CN102088604B - A method and device for compressing video thumbnails - Google Patents

Abstract

The embodiment of the invention discloses a method for compressing film thumbnails, comprising the following steps: carrying out loss compression on the film thumbnails to acquire film thumbnails with specific digits; carrying out run coding on the film thumbnails with the specific digits; and carrying out lossless compression on the film thumbnails which are subjected to the run coding to acquire compressed pictures. The embodiment of the invention also provides a device for compressing film thumbnails. In the invention, the loss compression, the run coding and the lossless compression are combined effectively to compress the film thumbnails, thus having the advantages of higher compression ratio, high compression efficiency, low degree of distortion and fast decoding speed.

Description

A method and device for compressing video thumbnails

technical field

The invention relates to the technical field of video thumbnail processing, in particular to a method and device for compressing video thumbnails.

Background technique

Lossy compression utilizes the insensitivity of human beings to certain frequency components in images or sound waves, allowing certain information to be lost during the compression process. Although the original data cannot be completely restored, the lost part has a greater impact on understanding the original image. Small and high compression ratio. Lossy compression is widely used in the compression of speech, image and video data. Generally, the compression of movie thumbnails usually adopts lossy compression based on floating-point operations.

Run-length coding, also known as "run-length coding" or "run-length coding", its basic principle is: use a symbol value or string length to replace consecutive symbols with the same value (continuous symbols constitute a continuous "stroke". encoding (hence the name), making the symbol length less than the length of the original data. Only when the code of each row or column of data changes, the code and the number of repetitions of the same code are recorded at one time, so as to realize data compression.

Lossless compression is to use the statistical redundancy of data for compression, which can completely restore the original data without causing any distortion. Common lossless compressions include LZ77 encoding compression or LZ78 encoding compression. Both LZ77 and LZ78 encoding compression are based on one-way search and matching of the sliding window dictionary, and the matching length depends on the size of the sliding window.

In the video resource management system, the file data volume of the video thumbnail is large. In order to effectively save the space of the video resource management system, it is necessary to compress the video thumbnail. Generally, any one of the above-mentioned three compression methods is generally used in the existing compression of movie thumbnails, and the inventors have found that there are defects in using any of the above-mentioned schemes to compress movie thumbnails:

1. The algorithm based on floating-point arithmetic performs lossy compression on movie thumbnails, and its compression rate is relatively high. However, due to the complexity of the algorithm of floating-point arithmetic, the compression rate of this scheme is low, easy to be distorted, and the decoding speed is slow. Fully restore the original image;

2. The compression rate of run-length coding is low, especially when the color of the video thumbnail is relatively rich, the compression effect of this scheme is not high, and the practicability is low;

3. The lossless compression scheme based on LZ77 or LZ78 coding compression has fast compression speed and is not easy to be distorted, but the compression rate of this scheme is low and the compression effect is not high.

Contents of the invention

The technical problem to be solved by the embodiments of the present invention is to provide a method and device for compressing film thumbnails, which can effectively combine lossy compression, run-length coding and lossless compression to compress film thumbnails. Higher, high compression rate, low distortion, fast decoding speed.

In order to solve the above technical problems, an embodiment of the present invention provides a method for compressing movie thumbnails, including:

Perform lossy compression on video thumbnails to obtain video thumbnails with a specific number of digits;

performing run-length encoding on the movie thumbnail image with a specific number of bits;

Perform lossless compression on the run-length coded video thumbnail to obtain a compressed image.

Wherein, the movie thumbnail is a 24-bit bmp bitmap in RGB (Red, Green, Blue) format.

Wherein, the lossy compression is performed on the thumbnail of the movie to obtain the thumbnail of the movie with a specific number of digits, including:

Create a unified palette and a unified header for all 24-bit bmp bitmaps in RGB format for the movie;

A lossy compression algorithm based on logical operations is used to compress the pixels of each 24-bit bmp bitmap in RGB format to obtain a 16-bit bmp bitmap in RGB format.

Wherein, performing run-length coding on the movie thumbnail with a specific number of digits includes:

Use matrix storage to store all 16-bit bmp bitmaps in RGB format of the movie;

Z word scanning is carried out to the 16-bit bmp bitmap of the RGB format stored in the matrix;

Perform run-length encoding on the 16-bit bmp bitmap in RGB format stored in the matrix according to the zigzag scanning result.

