CN110392260B

CN110392260B - Low-cost joint image compression transmission method

Info

Publication number: CN110392260B
Application number: CN201910630136.3A
Authority: CN
Inventors: 周诠; 张茗茗; 刘睿华; 刘娟妮; 回征; 呼延烺; 张怡; 王鸣涛; 蒙红英; 魏佳圆
Original assignee: Xian Institute of Space Radio Technology
Current assignee: Xian Institute of Space Radio Technology
Priority date: 2019-07-12
Filing date: 2019-07-12
Publication date: 2021-04-13
Anticipated expiration: 2039-07-12
Also published as: CN110392260A

Abstract

The invention provides a low-cost combined image compression and transmission method, which comprises the steps of preprocessing an original image, compressing the preprocessed image, embedding information, and further compressing the combined image in an interested area. The compression ratio is improved by the combined processing of block truncation and compression data; by embedding the interested region data, the image quality of the key region is ensured. The invention can reach 16 times compression ratio through preprocessing and compression hiding method, and the image PSNR is more than 30 dB. The invention can reach compression ratio of more than 16 times and even reach 200 times and still maintain certain image quality by a preprocessing and combined compression method. The method is far less complex than the standard data compression method, keeps better compression quality and provides technical support for the application of low-cost and large-compression image transmission.

Description

Low-cost joint image compression transmission method

Technical Field

The invention relates to a data transmission method, in particular to a low-cost joint image compression transmission method, and belongs to the field of communication (such as data communication technology and the like).

Background

Image processing and data transmission are currently important research topics in the field of communications. With the development of science and technology, people have greater and greater requirements on high-resolution images, and data compression is imperative.

Data compression is divided into a lossy compression method and a lossless compression method, the lossy compression method is high in compression ratio, but certain information loss exists between a compressed recovery image and an original image, and the application is not greatly influenced in practice as long as subjective and objective evaluation indexes meet requirements. The lossless compression method has no information loss between the recovered image and the original image after compression, but the compression ratio is particularly small, generally about 2 times, the use occasion is limited, and data transmission is inconvenient. The compression method adopted by the general high-speed data transmission system is mostly a lossy method, and for images, the peak signal-to-noise ratio (PSNR) is generally more than 30dB, and the effect is more than 35 dB.

There are many image data compression methods, representative of which are JPEG and JPEG 2000. The compression method is complex and difficult to realize by hardware, and the JPEG2000 compression algorithm is realized mostly based on foreign compression chips at present, has high development cost and is limited by people; or some satellite tasks do not need high image quality, such as satellite extravehicular monitoring, earth observation general survey, small satellites, micro-nano satellite image transmission, technical test satellite image transmission, space station internal monitoring, ground monitoring systems and the like, and the adoption of an overseas JPEG compression chip is not needed.

The Block Truncation Coding (BTC) compression method is a classic low-cost compression method, the typical compression ratio is 4 times, the peak signal-to-noise ratio (PSNR) is more than 30dB, and the method is very suitable for some low-cost satellite application occasions, and if the compression ratio can be increased to 16 times or even higher or the compression performance is improved when the compression ratio is unchanged, the method is more suitable for the occasions such as micro-nano satellites and the like requiring simple software and hardware realization, high compression ratio and meeting the requirements on performance.

Disclosure of Invention

The technical problem solved by the invention is as follows: the method overcomes the defects of the prior art, provides a low-complexity image joint compression method, and improves the compression ratio of block coding through image pre-compression, subsequent further compression and region-of-interest hiding compression. The invention improves the compression ratio through the image preprocessing and the further compression coding method of the block compression at the low complexity of the block compression coding.

The technical scheme of the invention is as follows: a low-cost joint image compression transmission method comprises the following steps:

(1) setting the size of an original image A0 as M × N, quantizing 8 bits, and firstly smoothing the original image A0 to obtain a smoothed image A1;

(2) performing scaling processing on the image A1 at a compression ratio of R0 to form a sub-image A2; dividing the sub-image a2 into non-overlapping blocks of size K1 × K2, each block yielding X1, X2, … Xk gray-values, K1 × K2;

(3) AMBTC compression is carried out on the sub-image A2, three groups of numbers L, H and B of each non-overlapped block after compression are continuously compressed to obtain compressed data, decompression is carried out to obtain a restored image A3 of the sub-image A2, and the restored image A3 is amplified; wherein, L is the average value of the gray scale smaller than the average gray scale value of the block in each block, H is the average value of the gray scale larger than or equal to the average gray scale value of the block in each block, B is 0 or 1, K1 × K2 bit values are total, 0 corresponds to the gray scale smaller than the average gray scale value of the block, and 1 corresponds to the gray scale larger than or equal to the average gray scale value of the block.

