CN100558166C - A kind of encoding and decoding method of graphic terminal interface - Google Patents

Abstract

The invention discloses a graphical terminal interface coding method. The method selects a reference position of a current encoding position and uniformly sets position marks. When encoding the current encoding position, the pixel values of the reference position are searched first. When the pixel value of the pixel point at the current coding position does not appear at the reference position, the position mark of the current coding position is coded into a code word, the code word is coded into a code stream, and then the pixel value of the pixel point at the current coding position is coded into the code stream. When the pixel value of the pixel point at the current coding position appears at the reference position, the position mark of the reference position is coded into a code word, the run length is searched backwards, the run length is coded into the code word, and the code word is coded into a code stream. The coding and decoding algorithm of the graphical terminal interface is simple, the coding efficiency is high, the operation amount is small, the requirement on the computing power of an application server and a terminal is not high, and the graphical terminal interface is suitable for terminal equipment with weak computing power.

Description

A kind of encoding and decoding method of graphic terminal interface

technical field

The present invention relates to the technical field of interface data transmission in the graphic terminal application protocol, and more specifically, the present invention relates to a coding and decoding method in the graphic terminal interface data transmission.

Background technique

With the rapid development of network infrastructure and Internet, the use of various terminal equipment including PC, PDA, NC, etc. is becoming more and more popular. Using a network terminal to access various applications on the network has become an important way for people to obtain information and various network services. Among various access methods, it has gradually become a trend to use various graphical interface applications on the network through the graphical terminal application protocol. Using this method can greatly reduce the maintenance and management costs of various software and hardware.

In the graphic terminal application, the application program is stored on the terminal server, and the terminal executes the application program on the server through the network. As shown in Figure 1, the graphical interface of the application program is drawn into the virtual screen on the terminal server. The graphic interface interaction between the terminal and the terminal server is mainly completed through the virtual screen. The terminal server checks the status of the virtual screen. If the data in some areas of the virtual screen (the shaded part in Figure 1) changes, the terminal server transmits the virtual screen of the changed area (usually a rectangular area) to the terminal at regular intervals. data and display it on the terminal screen. The data of the virtual screen is generally stored in BMP format, which sequentially describes the pixel value of each pixel in the virtual screen, and the number of bits (BPP) of the pixel value can be set to 8 bits, 16 bits, 24 bits or 32 bits.

In the graphics terminal application shown in Figure 1, the screen data is transmitted from the terminal server to the terminal through the network, and efficient and fast encoding and compression of the screen data to reduce the amount of network data transmission is a key to the graphics terminal application protocol technology. In the graphic terminal application process, the graphic interface data compression on the application server generally requires a lossless compression process, and the data format is BMP format. Since the terminal application server handles multiple user graphical interfaces at the same time, the compression algorithm must be simple and fast, and the terminal may include various devices from PC to PDA, which requires the decoding algorithm to be as simple as possible. In addition, in the graphic terminal application, the data transmission of the graphic interface will occupy a large amount of network bandwidth, and providing the highest possible compression rate is also a basic requirement of the compression algorithm.

Contents of the invention

The purpose of the present invention is to provide a simple and fast encoding and decoding method for graphic interface data in BMP format, and improve the compression rate as much as possible.

In order to achieve the above object, the present invention provides a graphical terminal interface encoding method, which is used to encode the virtual screen data of the rectangular area on the terminal server, so as to be transmitted to the terminal through the network. The method includes:

Selecting the reference position of the current encoding position, and uniformly setting a position mark for the current encoding position and the reference position; the reference position is located before the current encoding position in the encoding sequence;

When encoding a certain current encoding position, first search for the pixel value of its reference position, and compare the pixel values of the current encoding position and the pixel point of the reference position, and then encode according to two situations:

When the pixel value of the pixel point at the current encoding position does not appear at its reference position, the position mark of the current encoding position is encoded into the code word of the pixel point at the current encoding position; this code word is encoded into the code stream, and then the current The pixel value of the pixel at the encoding position is encoded into the code stream;

When the pixel value of the pixel point of the current coding position appears at its reference position, the position mark of the reference position is compiled into the codeword of the pixel point of the current coding position, and the pixel point with the same pixel point of the current coding position is searched backward. The run length of the pixel point of value, and this run length is coded into the code word of the pixel point of the current coding position; This code word is coded into the code stream;

The rectangular area is encoded sequentially from the starting position according to the aforementioned method until the encoding of the entire rectangular area is completed. When encoding, the starting position of the rectangular area is the pixel point in the upper left corner of the area, and the encoding sequence is from left to right and from top to bottom.