Wherein, the lossless compression is performed on the video thumbnail after the run-length encoding to obtain the compressed picture, including:

Scan the sliding window in both directions according to the run-length coded video thumbnail;

Judging whether the sliding window includes a character string that matches the run-length coded movie thumbnail, if the judgment result is yes, outputting a five-element symbol group as a compressed picture, and sliding the matching backwards in the sliding window Add 1 character to the length of the longest character string; if the judgment result is no, output a binary symbol group as a compressed image, and slide the sliding window backward by 1 character;

Repeat the above steps until all video thumbnails are compressed;

Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, next character c; the binary symbol group includes: encoding flag bit and The current character e.

Wherein, the off adopts fixed-length encoding, and the len adopts prefix encoding.

Correspondingly, an embodiment of the present invention also provides a compression device, including:

The lossy compression module is used to perform lossy compression on video thumbnails to obtain video thumbnails with a specific number of digits;

A run-length encoding module, configured to perform run-length encoding on the video thumbnail image of a specific number of bits obtained by the lossy compression module;

The lossless compression module is used to perform lossless compression on the movie thumbnail coded by the run-length coding module to obtain a compressed picture.

Wherein, the film thumbnail is a 24-bit bmp bitmap in RGB format, and the lossy compression module includes:

Creation unit for creating a unified palette and a unified header for all 24-bit bmp bitmaps in RGB format for the film;

The lossy compression unit is used to compress the pixels of each 24-bit bmp bitmap in RGB format by adopting a lossy compression algorithm based on logical operation to obtain a 16-bit bmp bitmap in RGB format.

Wherein, the run-length coding module includes:

The storage unit is used to store all 16-bit bmp bitmaps in RGB format of the film in a matrix storage mode;

A scanning unit, configured to perform Z-scanning on the 16-bit bmp bitmap in RGB format stored in the storage unit;

A run-length encoding unit, configured to perform run-length encoding on the 16-bit bmp bitmap in RGB format stored in the matrix according to the Z-shaped scanning result of the scanning unit.

Wherein, the lossless compression module includes:

The bidirectional scanning unit is used to scan the sliding window bidirectionally according to the video thumbnail after run-length encoding;

A judging unit, configured to judge whether the sliding window scanned by the two-way scanning unit includes a character string matching the run-length coded movie thumbnail;

The lossless compression unit is used to output the five-element symbol group as a compressed picture when the judgment result of the judgment unit is yes, and add 1 character to the length of the longest character string matched by sliding the sliding window backwards; When the judgment result of the judging unit is no, the binary symbol group is output as a compressed picture, and the sliding window is slid backward by 1 character;

Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, next character c; the binary symbol group includes: encoding flag bit and The current character e.

Implementing the embodiment of the present invention has the following beneficial effects:

The embodiment of the present invention first uses lossy compression to reduce the number of digits of the movie thumbnail to obtain a movie thumbnail with a specific number of digits. Here, the lossy compression is based on logical operations, and the operation is simple, which not only ensures the compression ratio, but also improves the compression ratio. rate, while reducing the degree of distortion; then perform run-length encoding on the video thumbnail with a specific number of digits, and perform lossless compression on the run-length encoded video thumbnail, such as bidirectional search and matching based on a sliding window dictionary, to obtain the final compressed image , the embodiment of the present invention improves the existing LZ77 or LZ78 encoding and compression, and improves the conventional one-way search matching into two-way search matching, which improves the compression rate while ensuring the original compression rate and distortion; the implementation of the present invention The example effectively combines lossy compression, run-length encoding and lossless compression, which effectively saves the space of the video resource management system, improves the decoding speed, makes the display of video thumbnails smoother, and improves the user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is the structural representation of the embodiment of compression device of the present invention;

Fig. 2 is a schematic structural view of an embodiment of a lossy compression module of the present invention;

FIG. 3 is a schematic structural diagram of an embodiment of the run-length encoding module of the present invention;

FIG. 4 is a schematic structural view of an embodiment of a lossless compression module of the present invention;

Fig. 5 is the flow chart of the first embodiment of the method for compressing movie thumbnails of the present invention;

FIG. 6 is a flow chart of the second embodiment of the video thumbnail compression method of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Please refer to Fig. 1, which is a schematic structural view of an embodiment of a compression device of the present invention; the compression device includes:

The lossy compression module 10 is used to perform lossy compression on the video thumbnail to obtain the video thumbnail with a specific number of digits.