(4) The restored original image a4 is obtained.

The method for continuously performing variable compression ratio transmission on the compressed three groups of numbers L, H and B in the step (3) is as follows:

31) decomposing B of the three sets of numbers of each block into C1, C2, wherein C1 is q bits of 1,3,5,7,9,11,13,15 … of B, C2 is q bits of 2,4,6,8,10,12,14,16 of B, and q is K1K 2/2;

32) comparing L and H, C1 with C2, d1 and d2 as threshold values

When H-L is not less than d1, if the bit number of C1 is not more than d2 and the bit number of C2 is not more than d 6335, the lowest 2 position of L is 00, and L is reserved; h is not transmitted;

when H-L is less than or equal to d1, if the number of bits of C1 is more than d2 and the number of bits of C2 is not less than d2, the lowest 2 position of L is 01,

l is reserved; h is not transmitted;

when H-L > d1, if the number of bits of C1 is not more than d2 and the number of bits of C2 is not more than d2, the lowest 2 position of L is 10,

l reserved, H reserved, C1 reserved; c2 does not transmit;

when H-L > d1, if the number of bits of C1 is different from that of C2 > d2, the lowest 2 position of L is 11;

l reserved, H reserved, C1 reserved, C2 reserved;

transmitting the data reserved after compression;

33) decompressing the retained values obtained in step 32) to restore image a3

If the data received by each non-overlapping block is X, Y, Z1, Z2, then

When the lowest 2 bits of X are 00 or 01,

restoring all the gray values of the non-overlapped blocks to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

obtaining L, H, B, restoring the gray value of the non-overlapped block through decoding:

if a certain bit in the block B is 0, the gray value corresponding to the block is L, and if a certain bit in the block B is 1, the gray value corresponding to the block is H;

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

b is obtained according to C1 and C2, and the gray value of the non-overlapped block is restored through decoding according to L, H and B:

the image A3 is restored and enlarged to a1 size based on the restored grayscale value for each non-overlapping block.

The method for transmitting the compressed three groups of numbers L, H and B with the fixed compression ratio in the step (3) is as follows:

31) decomposing three groups of numbers L, H and B of each block into L, H, C1 and C2, wherein C1 is q bits of B in 1,3,5,7,9,11,13 and 15 …, C2 is q bits of B in 2,4,6,8,10,12,14 and 16, and q is K1K 2/2;

processing the region-of-interest image F1 of the original image, and combining the position information of F1 and the processed data into data S, wherein the byte stream of S is S1, S2, S3, … … and Sn;

32) hiding the compressed data byte stream: comparing L and H, C1 with C2, d1 and d2 as threshold values

Firstly, comparing L and H, C1 of a first block with C2, wherein d1 and d2 are threshold values;

when H-L is less than or equal to d1, if the number of bits of C1 is not equal to d2 as compared with that of C2,

l lowest 2 position is 00, L is reserved; h was replaced by S1, C1 by S2 and C2 by S3;

when H-L is less than or equal to d1, if the bit numbers of C1 and C2 are different and are greater than d2,

the lowest 2 position of L is 01,

l is reserved; h was replaced by S1, C1 by S2 and C2 by S3;

when H-L > d1, if the number of bits of C1 is not equal to or less than d2 and C2 is not equal to or less than d2,

the L-lowest 2 position is 10,

l was left unchanged, H was unchanged, C1 was unchanged, C2 was replaced with S1;

when H-L > d1, if C1 is not the same as C2 in terms of number of bits > d2,

the L-lowest 2 position is 11,

l is reserved, H is unchanged, C1 is unchanged, and C2 is unchanged;

sequentially comparing L and H, C1 of the subsequent blocks with C2, wherein d1 and d2 are threshold values; and hiding the byte streams which are not hidden in the data S in sequence according to the same method.