The reference positions of the current encoding position are respectively one position to the left, two positions to the upper left, two positions directly to the upper side, and two positions to the upper right of the current encoding position. The position marks of the current encoding position and its reference position can be represented by three consecutive binary bits, which are located in the upper three bits of the codeword.

The position marks of the current encoding position and its reference position are respectively: the first position on the left is 011, the first position on the upper left is 110, the second position on the left is 010, the first position on the upper left is 101, and the second position on the upper left is 010. The first position is 001, the first position on the upper right is 100, the second position on the upper right is 000, and the current encoding position is 111.

The search for the pixel value of the reference position is carried out in a predetermined order, and when the pixel value of a reference position is found to be the same as the pixel value of the current encoding position during the search, the search is stopped.

When the pixel value of the pixel point at the current encoding position occurs at its reference position, the code word of the pixel point at the current encoding position also includes a run length flag bit, which is used to identify the number of bits occupied by the run length in the code word . The position mark, the run length flag bit and the run length are arranged next to each other in the code word.

When the pixel value of the pixel point at the current encoding position appears at its reference position, the code word of the pixel point at the current encoding position is composed of one or two bytes, wherein the high three bits of the first byte of the code word are the reference position The position mark, followed by the run length mark bit, according to the different states of the run length mark bit, the lower four bits of the first byte or the lower four bits of the first byte and the following byte are used as the run length .

When the pixel value of the pixel point at the current coding position does not appear at its reference position, the code word consists of one byte, wherein the upper three bits of the byte are the position mark of the current coding position, and the other bits are reserved bits.

The palette pixels of the rectangular area are greater than 16.

The present invention provides a graphic terminal interface decoding method for decoding data encoded by the aforementioned encoding method, the method comprising:

Reading the codeword at the current decoding position in the code stream, the current decoding position corresponds to the current encoding position in the encoding method;

The position markers in the codeword are read and then decoded according to two conditions:

If the read position mark is the position mark of the current encoding position in the encoding method, then the pixel value following the codeword in the code stream is used as the pixel value of the current decoding position pixel point;

If the read position mark is the position mark of a reference position in the encoding method, then read the run length n in the codeword, and set the pixel values of n consecutive pixel points starting from the current decoding position as the The pixel value of the reference position corresponding to the position marker;

Read the code words in the code stream in order until the entire data is decoded.

When reading the run length in the code word, according to the state of the run length position mark in the code word, read bits of different lengths in the code stream as the run length.

The encoding method of the graphical terminal interface of the present invention belongs to lossless encoding, the encoding method has a simple algorithm, and does not require high computing power of the application server and the terminal, and the corresponding decoding algorithm is relatively simple, which can significantly reduce the amount of calculation during decoding. It is suitable for mobile terminal equipment with weak computing power. The coding method of the present invention considers the two-dimensional information of the graphical interface during the coding process, and the coded value can be reference coded according to the 7 points of the current position, the left front, the upper left, the upper right, and the upper right, so as to reduce the correlation of the encoded graphical interface data, Improve coding efficiency.

Description of drawings

Fig. 1 is a schematic diagram of transmission of virtual screen data in a graphics terminal application;

Fig. 2 is an embodiment of selecting the reference position of the current encoding position in the encoding method of the present invention;

Fig. 3 is an embodiment of carrying out position mark setting to the current encoding position shown in Fig. 2 and its reference position;

Fig. 4 is the schematic diagram of codeword structure in the coding method of the present invention; Wherein Fig. 4 (a) is the direct coding to current coding position, Fig. 4 (b) is the indirect coding of current coding position relative to reference position.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

In the graphic terminal application protocol, when the terminal server detects that the virtual screen area changes, when the preset timer times out, the method provided by the present invention is used to encode the BMP data in the changed area, wherein the encoding area is generally rectangular , the description parameter is a quaternion array (x, y, w, h), where x and y indicate the coordinate position of the upper left corner of this area on the virtual screen, and w and h indicate the width and length of this area. In the drawing process of the graphical interface, there are currently two typical application program interfaces X and Windows. In these two cases, you can add a hook (Hook) function to the interface, that is, to wrap these interfaces again. , you can record all the operations of the application on the screen, and the areas involved in these operations are the changed areas. The description parameters (x, y, w, h) of the coding region can be obtained through the hook (Hook) function.