In a specific implementation, the lossy compression module 10 may be a device such as a processing chip integrated with a lossy compression algorithm based on logic operations. Generally, the original movie thumbnail is a 24-bit bmp bitmap in RGB format, and the lossy compression module 10 is mainly used to compress the 24-bit bmp bitmap in RGB format into a 16-bit bmp bitmap in RGB format.

The run-length encoding module 20 is configured to perform run-length encoding on the movie thumbnail image with a specific number of bits obtained by the lossy compression module 10 .

The run-length encoding module 20 encodes the 16-bit bmp bitmap in the RGB format obtained by the lossy compression module 10, and can describe more bmp bitmap data with a small amount of bytes, thereby achieving the purpose of further compression, Especially when the color of each bmp bitmap is relatively single, a better compression effect can be obtained by using the run-length encoding method, and the encoding process of the run-length encoding will not be distorted.

The lossless compression module 30 is configured to perform lossless compression on the video thumbnail encoded by the run-length encoding module 20 to obtain a compressed picture.

In a specific implementation, the lossless compression module 30 adopts an adaptive dictionary coding compression method based on a sliding window, and uses information that has been losslessly compressed (that is, information that has been encoded) as a sliding window, and the size of the sliding window can be determined according to actual needs. Make settings. Different from the one-way search matching method of the existing LZ77 or LZ78 encoding compression, the lossless compression of this embodiment adopts the two-way search matching method. When there is a matching character string in the sliding window, the lossless compression module 30 outputs a five-element The symbol group is used as a compressed picture, and the length of the longest string matched by the sliding window is added by 1 character; when there is no matching string in the sliding window, the lossless compression module 30 outputs binary symbols group as a compressed image, and slide the sliding window backward by 1 character. Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, and next character c. Wherein, the binary symbol group includes: a coding flag bit and a current character e.

The embodiment of the present invention effectively combines lossy compression, run-length coding and lossless compression to compress movie thumbnails. The compression rate is high, the compression rate is high, the degree of distortion is small, and the decoding speed is fast.

In order to illustrate the present invention more clearly, each module in the compression device of the present invention will be introduced in detail below.

Please refer to Fig. 2, which is a schematic structural diagram of an embodiment of the lossy compression module of the present invention; the lossy compression module 10 includes:

The creation unit 101 is used to create a unified palette and a unified header file for all 24-bit bmp bitmaps in RGB format of the movie.

The 24-bit bmp bitmap includes: a header file and a palette; wherein, the header file contains information such as the type of the bmp bitmap, display content, etc.; the palette is a flat surface for blending and shelving fresh paint. Each thumbnail of the movie (24-bit bmp bitmap in RGB format) may contain different header files and palettes. The movie resource management system needs to store the header files and palettes of each bmp bitmap at the same time, which wastes a lot of storage space. The creation unit 101 creates a unified palette and a unified header file for all 24-bit bmp bitmaps in RGB format of the movie, then the movie resource management system only needs to store the unified palette and the unified header file, and each bmp bitmap is shared The unified color palette and unified header file not only save the storage space of the movie resource management system, but also greatly compress the redundant data of the bmp bitmap.

The lossy compression unit 102 is configured to use a lossy compression algorithm based on logic operations to compress the pixels of each 24-bit bmp bitmap in RGB format to obtain a 16-bit bmp bitmap in RGB format.

