CN1558664A - A method for synthesizing subtitles and background images of high-definition digital TV programs - Google Patents

Abstract

Because high-definition digital TV subtitle images do not use data compression technology, a large amount of storage space is required, and they cannot be transmitted to other systems for overlay processing through the network. The invention provides a method for synthesizing subtitles and background images of high-definition digital television programs, comprising the following steps: the subtitle compression module serially receives the original subtitle image file through software, and after the original subtitle image is input into the subtitle compression module, the image is compressed using run-length coding technology , to generate a subtitle compressed file in a specific format; the subtitle decompression module receives the subtitle compressed file data through hardware serial, and decompresses the original subtitle image through run-length decoding technology; the subtitle synthesis module obtains the alpha of each pixel in the decompressed subtitle image Value overlays the subtitle image pixel-by-pixel over the background image. The invention can greatly save the storage space of subtitle images, and significantly reduce the limitation of subtitle image files in editing, playing and network transmission.

Description

A method for synthesizing subtitles and background images of high-definition digital TV programs

technical field

The invention relates to a method for synthesizing subtitles and background images of high-definition digital television programs.

Background technique

In digital television systems, it is often necessary to superimpose various text descriptions on images, and these text descriptions are usually called "subtitles". The current method is generally implemented by using hardware equipment similar to a subtitle machine. As shown in Figure 1, the original subtitle image and background image are sent to the synthesis module, and then the synthesis module is based on the Alpha value of each pixel of the input subtitle image or its own alpha value. Overlay setting, superimpose the subtitle image on the background image pixel by pixel, and send the superimposed video image to the digital TV for display. The original subtitle image file is usually saved in RGB format, and each pixel is composed of three components of R, G, and B. Each pixel of some subtitle image files also has its own independent 8-bit Alpha value.

From the above description, it can be found that since no data compression technology is used, the subtitle overlay solution of directly sending the subtitle image to the subtitle machine requires a large amount of storage space, especially for high-definition digital TV, a picture with Alpha value The subtitle image of 1920*1600 requires 7.91M storage space. If the subtitle of a video file requires 100,000 images, then 700G storage space is required. Obviously, such a storage requirement is unbearable for common video editing systems. This makes subtitle image files can only be superimposed on the local system, but cannot be transmitted to other systems for superimposition processing through the network, which is very inconvenient to use.

Contents of the invention

The purpose of the present invention is to provide a method for synthesizing subtitles and background images of high-definition digital television programs to overcome the limitation that uncompressed subtitle images require massive storage space.

The invention adopts the method of using software to compress subtitles and hardware decompression to synthesize subtitles and background images in the production of high-definition digital television programs. The present invention adopts run-length coding compression technology for subtitle image compression. The run-length coding compression technology refers to: the compression module records the first value, scans the input values one by one from the beginning, and stops at the place where the recorded value is not exactly the same, and then records the recorded value and the number of the same value Record to the compressed file, and then the compression module records the current value, and continues scanning from the current position until all data is processed. The run-length decoding technology based on the run-length coding compression technology is called the run-length decoding technology. Taking the compressed subtitle image in RGB format as an example, the method of using run-length coding to compress the subtitle image is: the compression module records the R, G, B value and Alpha value of the first pixel, scans the input pixels one by one from the beginning, and directly encounters Stop at pixels that are not exactly the same as the recorded R, G, B, and Alpha values. Then record the recorded R, G, B, Alpha values and the same number of pixels into the compressed file. Then the compression module records the R, G, B values and Alpha values of the pixels at the current position, and continues scanning from the current position until all pixels of the original subtitle image are processed. On the contrary, the run-length decoding technology decompresses the file obtained by the run-length coding compression method to obtain the original subtitle image.

For reaching above-mentioned purpose, the present invention is achieved like this:

A method for synthesizing subtitles and background images of high-definition digital television programs, comprising the following steps: 1) using a subtitle compression module, after the original subtitle image is input into the subtitle compression module, using run-length coding technology to compress the image, and generating a subtitle compression file in a specific format; 2) adopt the subtitle decompression module, after the subtitle decompression module receives the subtitle compressed file, decompress and obtain the original subtitle image; 3) adopt the subtitle synthesis module, receive the subtitle image that decompresses The alpha value superimposes the subtitle image on top of the background image.

In the present invention, the original subtitle image and the compressed subtitle file may be in the same format, for example, RGB format or YUV format.

The subtitle compression module of the present invention can realize serial reception of compressed image files saved in 32-bit BMP format through software.

The subtitle decompression module of the present invention can serially receive subtitle compressed file data through hardware.

The manner of overlaying subtitle images in the present invention may be overlaying pixel by pixel.

The beneficial effects of the present invention are: for subtitle images with a simple background, the compressed file generated by the run-length coding technique can save 90% to 98% of storage space, greatly saving the storage space of subtitle image files, thereby allowing a large number of subtitles to be stored. files, and significantly reduces the limitations of subtitle image files in editing, playback and network transmission.

Description of drawings

Figure 1 is a flow chart of the image subtitle overlay technology

Figure 2 is a flow chart of run-length coding and compressing the original subtitle image in RGB format

Figure 3 is the subtitle compression file format in RGB format

Figure 4 Subtitle compression file format in YUV format

Figure 5. Flowchart of superimposing subtitles with video after using run-length coding subtitle compression technology

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and examples of implementation.