Transmitting the data reserved after compression;

33) decompressing the retained values to recover image a3

Assuming that the data received by each block K1X K2 are X, Y, Z1 and Z2, then

When X is at the 2-lowest position of 00, L ═ X, H ═ X, a3 ═ X;

y for the recovered byte stream S1, Z1 for S2, Z2 for S3,

restoring all the gray values of the block to be X;

when the 2-lowest position of X is 01, L ═ X, H ═ X, A3 ═ X;

recovered byte stream S1 ═ Y, S2 ═ Z1, S3 ═ Z2

Restoring all the gray values of the block to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

recovering byte stream S1 ═ Z2;

if a bit in a certain block B is 0, the corresponding gray value of the block is L, and if a bit in a certain block B is 1, the corresponding gray value of the block is H;

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

no recovery byte stream;

restoring the image A3 according to the restored value of each block, and enlarging to a size of a 1;

then, the region-of-interest image F1 is restored from the byte stream S, and the corresponding region image of the enlarged image is replaced with F1.

Compared with the prior art, the invention has the beneficial effects that:

(1) the method firstly carries out smooth preprocessing on the original carrier image, the information quantity of each block is uniform, and the subsequent hiding processing is convenient for the image.

(2) A new method for further compressing AMBTC block compressed data is provided, 4 different states are coded skillfully, the data volume is reduced and the compression ratio is improved through the combined setting of thresholds d1 and d 2. Meanwhile, the uniqueness of information extraction is ensured through encoding, and a basis is provided for the quality of the compression method.

After AMBTC compression, three sets of values are obtained (L, H, B), L and H may pass only 1 value when L and H are relatively close; the two values C1, C2 formed by 16 bits B after the compression parameter are also comparable, when C1 and C2 are close, C1 and C2 only pass 1 value.

(3) The novel method for hiding the region of interest in the AMBTC compressed region is provided, information hiding and embedding are skillfully carried out aiming at 4 different states, the data volume is not increased, the uniqueness of information extraction is ensured, and a basis is provided for improving the image quality of the region of interest in the compression method.

After AMBTC compression, three sets of values are obtained (L, H, B), L and H may pass only 1 value when L and H are relatively close; two values C1 and C2 formed by 16 bits B after the compression parameter are compared, and when C1 and C2 are close, C1 and C2 can only transmit 1 value.

As long as L and H are relatively close, L and H can only transmit 1 value, and C1 and C2 can not transmit;

in order to transmit only one value such as L, the receiving end only decodes, the invention sets 4 different states, and codes based on the last two low significant bits of L, thus ensuring the information extraction only.

The original BTC method compresses 4 × 4 image blocks to have 2 average values (8 bits each) and 16 binary digits for each image block, which is 32 bits in total. The invention processes the last 16 bits to form two 8-bit values, so that the last 16 bits are processed jointly with the first 16 bits of data, and can be further compressed.

For example, 126,128,95,180, 126 has a binary value of "11111110", and the lowest 2-bit 10 becomes 00, which becomes "11111100", which becomes decimal 124.

126 and 128 in 126,128,95 and 180 are relatively close and can be represented by a number 124, and no transmission is needed for 128,95 and 180, and in the case of the variable compression ratio of the invention, only one number is transmitted for four numbers, so that the compression ratio is greatly improved.

In the case of the fixed compression ratio of the present invention, four numbers still transmit four numbers, but the first number is 124, and the last three numbers can be any three-byte number, such as the byte stream of data S, changed into 124, S1, S2, S3, thereby fixing the compression ratio and increasing the hiding amount.

(4) The invention fully utilizes the special property of the AMBTC block compression parameter to obtain a compressed image, hides the image of the region of interest and improves the quality of the recovered image through correction.

(5) In order to simplify the implementation of the method, the smoothing process may be omitted, and the original image may be a1 or a sub-image a2, thereby increasing the real-time performance of compression.

(6) The invention can realize any information hiding under the condition of fixed compression ratio. Any information is changed into the byte stream S, and the hiding method of the byte stream S is followed for hiding.

Drawings

FIG. 1 is a flow chart of the method of the present invention. Fig. 1(a) is a transmission scheme of a variable compression ratio of the present invention, and fig. 1(b) is a transmission scheme of a fixed compression ratio of the present invention.