BMP bitmap, also known as bitmap image or drawing image, is composed of pixels. Each pixel of a 24-bit bmp bitmap in RGB format is 24 bits (3 bytes), represented by RGB, and its organizational format is: R+G+B (R is 8bit, G is 8bit, B is 8bit ). The lossy compression unit 102 uses a lossy compression algorithm based on logic operations to compress each pixel of the 24-bit bmp bitmap, specifically: removing the last 3 bits of R in each pixel to compress 5bit; remove the last 2 bits of G in each pixel to compress it to 6bit; remove the last 3 bits of B in each pixel to compress it to 5bit; after the above compression, the 24-bit RGB format The 3-byte (24-bit) pixels of the bmp bitmap are compressed into 2-byte (16-bit) pixels, and its organization format is R+G+B (R is 5bit, G is 6bit, B is 5bit). Such as: the 24-bit bmp bitmap that color is red, its R=0xff, G=0x00, B=0x00, described lossy compression unit 102 adopts the lossy compression algorithm based on logic operation to compress, and its algorithm is: dev_red_color= ((((B) & 0xf8) >> 3)︱(((G)&0xfc)<<3)︱(((R)&0xf8)<<8)); Finally, all red 24-bit bmp The bitmaps are all 0xf800, which is a 16-bit bmp bitmap converted to red. It should be noted that the lossy compression unit 102 uses a lossy compression algorithm based on logical operations for compression, which is simple in operation and ensures a compression rate; in the above compression process, the last 3 bits of R in each pixel, The last 2 bits of G and the last 3 bits of B will not affect the display of the bmp bitmap, which ensures the low distortion of compression.

The embodiment of the present invention uses lossy compression to reduce the number of digits of the movie thumbnail, and obtains a movie thumbnail with a specific number of digits. Here, the lossy compression is based on logical operations, and the operation is simple, which not only ensures the compression rate, but also improves the compression rate. , while reducing distortion.

Please refer to FIG. 3 , which is a schematic structural diagram of an embodiment of the run-length encoding module of the present invention; the run-length encoding module 20 includes:

The storage unit 201 is used to store all 16-bit bmp bitmaps in RGB format of the film in a matrix storage manner.

After being compressed by the lossy compression module 10, the movie thumbnails are all 16-bit bmp bitmaps in RGB format, and the specifications of each movie thumbnail are the same. Format 16-bit bmp bitmap for storage. For example: suppose two 16-bit bmp bitmaps in RGB format are respectively: 000001111100000 and 000000000011111, wherein 0 and 1 both represent the color of the 16-bit bmp bitmap; the 16-bit bmp bitmaps in the two RGB formats are stored via the After unit 201 is stored, its matrix is as follows:

000001111100000

000000000011111

The scanning unit 202 is configured to perform zigzag scanning on the 16-bit bmp bitmap in RGB format stored in the storage unit 201 .

According to the above example, after the scanning unit 202 Z-scans the matrix stored in the storage unit 201 , sequentially obtain: 5 0s, 5 1s, 15 0s, and 5 1s.

The run-length encoding unit 203 is configured to perform run-length encoding on the 16-bit bmp bitmap in RGB format stored in the matrix according to the Z-shaped scanning result of the scanning unit 202 .

According to the above example, the run-length encoding unit 203 describes the bmp bitmap by using the encoding method of color plus color length according to the Z-scanning result of the scanning unit 202, and the result encoded by the run-length encoding unit 203 is: 0 #5+1#5+0#15+1#5; Among them, "#" indicates the color of the bmp bitmap, and "#" indicates the length of the color. The 16-bit bmp bitmap in RGB format is encoded by run-length coding, and more bmp bitmap data can be described with a small number of bytes, which achieves the purpose of further compression, especially when the color of each bmp bitmap is relatively single. A relatively good compression effect can be obtained by using the run-length encoding method, and the encoding process of the run-length encoding will not be distorted.

The embodiment of the present invention has simple operation, high compression rate and low distortion.

Please refer to Fig. 4, which is a schematic structural diagram of an embodiment of the lossless compression module of the present invention; the lossless compression module 30 includes:

A bi-directional scanning unit 301, configured to scan the sliding window bi-directionally according to the run-length coded film thumbnail;

In the specific implementation, the lossless compression in the embodiment of the present invention adopts the adaptive dictionary coding compression method based on the sliding window, and uses the information that has undergone lossless compression (that is, the information that has been encoded) as the sliding window. The size of the sliding window can be determined according to the actual situation. Settings are required. Different from the unidirectional search and matching method of the existing LZ77 or LZ78 encoding and compression, the lossless compression of this embodiment adopts a bidirectional search and matching method, and the bidirectional scanning unit 301 scans the sliding window bidirectionally according to the thumbnail image after the run-length encoding, To achieve bidirectional search matching.