A method for synthesizing subtitles and background images of a high-definition digital television program of the present invention comprises the following steps: 1) using a subtitle compression module, after the original subtitle image is input into the subtitle compression module, using run-length coding technology to compress the image, and generating subtitle compression in a specific format file; 2) adopt the subtitle decompression module, after the subtitle decompression module receives the subtitle compressed file, decompress and obtain the original subtitle image; 3) adopt the subtitle synthesis module, receive the subtitle image obtained by decompression, according to each The alpha value of the pixel to overlay the subtitle image on top of the background image.

Wherein, the formats of the original subtitle image and the generated subtitle compressed file are consistent, which may be RGB format or YUV format.

Fig. 2 is a flow chart of run-length coding and compressing the original subtitle image in RGB format, and the original subtitle image is sent to the buffer pixel by pixel in the order of progressive scanning from bottom to top. The compression module saves the R, G, B, and Alpha values of the first pixel into a register, and sets the counter value to one, and then compares the R, G, B, Alpha values of the next pixel with the R, G saved in the register , B, Alpha value. If the exact correspondence is equal, add one to the counter value, and then take the next pixel sent to the buffer, and compare its R, G, B, Alpha value with the R, G, B, Alpha value saved in the register until it meets A pixel that is not exactly the same as the R, G, B, Alpha values held in the register stops, and then the compression module writes the R, G, B, Alpha values in the registers and the value of the counter to the compressed file. The compression module saves the R, G, B, and Alpha values of the pixel at the current position into the register, and sets the counter value to one. If the above-mentioned process of searching for non-identical pixels does not stop and the value of the counter reaches the upper limit of 65535, then write the R, G, B, Alpha values in the register and the value of the counter into the compressed file without changing the original saved R, G, B. Alpha value, set the counter to one. Continue to search for non-identical pixels, and repeat the above process until all pixels of the original subtitle image are processed.

Fig. 3 is the compressed file format generated after the subtitle image in RGB format is compressed. The beginning is the length of the original subtitle image file and the length of the compressed file, both of which are four bytes long, and the decompression module needs to use these two lengths. Thereafter, the R, G, B, and Alpha values of the pixels and the number of the same pixels are arranged in sequence, wherein the R, G, B, and Alpha values are all one Byte long, and the number of the same pixels is two Byte long.

The method of using run-length coding to compress the original subtitle image in YUV format is basically the same as that in RGB format. The main difference is that when processing subtitle video data, two pixels are obtained continuously, and the counter counting method is adjusted accordingly. Therefore, the run length is not drawn here. Flow chart of encoding and compressing raw subtitle images in YUV format. Fig. 4 is the compressed file format generated after the subtitle image in YUV format is compressed. The beginning is the length of the original subtitle image file and the length of the compressed file, both of which are four bytes long, and the decompression module needs to use these two lengths. Afterwards, the Y1, Y2, U, V, Alpha values of the pixels and the number of the same pixels are arranged in sequence, wherein the Y1, Y2, U, V, and Alpha values are all one Byte long, and the same pixel number is two Byte long.

It can be seen from the above description that the subtitle image with a simple background can be effectively compressed by using the run-length coding compressed subtitle technology, saving storage space.

As shown in Figure 5, before subtitles are used, the compressed subtitle file generated by compression passes through the hardware decompression module to generate the original subtitle image. Then the subtitle image is sent to the synthesis module, and the subtitle is superimposed on the background image by the synthesis module. The synthesis module receives the decompressed subtitle image and background image, and according to the Alpha value of each pixel in the subtitle image, superimposes the pixels in the subtitle image on the corresponding pixels of the background image with a specific superimposition formula, which results in the superimposed video frames. If the Alpha value of the subtitle image pixel is 0, this pixel will not be replaced by the corresponding position of the background image; if the Alpha value of the subtitle image pixel is 1, the RGB value of this pixel will be replaced by the corresponding position pixel of the background image, thus completing HD Synthesis of digital TV program subtitles and background images.

Claims (6)

1. the synthetic method of digital TV in high resolution program captions and background image, it may further comprise the steps:

1) adopts captions to press and compile module, behind the original captions image input captions compression module, use Run-Length Coding technique compresses image, generate the captions compressed file of specific format;

2) adopt the captions decompression module, after the captions decompression module receives the captions compressed file, obtain original captions image by runs decoding technology decompress(ion);

3) adopt the captions synthesis module, receive the captions image that decompresses and obtain, according to the Alpha value of each pixel in the captions image with the captions image overlay to background image.

2. the synthetic method of digital TV in high resolution program captions according to claim 1 and background image is characterized in that: the captions compressed file of described original captions image and generation is rgb format.

3. the synthetic method of digital TV in high resolution program captions according to claim 1 and background image is characterized in that: the captions compressed file of described original captions image and generation is yuv format.

4. the synthetic method of digital TV in high resolution program captions according to claim 1 and background image is characterized in that: described captions compression module is realized the original captions image file that serial received is preserved with 32 BMP forms by software.

5. the synthetic method of digital TV in high resolution program captions according to claim 4 and background image is characterized in that: described captions decompression module is by hardware serial received captions compressed file data.

6. the synthetic method of digital TV in high resolution program captions according to claim 5 and background image is characterized in that: the mode of described captions image overlay is for superposeing by pixel.

Images