In the present invention, the coding of the rectangular coding area starts from the pixel point at the upper left corner thereof, and proceeds in order from left to right and from top to bottom.

In the process of encoding, the encoding of the pixel point at the current encoding position needs to refer to the encoding of its reference position, so here it is first necessary to define the reference position of the current encoding position. As shown in Fig. 2, in one embodiment of the present invention, a total of 7 reference positions are selected, wherein No. 7 position in Fig. 2 is the current encoding position, and positions 0-6 are the reference positions of the current encoding position. The reference positions in Fig. 2 are respectively the left front position (No. 1), the upper left two positions (No. 2 and 4), the two positions directly above (No. 0 and 5), and the upper right two positions of the current encoding position (No. 7). location (No. 3 and 6). It can be seen from Fig. 2 that the reference position is located before the current encoding position in the encoding order.

After the selection of the reference position is completed, the current encoding position and the reference position need to be marked uniformly, so that the position mark can be encoded into the codeword during encoding, and it is also necessary to judge the position mark according to the position mark in the codeword during decoding. Whether the codeword is a direct encoding of the current encoding position or an encoding relative to a reference position. The role of position markers will be explained in more detail below. Fig. 3 shows an embodiment of the present invention to mark the position of the reference position, corresponding to the position in Fig. 2, the position mark of the current encoding position is 111, the position before the left of the current encoding position is 011, and the first position on the upper left It is 110, the second upper left position is 010, the first upper right position is 101, the second upper right position is 001, the first upper right position is 100, and the second upper right position is 000.

After the position marks of the current encoding position and its reference position are set, the rectangular area to be encoded can be encoded. The encoding strategy of the present invention is:

When encoding a certain current coding position, first search for the pixel value of its reference position, an exemplary search sequence is that the pixel values of the current coding position in the region are in sequence from 0 to 6 according to the labeling sequence in Figure 2 Compare. Then encode according to two cases (A) and (B):

(A) after the search finishes, the pixel value of the current coding position pixel point does not appear at its reference position, then the position mark of the current coding position is coded into the code word of the current coding position pixel point; this code word is coded into code stream, and then encode the pixel value of the pixel at the current coding position into the code stream.

In an embodiment of this case, the encoding of the pixel point at the current encoding position is shown in FIG. 4(a), which includes two parts, the code word and the pixel value of the pixel point at the current encoding position. Wherein in the embodiment shown in Fig. 4 (a), the code word occupies one byte, and the position mark is located at the upper three bits t ₇₆₅ of the code word, that is, the 5-7 bits of the code word, corresponding to the aforementioned setting of the current encoding position It can be known that t ₇₆₅ =111. The other bits t ₄₃₂₁₀ in the codeword are reserved bits in this encoding scheme and in one embodiment they are all set to zero. The BPP/8 bytes following the code word store the pixel value of the pixel at the current encoding position, where BPP is the number of BMP data bits for each pixel, and BPP=8/16/24/32.

(B) in the search process, when the pixel value of the pixel point of the current encoding position occurs at a certain reference position, then stop searching and code the position mark of the reference position into the code word of the pixel point of the current encoding position, and then send Then search for the run length of the pixel point with the same pixel value as the pixel point at the current encoding position, and encode the run length into the code word of the pixel point at the current encoding position; encode the code word into the code stream.

