discloses a compression and decompression method of acoustic logging data, which comprises the following steps: carrying out matrix blocking on data acquired by the acoustic logging instrument according to the number of acquisition channels; respectively carrying out discrete cosine transform on each block matrix to obtain a DCT coefficient matrix; according to the characteristics of the sound wave signals, a proper quantization matrix is designed, and the DCT coefficient matrix is quantized; carrying out run length coding and Huffman coding on the quantized matrix; and obtaining and transmitting the compressed code stream. The compression method has higher compression ratio on the basis of improving the calculation performance of the underground main control device; the method can compress data in real time underground, greatly reduces the transmission quantity of the data, and improves the logging speed of the acoustic logging instrument.

Description

Compression and decompression method for acoustic logging data

Technical Field

The invention relates to the technical field of data compression, in particular to a method for compressing and decompressing acoustic logging data.

Background

With the development of logging technology and integrated circuits, higher requirements are put on the detection accuracy, speed and detection distance of logging instruments. Among various acoustic logging instruments, the remote detection multipole array acoustic logging instrument has the largest data volume and is limited by the communication speed of a data transmission system, and the remote detection multipole array acoustic logging instrument usually has a slower logging speed.

In the existing acoustic logging data compression technology, an ADPCM compression method is used, 16-bit data is input, the compression ratio is 4:1, 12-bit data is input, and the compression ratio is 2: 1; chinese patent application (CN201110408530) discloses a data compression method in a "real-time compression method of three-dimensional acoustic logging data", which splits original data into high and low bits to be compressed, then performs DCT to obtain mapping data, then retains more bits for data with large numerical value in the mapping data, retains less bits for data with small numerical value, and realizes data compression, but the method can only achieve 50% of compression ratio; chinese patent application (CN201610430622.7) discloses a data compression method in "a method for compressing and decompressing logging while drilling data in real time", which comprises designing a multi-resolution quantization model according to specific data, making a quantization table, calculating a difference between adjacent data, processing the difference by using the quantization table to obtain a quantization value and coordinates thereof, and finally transmitting the coordinates.

With the development of logging instruments, the compression ratio achieved by the above logging data compression method cannot meet the actual logging requirements.

Disclosure of Invention

The invention aims to overcome the technical defects, provides a compression method and a decompression method of acoustic logging data from actual requirements during underground operation, and improves the acoustic logging efficiency under the conditions that the transmission rate of a field cable is not changed and an internal circuit of an instrument is not changed.

In order to achieve the above object, the present invention provides a method for compressing acoustic logging data, the method comprising: carrying out matrix blocking on data acquired by the acoustic logging instrument according to the number of acquisition channels; respectively carrying out discrete cosine transform on each block matrix to obtain a DCT coefficient matrix; according to the characteristics of the sound wave signals, a proper quantization matrix is designed, and the DCT coefficient matrix is quantized; carrying out run length coding and Huffman coding on the quantized matrix; and obtaining and transmitting the compressed code stream.

As an improvement of the above method, the method specifically comprises the following steps:

step S1), carrying out analog-to-digital conversion on data collected by the acoustic logging instrument to obtain an M x N matrix;

step S2) dividing the M × N matrix into M/K × N/K block matrixes with the size of K × K based on the number K of the acquisition channels, and performing discrete cosine transform on each block matrix to obtain a DCT coefficient matrix;

step S3), a quantization matrix with the size of K x K is designed according to the characteristics of the acoustic logging data; quantizing each DCT coefficient matrix of step S2) with a quantization matrix; dividing the quantized matrix into a direct current coefficient and an alternating current coefficient;

step S4) using zigzag scanning for the quantized coefficients, rearranging K × K-1 ac coefficients into an integer array;

step S5) DPCM encoding the dc coefficient; carrying out run-length coding on the integer array formed by the alternating current coefficients in the step S4); finally, respectively carrying out two-dimensional Huffman codes on the direct current coefficient and the run-length coded integer array; and transmitting the code stream formed after encoding.

As an improvement of the above method, the specific step of performing discrete cosine transform on each block matrix is as follows: and performing two-dimensional discrete cosine transform on the block matrix from left to right and from top to bottom to obtain a discrete cosine transform coefficient matrix.

As an improvement of the above method, the specific step of quantizing each DCT coefficient matrix in step S2) by using a quantization matrix is: the quantization matrix by which the DCT coefficient matrix is divided is then rounded.

As an improvement of the above method, the specific steps of DPCM encoding the dc coefficient are: and subtracting the direct current coefficient of the last block matrix from the direct current coefficient of the current block matrix to obtain a difference value, and coding the difference value.

As an improvement of the above method, the specific step of run-length encoding the integer array composed of the ac coefficients in step S4) is: and representing the numerical values which continuously appear in the integer array by using numerical values of two bytes, wherein the first numerical value is the number of times of the repetition of the numerical value, the second numerical value is the numerical value per se, and when the rest numerical values are all 0, the special symbol is used for marking.