A judging unit 302, configured to judge whether the sliding window scanned by the two-way scanning unit 301 includes a character string that matches the run-length coded movie thumbnail;

A lossless compression unit 303, configured to output a five-element symbol group as a compressed picture when the determination result of the determination unit 302 is yes, and add 1 character to the length of the longest character string matched by sliding the sliding window backwards; When the judgment result of the judging unit is negative, the binary symbol group is output as a compressed picture, and the sliding window is slid backward by 1 character.

In the specific implementation, the five-element symbol group includes: encoding flag bit: indicates whether the compressed picture has been encoded through a dictionary based on a sliding window, bit=1, indicating that it has been encoded; bit=0, indicating that it is not encoded; the direction flag dir : Indicates the scanning direction of the sliding window when the compressed image is compressed by dictionary encoding based on the sliding window, dir=1 indicates scanning from front to back, dir=0 indicates scanning from back to front; offset within the window off: indicates the offset position of the character string with the maximum matching length in the sliding window; the maximum matching length len: indicates the length of the longest character string in the sliding window that matches the movie thumbnail after the run-length encoding; the next character c. The lossless compression unit 303 outputs a five-element symbol group (bit, dir, off, len, c) as a compressed picture when the determination result of the determination unit 302 is yes, and slides the sliding window backward by len+1 characters.

编码等前缀编码方式。 The off uses fixed-length coding, and the number of bits required for coding off is bitnum=upper_bound (MAX_WND_SIZE), where MAX_WND_SIZE is the maximum sliding window size. The len is coded by a prefix, preferably, the len is coded in normal form Huffman, and in practical applications, the len can also be coded by Golomb or

Encoding and other prefix encoding methods.

The binary symbol group includes: a coding flag bit and a current character e. When the determination result of the determination unit 302 is negative, the lossless compression unit 303 outputs a five-element symbol group (bit, e) as a compressed picture, and slides the sliding window backward by 1 character.

It can be understood that each unit in the lossless compression module 30 repeats the above work until all thumbnails of the movie are compressed.

The embodiment of the present invention improves the existing LZ77 or LZ78 encoding and compression, and improves the conventional one-way search and matching into two-way search and matching, which improves the compression rate and ensures the original compression rate and distortion.

In order to illustrate the present invention more clearly, the compression method performed by the compression device on movie thumbnails will be described in detail below.

Please refer to Fig. 5, which is a flow chart of the first embodiment of the method for compressing movie thumbnails of the present invention; the method includes:

S101. Perform lossy compression on the video thumbnail to obtain a video thumbnail with a specific number of bits.

In a specific implementation, the S101 may be executed by a device such as a processing chip integrated with a lossy compression algorithm based on logical operations. Generally, the original movie thumbnail is a 24-bit bmp bitmap in RGB format, and the S101 compresses the 24-bit bmp bitmap in RGB format into a 16-bit bmp bitmap in RGB format.

S102. Perform run-length encoding on the video thumbnail image with a specific number of digits.

Said S102 encodes the 16-bit bmp bitmap in RGB format obtained in said S101, and can use a small amount of bytes to describe more bmp bitmap data, thereby achieving the purpose of further compression, especially when each bmp bitmap When the color is relatively single, a better compression effect can be obtained by using the run-length encoding method, and the encoding process of the run-length encoding will not be distorted.

S103. Perform lossless compression on the run-length coded movie thumbnail to obtain a compressed picture.

In the specific implementation, the S103 adopts the adaptive dictionary encoding compression method based on the sliding window, and uses the information that has undergone lossless compression (that is, the information that has been encoded) as the sliding window, and the size of the sliding window can be set according to actual needs. . Different from the one-way search matching method of the existing LZ77 or LZ78 encoding compression, the lossless compression of this embodiment adopts the two-way search matching method. When there is a matching character string in the sliding window, the S103 outputs the five-element symbol group as Compress the picture, and add 1 character to the length of the longest character string matched by sliding the sliding window backwards; when there is no matching character string in the sliding window, the S103 outputs a binary symbol group as a compressed picture, and Slide the sliding window backward by 1 character. Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, and next character c. Wherein, the binary symbol group includes: a coding flag bit and a current character e.