In an embodiment of this situation, as shown in Figure 4 (b) to the coding codeword of current coded position pixel point, the upper three t ₇₆₅ of codeword are stored and searched for the position mark of the reference position, and in The lower four bits t ₃₂₁₀ of this byte store the run length. In order to expand the range of the run length that can be stored in the code word, the fourth bit _t4 in the code word can be used as the run length mark bit. When the run-length flag bit _t4 was set to 0, the codeword length was one byte, and the lower four bits of the byte represented the run length in t ₃₂₁₀ , and the maximum run length that could be represented was 15; when the run-length flag bit was set When it is 1, the length of the code word is two bytes, the lower four bits t ₃₂₁₀ of the first byte in the code word and the following byte represent the length of the run length with a total of 12 bits, and the maximum run length that can be represented is 4095.

The rectangular area is encoded sequentially from the starting position according to the above-mentioned encoding method of the present invention, and is sequentially encoded into the code stream until the encoding of the entire rectangular area is completed.

The encoding method of the present invention will be further described in detail below by taking a specific rectangular block of 16*16 pixels as an example. As shown in Table 1, the number in the first column indicates the row of the rectangular block, and the number in the first row indicates the column of the rectangular block. Other elements in Table 1 represent pixels, and their values represent the pixel value of the pixel point. A pixel can be represented by its corresponding row number and column number as coordinates, for example, the pixel in row 0 and column 8 is represented by (0, 8). In this embodiment, BPP=32, that is, the pixel value is represented by four bytes.

Table 1

Encoding starts from (0,0) pixel. From the reference position shown in Figure 2, it can be seen that in the rectangular area shown in Table 1, the (0,0) pixel has no reference position, so the pixel value of the (0,0) pixel point cannot appear at the reference position , so the upper three bits t ₇₆₅ of the codeword are the position mark of the current encoding position, and it can be seen that the position mark is 111 in FIG. 3 . The other bits in the codeword are set to 0, that is, the codeword is 11100000. After the code word is encoded into the code stream, following the code word, the pixel value (00000000, 00000000, 00000000, 00000000) of the (0,0) pixel is encoded into the code stream, and the (0, 0) pixel encoding.

Then, (0,1) pixel becomes the current encoding position, (0,0) pixel is the reference position of (0,1) pixel, and (0,1) pixel and (0,0) pixel The pixel values are the same. It can be seen from the position mark in FIG. 3 that when the (0,0) pixel is used as the reference position of the (0,1) pixel, its position mark is 011, so 011 is coded into the upper three bits of the code word. From (0,1) pixel to (0,7) pixel, 7 consecutive pixels have the same pixel value, so the run length is 7, and the run length is less than 15, so the run length mark position in the codeword Zero, the lower four bits in the codeword represent a runlength of 0111. Finally, the encoded codeword from (0,1) pixel to (0,7) pixel is 01100111.

Then, the (0, 8) pixel becomes the current coding position, and the coding continues according to the coding method of the present invention.

When the (2,4) pixel becomes the current coding position, at its reference position, the pixel values of the (1,4) and (1,5) pixels are the same as (2,4) and both are 3. Because the search of the reference position is carried out in the order from 0 to 6 in Fig. 2, the pixel value of the (1,4) pixel and the pixel value of the current coding position (2,4) pixel are first searched Likewise, the search can be stopped at this point. It can be seen from FIG. 3 that when the (1,4) pixel is used as the reference position of the (2,4) pixel, its position is marked as 101, so 101 is coded into the upper three bits of the currently encoded codeword. From (2, 4) pixel to (2, 15) pixel, 12 consecutive pixels have the same pixel value, so the run length is 12, and the run length is less than 15, so the run length mark position in the codeword Zero, the lower four bits in the codeword represent a runlength of 1100. Finally, the encoded codeword from (2, 4) pixels to (2, 15) pixels is 10101100.

When the (13,0) pixel becomes the current coding position, its pixel value is 17, and there is no identical pixel value at its reference position, so the upper three bits t ₇₆₅ of the codeword are the position mark of the current coding position, as shown in Fig. 3, it can be seen that the position is marked as 111. The other bits in the codeword are set to 0, that is, the codeword is 11100000. After the code word is encoded into the code stream, following the code word, the pixel value (00000000, 00000000, 00000000, 00010001) of the (13, 0) pixel is encoded into the code stream, and the (13, 0) pixel encoding.