Detailed Description

Taking a typical block (4 × 4 blocks) as an example, the conventional AMBTC compression ratio is 4 times, and the compression PSNR is about 30dB, which cannot be more than 4 times, or even more than 8 times.

The invention firstly carries out smooth processing and zooming by a certain multiple, 2-4 times, and then carries out conventional AMBTC compression coding, such as 4 times compression coding, on the zoomed image to obtain the compression ratio of 8-16 times. Although the PSNR of the scaled image itself may satisfy the requirement, the original image quality deteriorates. Generally, the image quality is poor, for example, the PSNR is lower than 30dB, or about 25dB, and the user requirement may not be satisfied.

However, the invention provides a combined compression method for hiding the compressed data of the interested region image in the AMBTC image compressed data and extracting the interested region data at the receiving end to correct the original image, and the recovery quality of the whole image is ensured and the quality of the interested region is also improved through the ingenious coding of the compression parameters.

The invention provides a simple and efficient combined compression method under the condition of keeping a certain PSNR performance, and the combined compression method can generally reach 4-32 under the condition of a fixed compression ratio; the compression ratio under the condition of variable compression ratio can reach 8-64, and even the PSNR under the condition of 64-200 times can reach about 25dB (lena. bmp).

The invention relates to a low-cost joint image compression transmission method, which comprises the following steps:

(1) setting an original image as A0, wherein the size of the original image is M × N, 8 bits of quantization are used for quantization, firstly, the image is subjected to smoothing processing, and the smoothed image is set as A1; original image a0 is lena. bmp, M × N ═ 512 × 512.

(2) Scaling the image a1 to form a sub-image a2 compressed at R0 and divided into non-overlapping blocks of size K1 × K2; each block obtained X1, X2, … Xk gray values, K-K1 × K2;

R0＝4，K1*K2＝4*4＝16

(3) when the sub-image a2 and the image a0 are 4, the size of a2 is 256 × 256, AMBTC compression is performed, the compressed three groups L, H, and B are further compressed and transmitted to obtain compressed data, the compression ratio is R2 × 4, the compressed data is decompressed to obtain a restored image A3 of a2, the size of A3 is 256 × 256, and the restored image A3 is amplified by R0 times.

(4) The restored original image A4 is obtained, A4 is 512 x 512

The method for continuously transmitting the compressed three groups of numbers L, H and B with the variable compression ratio in the step (3) is as follows:

1) decomposing three groups of numbers L, H and B of each block into L, H, C1 and C2, wherein C1 is 1-q bits of B, C2 is 1-q bits of B, and q is B/2;

B-K1-K2-16, C1-8, C2-8, q-8, C1 takes 1,3,5,7,9,11,13,15, C2 of 16 bits of B takes 2,4,6,8,10,12,14,16 of B

(alternative: C2 takes 1,3,5,7,9,11,13,15 of the 16 bits of B, C1 takes 2,4,6,8,10,12,14,16 of B)

2) Comparing L and H, C1 with C2 and encoding (d1, d2 are threshold values)

The general setting d1 is 0-64, d2 is 0-4

Typical values are d 1-4 and d 2-1

l is reserved; h is not transmitted;

l reserved, H reserved, C1 reserved; c2 does not transmit;

l reserved, H reserved, C1 reserved, C2 reserved;

transmitting the data reserved after compression;

3) decompressing the retained values obtained in step 2) to restore the image a3

If the data received by each non-overlapping block is X, Y, Z1, Z2, then

When the lowest 2 bits of X are 00 or 01,

restoring all the gray values of the non-overlapped blocks to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

The region-of-interest image F1 of the original image is a rectangular region, and is subjected to AMBTC compression processing to obtain processed data S, wherein the byte stream of the data S is S1, S2, S3, S4 … … and Sn

2) Hiding of compressed data byte stream:

comparing L and H, C1 with C2 and encoding (d1, d2 are threshold values)

Generally, d1 is 0-64, d2 is 0-4,

typical values are d 1-4 and d 2-1

the lowest 2 position of L is 01,

l is reserved; h was replaced by S1, C1 by S2 and C2 by S3;

the L-lowest 2 position is 10,

when H-L > d1, if C1 is not the same as C2 in terms of number of bits > d2,

the L-lowest 2 position is 11,

l is reserved, H is unchanged, C1 is unchanged, and C2 is unchanged;