The embodiment of the present invention effectively combines lossy compression, run-length coding and lossless compression to compress movie thumbnails. The compression rate is high, the compression rate is high, the degree of distortion is small, and the decoding speed is fast.

Please refer to FIG. 6, which is a flow chart of the second embodiment of the method for compressing movie thumbnails of the present invention; the method includes:

S201, creating a unified palette and a unified header file for all 24-bit bmp bitmaps in RGB format of the movie.

The 24-bit bmp bitmap includes: a header file and a palette; wherein, the header file contains information such as the type of the bmp bitmap, display content, etc.; the palette is a flat surface for blending and shelving fresh paint. Each thumbnail of the movie (24-bit bmp bitmap in RGB format) may contain different header files and palettes. The movie resource management system needs to store the header files and palettes of each bmp bitmap at the same time, which wastes a lot of storage space. Said S201 creates a unified palette and a unified header file for all 24-bit bmp bitmaps in RGB format of the movie, then the movie resource management system only needs to store the unified palette and the unified header file, and each bmp bitmap shares the unified palette. The color palette and the unified header file not only save the storage space of the video resource management system, but also greatly compress the redundant data of the bmp bitmap.

S202, using a lossy compression algorithm based on logical operations to compress the pixels of each 24-bit bmp bitmap in RGB format to obtain a 16-bit bmp bitmap in RGB format.

BMP bitmap, also known as bitmap image or drawing image, is composed of pixels. Each pixel of a 24-bit bmp bitmap in RGB format is 24 bits (3 bytes), represented by RGB, and its organizational format is: R+G+B (R is 8bit, G is 8bit, B is 8bit ). The S202 uses a lossy compression algorithm based on logic operations to compress each pixel of the 24-bit bmp bitmap, specifically: removing the last 3 bits of R in each pixel to compress it to 5 bits; removing The last 2 bits of G in each pixel make it compressed to 6 bits; remove the last 3 bits of B in each pixel to make it compressed to 5 bits; after the above compression, the 24-bit bmp bitmap in RGB format 3-byte (24-bit) pixels are compressed into 2-byte (16-bit) pixels, and its organization format is R+G+B (5 bits for R, 6 bits for G, and 5 bits for B). For example: a 24-bit bmp bitmap whose color is red, its R=0xff, G=0x00, B=0x00, the S202 is compressed using a lossy compression algorithm based on logical operations, and its algorithm is: dev_red_color=(((( B) & 0xf8) >> 3)︱(((G)&0xfc)<<3)︱(((R)&0xf8)<<8)); Finally, all red 24-bit bmp bitmaps are 0xf800, which is a 16-bit bmp bitmap converted to red. It should be noted that the S202 uses a logical operation-based lossy compression algorithm for compression, the operation is simple, and the compression rate is guaranteed; in the above compression process, the last 3 bits of R and the last 2 bits of G in each pixel are removed. bit and the last 3 bits of B will not affect the display of the bmp bitmap, which ensures low compression distortion.

In this embodiment, S201-S202 are specific detailed steps of S101 in the previous embodiment.

S203, adopting a matrix storage method to store all 16-bit bmp bitmaps in RGB format of the film.

After being compressed by said S202, the movie thumbnails are all 16-bit bmp bitmaps in RGB format, and the specifications and sizes of each movie thumbnail are the same, and said S203 uses matrix storage to store all 16-bit bmp bitmaps in RGB format to store. For example: suppose two 16-bit bmp bitmaps in RGB format are respectively: 000001111100000 and 000000000011111, wherein 0 and 1 represent the color of 16-bit bmp bitmaps; the 16-bit bmp bitmaps in these two RGB formats are passed through the S203 After storage, its matrix is as follows:

000001111100000

000000000011111

S204. Z-scan the 16-bit bmp bitmap in RGB format stored in the matrix.

According to the above example, after the said S204 Z-scan the matrix stored in the said S203, sequentially obtain: 5 0s, 5 1s, 15 0s, 5 1s.

S205. Perform run-length encoding on the 16-bit bmp bitmap in RGB format stored in the matrix according to the Z-scanning result.