Then, the (13, 1) pixel becomes the current coding position, and at its reference position, the (13, 0) pixel has the same pixel value as the current coding position (13, 1) pixel. It can be seen from the position mark in Fig. 3 that when the (13,0) pixel is used as the reference position of the (13,1) pixel, its position mark is 011, so 011 is coded into the upper three bits of the code word. From (13, 1) pixel to (15, 15) pixel, 47 consecutive pixels have the same pixel value, so the run length is 47, and the run length is greater than 15, so the run length mark position in the codeword is 1, the lower four bits in the codeword and the subsequent byte are 12 to represent the run length (0000 00101111). Finally, the encoded codeword from (13,1) pixel to (15,15) pixel is (01110000 00101111).

Finally, the code stream of the entire rectangular block encoded according to the method of the present invention is: a total of 952 (bit).

11100000, 00000000, 00000000, 00000000, 00000000, 01100111,

11100000, 00000000, 00000000, 00000000, 00000001, 01100111,

11100000, 00000000, 00000000, 00000000, 00000010, 01100011,

11100000, 00000000, 00000000, 00000000, 00000011, 01101001,

11100000, 00000000, 00000000, 00000000, 00000100, 01100101,

10101100,

11100000, 00000000, 00000000, 00000000, 00000101, 01100011,

11000100,

11100000, 00000000, 00000000, 00000000, 00000110, 01100111,

11100000, 00000000, 00000000, 00000000, 00000111, 01100110,

10001000,

11100000, 00000000, 00000000, 00000000, 00001000, 01100011,

01000100,

11100000, 00000000, 00000000, 00000000, 00001001, 01100111,

11100000, 00000000, 00000000, 00000000, 00001010, 01100111,

11100000, 00000000, 00000000, 00000000, 00001011, 01100011,

11100000, 00000000, 00000000, 00000000, 00001100, 01100011,

10101000,

11100000, 00000000, 00000000, 00000000, 00001101, 01100011,

11100000, 00000000, 00000000, 00000000, 00001110, 01100011,

11100000, 00000000, 00000000, 00000000, 00001111, 01110000,

00010001,

11100000, 00000000, 00000000, 00000000, 00010000, 01110000,

00010101, 01000110, 10110000, 00100010,

11100000, 00000000, 00000000, 00000000, 00010001, 01110000,

00101111,

After understanding the above-mentioned encoding method of the present invention, the corresponding decoding method is also very obvious. When decoding the code stream, read a code word (or a byte) in the code stream, and judge the position mark in the code word. For example, in the foregoing embodiments, the position mark is located at the end of the code word High three. When the position mark is the position mark of the current encoding, which is 111 in the aforementioned embodiment, then the pixel value following the codeword in the code stream is taken as the pixel value of the pixel point at the current decoding position. When the position mark is a position mark of a reference position, in the aforementioned embodiment, it is the position mark of each reference position defined in Fig. 3, then reading the run length n in the codeword will start from the current decoding position The pixel values of n consecutive pixel points in are set as the pixel values of the reference position corresponding to the position marker. When the run length flag bit is defined in the code word, bits of different lengths in the code stream are read as the run length according to the state of the run length position mark in the code word.

Corresponding to the aforementioned encoding embodiment, in a decoding embodiment, the following decoding steps can be used:

(1) get the upper three bits of the codeword;

(2) if the value of the upper three bits of the codeword is the position marker of the current encoding position (that is, the current decoding position), i.e. 111 among Fig. 3, then the pixel value of the current position begins from the next byte;

(3) If the value of the upper three bits of the codeword is the position mark of the reference position of the current encoding position (that is, the current decoding position), that is, other values except 111 in FIG. 3 . Continue to read the run length flag bit adjacent to the position mark in the codeword, that is, the fourth bit t ₄ in the codeword:

(1) when the fourth t ₄ is 0, it means that the length of the code word is a byte, and the value of the lower four t ₃₂₁₀ adjacent to the run length mark position in the read code word is used as the length of the run;

(II) When the fourth bit t ₄ is 1, it means that the length of the code word is two bytes, read the subsequent 12-bit value adjacent to the run-length flag in the code word, i.e. t in Fig. 4(b) ₃₂₁₀ bits and the following byte are used as the length of the run, and the length is set to n;

(4) Set the pixel values of n consecutive pixel points starting from the current decoding position as the pixel value of the reference position corresponding to the position mark in the codeword and the pixel value of the current decoding position. Especially, when the position mark in the code word is 011, as shown in Figure 3, the pixel value of setting and the previous position pixel value of current decoding position are identical; When the position mark in the code word is outside 011 When other reference positions are marked, the set pixel value is the same as the pixel value of the position shown in the following formula:

c-(2-t ₇ )×w+(1-t ₆₅ )

Where c is the current decoding position, t ₇ is the value of the seventh bit of the codeword, w is the width of the encoded rectangular area, and t ₆₅ is the value of the sixth and fifth bits of the codeword.