Transmitting the data reserved after compression;

3) decompressing the retained values to recover image a3

Assuming that the data received by each block K1X K2 are X, Y, Z1 and Z2, then

When X is at the 2-lowest position of 00, L ═ X, H ═ X, a3 ═ X;

y for the recovered byte stream S1, Z1 for S2, Z2 for S3,

restoring all the gray values of the block to be X;

when the 2-lowest position of X is 01, L ═ X, H ═ X, A3 ═ X;

recovered byte stream S1 ═ Y, S2 ═ Z1, S3 ═ Z2

Restoring all the gray values of the block to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

recovering byte stream S1 ═ Z2;

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

no recovery byte stream;

the image A3 is restored according to the restored value of each block, and enlarged to a size of a 1.

The invention relates to a low-cost combined image compression transmission method, which comprises the following steps:

variable compression ratio case:

r0 ═ 1 (without pretreatment)

K1*K2＝4*4

R1＝4

R>4

R0 ═ 2 (pretreatment)

K1*K2＝4*4

R1>4

R>8

R0 ═ 4 (pretreatment)

K1*K2＝4*4

R1>4

R>16

R0 ═ 8 (pretreatment)

K1*K2＝4*4

R1>4

R>32

R0 ═ 16 (pretreatment)

K1*K2＝4*4

R1>4

R>64

R0 ═ 32 (pretreatment)

K1*K2＝4*4

R1>4

R>128

Fixed compression ratio case:

r0 ═ 1 (without pretreatment)

K1*K2＝4*4

R1＝4

R＝4

R0 ═ 2 (pretreatment)

K1*K2＝4*4

R1＝4

R＝8

R0 ═ 4 (pretreatment)

K1*K2＝4*4

R1＝4

R＝16

R0 ═ 8 (pretreatment)

K1*K2＝4*4

R1＝4

R＝32

R0 ═ 16 (pretreatment)

K1*K2＝4*4

R1＝4

R＝64

R0 ═ 32 (pretreatment)

K1*K2＝4*4

R1>4

R＝128

To verify the performance of the algorithm presented herein, the experiment was simulated using an 8-bit grayscale image of size 512 × 512. The degree of change of the image is expressed in terms of peak signal-to-noise ratio (PSNR).

Bmp is taken as an example, K1 × K2 ═ 4 × 4.

And compressing the image by 4 times by utilizing an AMBTC conventional compression method, and calculating the PSNR.

Image a1 and recovery a 4: compression recovery PSNR 33 dB;

the compression method of the invention is used for compressing the image by 4 times, and the PSNR is calculated, so that the compression recovery PSNR can reach 36dB and is better than 33 dB.

The compression method is utilized to compress the image by 8 times, PSNR is calculated, and the compression recovery PSNR of the image A3 and the recovery A4 is approximately equal to 32 dB;

the image is compressed by 16 times by using the compression method of the invention, and the PSNR is calculated, so that the image A3, the image A3 and the recovery image A4 are as follows: compression recovery PSNR ≈ 31 dB.

The invention provides a new method for compressing image data with low cost, which breaks through the defect of low compression ratio of the original block compression method, achieves the purpose that 4-16 times of compression has better image quality, even the compression ratio between 16 times and 200 times keeps certain image quality, has the characteristics of low realization complexity, strong practicability and the like, and has far lower realization complexity than JPEG (joint photographic experts group) compression methods, thereby having practical value in a low-cost image transmission system.

The invention is not described in detail and is within the knowledge of a person skilled in the art.