According to the above example, the S205 describes the bmp bitmap by using the encoding method of color plus color length according to the Z-scanning result of the S204, and the encoded result of the S205 is: 0#5+1#5+0 #15+1#5; Among them, the front of "#" indicates the color of the bmp bitmap, and the back of "#" indicates the length of the color. The 16-bit bmp bitmap in RGB format is encoded by run-length coding, and more bmp bitmap data can be described with a small number of bytes, which achieves the purpose of further compression, especially when the color of each bmp bitmap is relatively single. A relatively good compression effect can be obtained by using the run-length encoding method, and the encoding process of the run-length encoding will not be distorted.

In this embodiment, S203-S205 are specific detailed steps of S102 in the previous embodiment.

S206. Bidirectionally scan the sliding window according to the run-length coded video thumbnail.

In the specific implementation, the lossless compression in the embodiment of the present invention adopts the adaptive dictionary coding compression method based on the sliding window, and uses the information that has undergone lossless compression (that is, the information that has been encoded) as the sliding window. The size of the sliding window can be determined according to the actual situation. Settings are required. Different from the unidirectional search and matching method of the existing LZ77 or LZ78 encoding and compression, the lossless compression of this embodiment adopts a two-way search and matching method, and the S206 scans the sliding window in both directions according to the thumbnail of the run-length coded film, so as to realize two-way Search for matches.

S207, judging whether the sliding window includes a character string matching the run-length coded movie thumbnail, if the judging result is yes, go to S208; otherwise, go to S209.

S208, outputting a five-element symbol group as a compressed picture, and adding 1 character to the length of the longest character string matched by sliding the sliding window backwards; after that, turn to S210.

编码等前缀编码方式。 In the specific implementation, the five-element symbol group includes: encoding flag bit: indicates whether the compressed picture has been encoded through a dictionary based on a sliding window, bit=1, indicating that it has been encoded; bit=0, indicating that it has not been encoded; the direction flag bit dir : Indicates the scanning direction of the sliding window when the compressed image is compressed by dictionary encoding based on the sliding window. dir=1 indicates scanning from front to back, dir=0 indicates scanning from back to front; the offset within the window off: indicates the offset position of the character string with the maximum matching length in the sliding window; the maximum matching length len: indicates the length of the longest character string in the sliding window that matches the movie thumbnail after the run-length encoding; the next character c. In S208, when the judgment result in S207 is yes, output a five-element symbol group (bit, dir, off, len, c) as a compressed picture, and slide the sliding window backward by len+1 characters. The off uses fixed-length coding, and the number of bits required for coding off is bitnum=upper_bound(MAX_WND_SIZE). The len is coded by a prefix, preferably, the len is coded in normal form Huffman, and in practical applications, the len can also be coded by Golomb or

Encoding and other prefix encoding methods.

S209. Output binary symbol groups as compressed pictures, and slide the sliding window backward by 1 character.

The binary symbol group includes: a coding flag bit and a current character e. In said S209, when the determination result of said S207 is No, output a five-element symbol group (bit, e) as a compressed picture, and slide the sliding window backward by 1 character.

S210, judging whether all the movie thumbnails have been compressed, if the judging result is yes, then end; otherwise, go to S207.

In this embodiment, S206-S210 are detailed steps of S103 in the previous embodiment.

The embodiment of the present invention effectively combines lossy compression, run-length coding and lossless compression to compress movie thumbnails. The compression rate is high, the compression rate is high, the degree of distortion is small, and the decoding speed is fast.

Through the description of the above embodiments, the embodiment of the present invention first uses lossy compression to reduce the number of digits of the movie thumbnail to obtain a movie thumbnail with a specific number of digits. Here, the lossy compression is based on logical operations, and the operation is simple, which ensures The compression rate increases the compression rate and reduces the distortion; then run-length encoding is performed on the video thumbnail of the specific number of digits, and the lossless compression is performed on the video thumbnail after the run-length encoding, such as two-way based on the sliding window dictionary Search and match to obtain the final compressed picture. The embodiment of the present invention improves the existing LZ77 or LZ78 encoding and compression, and improves the conventional one-way search and match to two-way search and match, which improves the compression rate and ensures the original compression rate at the same time and distortion; the embodiment of the present invention effectively combines lossy compression, run-length coding and lossless compression three compression methods, effectively saves the space of the film resource management system, improves the decoding speed, and makes the display of film thumbnails smoother , which improves the user experience.