(5) Read the next codeword, repeat the process of steps (1) to (4) until all codewords are solved.

Although in the above-mentioned specific embodiment, described adopting the method of the present invention to encode and decode a rectangular area of 16 * 16 pixels, and selected the reference position as shown in Figure 2, and adopted BPP=32 Bits are used to describe the pixel value of a pixel point, but the present invention is not limited to this embodiment. Those of ordinary skill in the art can understand that the present invention can also be used for the encoding of rectangular areas of other sizes. For example, in the embodiment of the present invention, the longest run length is represented by 12 binary bits, and the maximum representable run length is 4095 , so any rectangular area no more than 4096 pixels can adopt the method of the present invention. Moreover, for a larger rectangular area, the method of the present invention can be adopted by dividing it into a plurality of small rectangular areas. Those skilled in the art can also understand that the present invention can use other bits such as 8/16/24 to describe the pixel value, and these changes should be included in the technical solution of the present invention.

In the present invention, the number of palette pixels contained in the rectangular area to be coded (that is, the number of different pixel values contained in the rectangular area) is not limited, but the present invention is preferably applied to color palette A rectangular area with more than 16 board pixels. For the case where the number of pixels in the palette is less than or equal to 16, it is preferable to use some existing coding schemes. As shown in Fig. 4 (a), when the pixel value of the pixel point of the current encoding position does not appear in its reference position, some reserved bits t ₄₃₂₁₀ are left in the codeword of the present invention, which can be used to combine with other existing encoding The interface of the method, by defining these reserved bits, the technical solution of the present invention can be conveniently used in combination with some existing coding schemes, so as to obtain better effects. For example, a preferred encoding scheme is shown below:

(a1) If the whole rectangular block is a color, then the code word is directly set as (11110000), that is, the reserved bit t ₄₃₂₁₀ is set to 10000, then the current pixel value is stored in (BPP/8) byte, and returns.

(a2) If the number of palette pixels in the entire rectangular area is greater than 1 and less than or equal to 16, then the upper four bits of the codeword are set to (1111), and the lower four bits are the number of palette pixels, which are about to be reserved. Set to (1+number of pixels in the palette). Then use (BPP/8×the number of palette pixels) bytes to store the palette pixel value, followed by the pixel index value, and return. where, if:

If the number of pixels in the palette is 2, use 1 bit to represent an index;

If the number of pixels in the palette is [3...4], use 2 bits to represent an index;

If the number of pixels in the palette is [5...16], 4 bits are used to represent an index.

(a3) If the number of palette pixels in the entire rectangular area is greater than 16, directly adopt the encoding method provided in the present invention for encoding.

The corresponding decoding scheme is:

(b1) get the upper three bits of the codeword;

(b2) If the value of the upper three bits of the codeword is 111,

The fourth bit of the code word, if it is 1,

The lower four bits of the code word, if it is 0,

Indicates that the entire rectangular area is a pixel value, and the pixel value starts from the next byte;

otherwise

Indicates that the number of pixels in the palette of the entire rectangular block is less than 16, and the pixel value starts from the next byte, and the number is the lower four bits of the codeword, followed by the palette index of each pixel of the entire rectangular block , the number of index digits varies with the number of pixels in the palette;

otherwise

Decoding according to the decoding method provided by the present invention;

(b3) If the value of the upper three bits of the codeword is not 111, then perform decoding according to the decoding method provided by the present invention.