Claims

1. A low-cost joint image compression transmission method is characterized by comprising the following steps:

setting the size of an original image A0 as M x N, quantizing 8 bits, and firstly smoothing the original image A0 to obtain a smoothed image A1;

step (2), carrying out scaling treatment with the compression ratio of R0 on the image A1 to form a sub-image A2; dividing the sub-image a2 into non-overlapping blocks of size K1 × K2, each block yielding X1, X2, … Xk gray-values, K1 × K2;

step (3), performing AMBTC compression on the sub-image A2, continuously compressing three groups of numbers L, H and B of each non-overlapped block after compression to obtain compressed data, decompressing to obtain a restored image A3 of the sub-image A2, and amplifying the restored image A3; wherein L is the average value of the gray levels smaller than the average gray level value of the block in each block, H is the average value of the gray levels larger than or equal to the average gray level value of the block in each block, B is 0 or 1, K1 × K2 bit values are total, 0 corresponds to the gray level smaller than the average gray level value of the block, and 1 corresponds to the gray level larger than or equal to the average gray level value of the block;

step (4), obtaining a restored original image A4;

step 131), decomposing B of the three groups of blocks into C1 and C2;

step 132), setting d1 and d2 as threshold values, and performing data transmission according to L, H and the values of C1 and C2;

step 133), decompressing the retained value obtained in step 132) to restore the image a 3;

in the step 131), C1 is q bits of B1, 3,5,7,9,11,13,15 …, C2 is q bits of B2, 4,6,8,10,12,14, 16.. q is K1 × K2/2;

the specific process of step 132) is as follows:

l is reserved; h is not transmitted;

l reserved, H reserved, C1 reserved; c2 does not transmit;

l reserved, H reserved, C1 reserved, C2 reserved;

and transmitting the data reserved after compression.

2. A low-cost joint image compression transmission method according to claim 1, wherein: the specific process of the step 133) is as follows:

if the data received by each non-overlapping block is X, Y, Z1, Z2, then

When the lowest 2 bits of X are 00 or 01,

restoring all the gray values of the non-overlapped blocks to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

3. A low-cost joint image compression transmission method according to claim 1, wherein: the method for transmitting the compressed three groups of numbers L, H and B with the fixed compression ratio in the step (3) is as follows:

step 231) decomposing three groups of numbers L, H and B of each block into L, H, C1 and C2; processing the region-of-interest image F1 of the original image, and combining the position information of F1 and the processed data into data S, wherein the byte stream of S is S1, S2, S3, … … and Sn;

step 232), hiding the compressed data byte stream: setting d1 and d2 as threshold values, hiding the compressed data byte stream and transmitting data according to L, H and the values of C1 and C2;

step 233), the retained values are decompressed to recover image a 3.

4. A low-cost joint image compression transmission method according to claim 3, wherein: in the step 231), C1 is q bits of B1, 3,5,7,9,11,13,15 …, C2 is q bits of B2, 4,6,8,10,12,14, 16.. q is K1 × K2/2.

5. A low-cost joint image compression transmission method according to claim 4, characterized in that: the specific method of the step 232) is as follows:

the lowest 2 position of L is 01,

l is reserved; h was replaced by S1, C1 by S2 and C2 by S3;

the L-lowest 2 position is 10,

when H-L > d1, if C1 is not the same as C2 in terms of number of bits > d2,

the L-lowest 2 position is 11,

l is reserved, H is unchanged, C1 is unchanged, and C2 is unchanged;

sequentially comparing L and H, C1 of the subsequent blocks with C2, wherein d1 and d2 are threshold values; sequentially hiding the byte streams which are not hidden in the data S according to the same method;

and transmitting the data reserved after compression.

6. A low-cost joint image compression transmission method according to claim 5, wherein: the specific method of the step 233) is as follows:

assuming that the data received by each block K1X K2 are X, Y, Z1 and Z2, then

When X is at the 2-lowest position of 00, L ═ X, H ═ X, a3 ═ X;

y for the recovered byte stream S1, Z1 for S2, Z2 for S3,

restoring all the gray values of the block to be X;

when the 2-lowest position of X is 01, L ═ X, H ═ X, A3 ═ X;

recovered byte stream S1 ═ Y, S2 ═ Z1, S3 ═ Z2

Restoring all the gray values of the block to be X;

when the lowest 2 position of X is 10, L is X, H is Y, C1 is Z1, C2 is Z1;

recovering byte stream S1 ═ Z2;

7. A low-cost joint image compression transmission method according to claim 6, wherein: in the step 233), the specific method for restoring the gray value of the non-overlapping block through decoding includes:

when the lowest 2 position of L is 11, L is X, H is Y, C1 is Z1, C2 is Z2;

there is no recovery byte stream.