Those of ordinary skill in the art can understand that realizing all or part of the runs in the methods of the above-mentioned embodiments can be completed by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, the program During execution, the runs as in the embodiments of the above-mentioned methods may be included. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM), etc.

What is disclosed above is only a preferred embodiment of the present invention, and certainly not to limit the scope of rights of the present invention. Those of ordinary skill in the art can understand that all or part of the above-mentioned embodiments can be realized, and according to the rights of the present invention The equivalent changes required still belong to the scope covered by the invention.

Claims (2)

1. A method for compressing movie thumbnails, comprising: Perform lossy compression on video thumbnails to obtain video thumbnails with a specific number of digits; performing run-length encoding on the movie thumbnail image with a specific number of bits; Perform lossless compression on the run-length coded video thumbnail to obtain the compressed picture; The described method of performing lossless compression on the run-length encoded film thumbnail to obtain a compressed picture includes: Scan the sliding window in both directions according to the run-length coded video thumbnail; Judging whether the sliding window includes a character string that matches the run-length coded movie thumbnail, if the judgment result is yes, outputting a five-element symbol group as a compressed picture, and sliding the matching backwards in the sliding window Add 1 character to the length of the longest character string; if the judgment result is no, output a binary symbol group as a compressed image, and slide the sliding window backward by 1 character; Repeat the above steps until all video thumbnails are compressed; Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, next character c; The binary symbol group includes: encoding flag bit and current character e; The film thumbnail is a bitmap of 24-bit bmp in RGB format; The lossy compression of the movie thumbnail to obtain a movie thumbnail with a specific number of digits includes: Create a unified palette and a unified header for all 24-bit bmp bitmaps in RGB format for the movie; Use a lossy compression algorithm based on logical operations to compress the pixels of each 24-bit bmp bitmap in RGB format to obtain a 16-bit bmp bitmap in RGB format; performing run-length encoding on the video thumbnail of the specified number of bits, including: Use matrix storage to store all 16-bit bmp bitmaps in RGB format of the movie; Z word scanning is carried out to the 16-bit bmp bitmap of the RGB format stored in the matrix; According to the Z word scanning result, the 16-bit bmp bitmap of the RGB format of described matrix storage is carried out run-length coding; The off uses fixed-length coding, and the len uses prefix coding. 2. A compression device, characterized in that, comprising: The lossy compression module is used to perform lossy compression on video thumbnails to obtain video thumbnails with a specific number of digits; A run-length encoding module, configured to perform run-length encoding on the video thumbnail image of a specific number of bits obtained by the lossy compression module; A lossless compression module, configured to perform lossless compression on the video thumbnail encoded by the run-length encoding module to obtain a compressed picture; The lossless compression module includes: The bidirectional scanning unit is used to scan the sliding window bidirectionally according to the video thumbnail after run-length encoding; A judging unit, configured to judge whether the sliding window scanned by the two-way scanning unit includes a character string matching the run-length coded movie thumbnail; The lossless compression unit is used to output the five-element symbol group as a compressed picture when the judgment result of the judgment unit is yes, and add 1 character to the length of the longest character string matched by sliding the sliding window backwards; When the judgment result of the judging unit is no, the binary symbol group is output as a compressed picture, and the sliding window is slid backward by 1 character; Wherein, the five-element symbol group includes: encoding flag bit, direction flag bit dir, offset off in the window, maximum matching length len, next character c; The binary symbol group includes: encoding flag bit and current character e; The film thumbnail is a bitmap of 24 bit bmp in RGB format, and the lossy compression module includes: Creation unit for creating a unified palette and a unified header for all 24-bit bmp bitmaps in RGB format for the film; The lossy compression unit is used to compress the pixels of the 24-bit bmp bitmap of each RGB format by using a lossy compression algorithm based on logic operations to obtain a 16-bit bmp bitmap of the RGB format; The run-length encoding module includes: The storage unit is used to store all 16-bit bmp bitmaps in RGB format of the film in a matrix storage mode; Scanning unit, for carrying out Z word scanning to the 16-bit bmp bitmap of the RGB format of described matrix storage; A run-length encoding unit, configured to perform run-length encoding on the 16-bit bmp bitmap in RGB format stored in the matrix according to the Z-shaped scanning result.

Images