The invention also provides a method for decompressing acoustic logging data, which is used for decompressing the data stream based on the acoustic logging data compression method and comprises the following steps:

step T1) receiving the compressed data stream;

step T2) constructing a Huffman tree of the direct current difference value and a two-dimensional Huffman tree of the alternating current quantity respectively;

step T3) decoding the DC coefficient and the AC coefficient respectively by using a two-dimensional Huffman tree;

step T4) restoring the decoded data to a matrix of K × K size;

step T5) multiplying the matrix of the step T4) with the quantization matrix of the step S3) to obtain an inverse quantization matrix;

step T6) inverse discrete cosine transform is carried out on the matrix after inverse quantization to obtain an original block matrix;

step T7) synthesizes the M/K × N/K original patch matrices into a matrix of M × N.

Compared with the existing acoustic logging data compression technology, the invention has the advantages that:

1. the method can design different quantization matrixes according to the precision of different acquisition modules of different logging instruments, thereby greatly improving the flexibility of acoustic logging data compression;

2. the compression method has higher compression ratio on the basis of improving the calculation performance of the underground main control device; the method can compress data in real time underground, greatly reduces the transmission quantity of the data, and improves the logging speed of the acoustic logging instrument;

3. the compression and decompression method of the acoustic logging data has higher compression ratio, and improves logging speed and efficiency under the condition that a transmission system is not changed;

4. the compression ratio of the single-stage acoustic signal in the acoustic logging data compression method can reach 20: 1, even-level sound wave signals can reach a compression ratio of 10:1, real-time compression can be performed underground at a higher compression ratio, and logging efficiency is effectively improved;

5. the method is not limited to compression of acoustic logging data, and can also be used for data compression of data of other geological parameters.

Drawings

FIG. 1 is a general schematic diagram of a method of compressing and decompressing sonic logging data in accordance with the present invention;

FIG. 2 is a flow chart of a method of compression of sonic logging data in accordance with the present invention;

fig. 3 is a flow chart of the decompression method of the present invention.

Detailed Description

The present invention will be described in detail with reference to the accompanying drawings.

As shown in fig. 1, the present invention provides a method for compressing and decompressing acoustic logging data, which comprises the following steps:

step 101: during the logging process, the acquisition system first receives the source signal.

Step 102: and the acquisition system performs analog-to-digital conversion after receiving the information source signal to obtain signals of a plurality of channels, and the converted digital signals are original data to be coded.

Step 103: the main controller mainly realizes the compression process of data, and the specific steps are as follows:

(103-1) writing original data to be encoded in a matrix form;

(103-2) performing discrete cosine transform on the source matrix;

(103-3) quantizing the discrete cosine transformed matrix;

and (103-4) encoding the quantized matrix.

Step 104: and sending the data stream coded in the MCU into a transmission system for transmission.

Step 105: the upper computer receives the coded code stream to perform decoding operation, and the specific steps are as follows:

(105-1) decoding the code stream and writing the code stream into a matrix form with the same size as the original matrix;

(105-2) quantizing the decoded matrix;

(105-3) performing inverse discrete cosine transform on the quantized matrix;

(105-4) restoring the original data.

As shown in fig. 2, the invention provides a compression method of sonic logging data, which compresses in an MCU, and includes the following steps:

step 201: during field operation, data received by a remote detection acoustic logging instrument acquisition system is subjected to analog-to-digital conversion to obtain data of a plurality of channels, and the data of the plurality of channels are written into an M-N matrix form.

Step 202: the acquisition module of the remote detection acoustic logging instrument is 8 channels, the master controller blocks the acquired data according to the acquisition channels, and the matrix of M x N in the step 201 is divided into M/8 x N/8 matrices of 8 x 8.

Step 203: and respectively carrying out discrete cosine transform on the divided 8-by-8 matrixes to obtain discrete cosine transform coefficient matrixes, wherein low frequency is mainly concentrated at the upper left corner of the matrixes. Compared with direct DCT on the M-N matrix, the method reduces the operation amount on one hand, facilitates the following quantization and coding on the other hand, and can control the transmission error within one block without influencing the waveform of the whole data.

Step 204: and designing an appropriate 8 x 8 quantization matrix according to the characteristics of the source signal of the far detection acoustic wave instrument. When different acoustic logging instruments are used, different quantization matrixes need to be designed due to different signal characteristics so as to adapt to different requirements.

Step 205: the discrete cosine coefficient matrix obtained after the step 203 is divided by the quantization matrix designed in the step 204 for quantization, and the above operations are all lossless. The data of the coefficient matrix obtained by discrete cosine transform contains decimal, and the input data is integer when coding, so the quantized matrix needs to be subjected to rounding operation, and the decimal part is cut off, and the rounding operation is the lossy part in the compression method.

Step 206: the matrix quantized in step 205 is divided into a dc coefficient part and an ac coefficient part, and DPCM encoding is performed on the dc coefficient part, that is, the difference is obtained by subtracting the quantized dc coefficient of the previous block from the quantized dc coefficient of the current block, and the difference is encoded. The quantized dc coefficients are not coded directly because coding the difference values is advantageous to remove the correlation between the blocks.