Claims (10)

1. A graphical terminal interface encoding method, which is used to encode virtual screen data in a rectangular area on a terminal server, so as to be transmitted to the terminal through the network; the method includes: Selecting the reference position of the current encoding position, and uniformly setting a position mark for the current encoding position and the reference position; the reference position is located before the current encoding position in the encoding order; The reference positions of the current coding position are respectively the previous left position, the upper left two positions, the two upper right positions, and the upper right two positions of the current encoding position; When encoding a certain current encoding position, first search for the pixel value of its reference position, and compare the pixel values of the current encoding position and the pixel point of the reference position, and then encode according to two situations: The position marks of the current coding position and its reference position are represented by three consecutive binary bits, which are located in the upper three bits of the codeword; The position marks of the current encoding position and its reference position are respectively: the first position on the left is 011, the first position on the upper left is 110, the second position on the left is 010, the first position on the upper left is 101, and the second position on the upper left is 010. The first position is 001, the first position on the upper right is 100, the second position on the upper right is 000, and the current encoding position is 111; When the pixel value of the pixel point at the current encoding position does not appear at its reference position, the position mark of the current encoding position is encoded into the code word of the pixel point at the current encoding position; this code word is encoded into the code stream, and then the current The pixel value of the pixel at the encoding position is encoded into the code stream; When the pixel value of the pixel point of the current coding position appears at its reference position, the position mark of the reference position is compiled into the codeword of the pixel point of the current coding position, and the pixel point with the same pixel point of the current coding position is searched backward. The run length of the pixel point of value, and this run length is coded into the code word of the pixel point of the current coding position; This code word is coded into the code stream; The rectangular area is encoded sequentially from the starting position according to the aforementioned method until the encoding of the entire rectangular area is completed. 2. The graphic terminal interface coding method according to claim 1, wherein the starting position of the rectangular area is the pixel point in the upper left corner of the area, and the coding sequence is from left to right and from top to bottom . 3. The graphical terminal interface coding method according to claim 1, wherein the search for the pixel value of the reference position is compared sequentially from 0 to 6 according to the order of the labels, and when a pixel value of the reference position is found in the search process Stop searching when it is the same as the pixel value of the current coding position. 4. The graphic terminal interface coding method according to claim 1, wherein when the pixel value of the pixel point at the current coding position appears at its reference position, the codeword of the pixel point at the current coding position also includes a The run length flag bit is used to identify the number of bits occupied by the run length in the codeword. 5. The graphic terminal interface encoding method according to claim 4, wherein, when the pixel value of the pixel point at the current encoding position appears at its reference position, the position mark, the run length flag and the run length are in the code The order of words is arranged next to each other. 6. The graphic terminal interface coding method according to claim 1, wherein, when the pixel value of the pixel point at the current coding position appears at its reference position, the code word of the pixel point at the current coding position is composed of one or Composed of two bytes, the high three bits of the first byte of the codeword are the position mark of the reference position, followed by the run length mark bit, according to the different states of the run length mark bit, the low four bits of the first byte or The lower four bits of the first byte and the following byte are the run length. 7. The graphic terminal interface coding method according to claim 1, wherein, when the pixel value of the pixel point at the current coding position does not appear at its reference position, the codeword is composed of one byte, wherein the byte of the byte The upper three bits are the position mark of the current encoding position, and the other bits are reserved bits. 8. The graphic terminal interface coding method according to claim 1 or 7, characterized in that the palette pixels of the rectangular area are greater than 16. 9. A graphic terminal interface decoding method for decoding data encoded by the encoding method according to claim 1, the method comprising: Reading the codeword at the current decoding position in the code stream, the current decoding position corresponds to the current encoding position in the encoding method; The position markers in the codeword are read and then decoded according to two conditions: If the read position mark is the position mark of the current encoding position in the encoding method, then the pixel value following the codeword in the code stream is used as the pixel value of the current decoding position pixel point; If the read position mark is the position mark of a reference position in the encoding method, then read the run length n in the codeword, and set the pixel values of n consecutive pixel points starting from the current decoding position as the The pixel value of the reference position corresponding to the position mark; the run length n is 1-4095; Read the code words in the code stream in order until the entire data is decoded. 10. The graphic terminal interface decoding method according to claim 9, characterized in that, when reading the run length in the code word, it is necessary to read the run length in the code stream according to the state of the run length position mark in the code word Bits of different lengths are used as run lengths.

Images