Step 207: removing the dc coefficients from the matrix quantized in step 205, performing zigzag scanning on the remaining 63 ac coefficients, and rearranging the 63 ac coefficients into an integer array.

Step 208: the encoding is carried out, and the specific steps are as follows:

(208-1) run-length coding the ac coefficients;

and (208-2) performing Huffman coding on the direct current coefficient and the alternating current coefficient after run-length coding respectively.

Step 209: and sending the code stream formed after the coding (208-2) to a transmission system for transmission.

As shown in fig. 3, the present invention provides a method for decompressing a data stream compressed by the above compression method, wherein during field logging, an upper computer performs a decoding operation after receiving a code stream, and the steps are as follows:

step 301: and respectively constructing a direct current Huffman tree and an alternating current two-dimensional Huffman tree.

Step 302: and respectively decoding the direct current coefficient and the alternating current coefficient according to the Huffman tree constructed in the step 301, and recovering into an array of 64 coefficients after each decoding.

Step 303: the array recovered after decoding in step 302 is written in the form of an 8 x 8 matrix.

Step 304: according to the characteristics of the far detection acoustic logging data, the same quantization matrix as that used in encoding is designed so as to accurately recover the data.

Step 305: the 8 × 8 matrix generated in step 303 is multiplied by the quantization matrix in step 304 to perform inverse quantization.

Step 306: and performing inverse discrete cosine transform on the matrix subjected to inverse quantization in the step 305 to restore original data.

Step 307: writing the recovered data in matrix form of M × N according to the channels.

Step 308: and outputting the recovery value, and ending decompression.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (7)

1. A method of compression of sonic logging data, the method comprising: carrying out matrix blocking on data acquired by the acoustic logging instrument according to the number of acquisition channels; respectively carrying out discrete cosine transform on each block matrix to obtain a DCT coefficient matrix; according to the characteristics of the sound wave signals, a proper quantization matrix is designed, and the DCT coefficient matrix is quantized; carrying out run length coding and Huffman coding on the quantized matrix; and obtaining and transmitting the compressed code stream.

2. The method of compressing sonic logging data of claim 1, the method comprising in particular the steps of:

step S1), carrying out analog-to-digital conversion on data collected by the acoustic logging instrument to obtain an M x N matrix;

step S2) dividing the M × N matrix into M/K × N/K block matrixes with the size of K × K based on the number K of the acquisition channels, and performing discrete cosine transform on each block matrix to obtain a DCT coefficient matrix;

step S3), a quantization matrix with the size of K x K is designed according to the characteristics of the acoustic logging data; quantizing each DCT coefficient matrix of step S2) with a quantization matrix; dividing the quantized matrix into a direct current coefficient and an alternating current coefficient;

step S4) using zigzag scanning for the quantized coefficients, rearranging K × K-1 ac coefficients into an integer array;

step S5) DPCM encoding the dc coefficient; carrying out run-length coding on the integer array formed by the alternating current coefficients in the step S4); finally, respectively carrying out two-dimensional Huffman codes on the direct current coefficient and the run-length coded integer array; and transmitting the code stream formed after encoding.

3. The method of claim 2, wherein the discrete cosine transforming each block matrix comprises: and performing two-dimensional discrete cosine transform on the block matrix from left to right and from top to bottom to obtain a discrete cosine transform coefficient matrix.

4. The method for compressing sonic logging data of claim 3, wherein the step of quantizing each DCT coefficient matrix of step S2) with a quantization matrix comprises: the quantization matrix by which the DCT coefficient matrix is divided is then rounded.

5. The method of claim 4, wherein the DPCM encoding of the dc coefficients comprises: and subtracting the direct current coefficient of the last block matrix from the direct current coefficient of the current block matrix to obtain a difference value, and coding the difference value.

6. The method for compressing sonic logging data of claim 5, wherein the step of run-length encoding the array of integers consisting of the ac coefficients of step S4) comprises the steps of: and representing the numerical values which continuously appear in the integer array by using numerical values of two bytes, wherein the first numerical value is the number of times of the repetition of the numerical value, the second numerical value is the numerical value per se, and when the rest numerical values are all 0, the special symbol is used for marking.

7. A method for decompressing sonic logging data for a data stream using the sonic logging data compression-based method of any of claims 1-6, comprising the steps of:

step T1) receiving the compressed data stream;

step T2) constructing a Huffman tree of the direct current difference value and a two-dimensional Huffman tree of the alternating current quantity respectively;

step T3) decoding the DC coefficient and the AC coefficient respectively by using a two-dimensional Huffman tree;

step T4) restoring the decoded data to a matrix of K × K size;

step T5) multiplying the matrix of the step T4) with the quantization matrix of the step S3) to obtain an inverse quantization matrix;

step T6) inverse discrete cosine transform is carried out on the matrix after inverse quantization to obtain an original block matrix;

step T7) synthesizes the M/K × N/K original patch matrices into a matrix of M × N.

Images