CN107360384B - High-speed video acquisition and transmission method - Google Patents

Abstract

The invention provides a high-speed video acquisition and transmission method, comprising the following steps: 1) acquiring a high-speed image in a non-standard format; the high-speed image in the non-standard format is 640*480@K ₁ fps, and the high-speed image format is YCbCr; wherein, K ₁ = 200 or 400; 2) Map the high-speed image in a non-standard format into high-definition video data in a standard format according to the set mapping rules; the high-definition video data in the standard format is 1920*1080@K ₂ fps, and the high-definition image format is YCbCr; wherein: K ₂ =30 or 60; 3) restore the high-definition video data of the standard format to the high-speed image data of the non-standard format according to the set demapping rule; 4) will restore the high-speed image of the non-standard format The data is compressed and encoded by HEVC/H265; 5) The compressed code stream of the high-speed image in the non-standard format is output. Through this method, the non-standard format high-speed image long-distance lossless high-speed transmission is realized.

Description

A high-speed video capture and transmission method

technical field

The invention belongs to the technical field of information data transmission, and in particular relates to a high-speed video acquisition and transmission method.

Background technique

In some special occasions or some specialized fields, it is necessary to use special high-speed cameras to shoot and record the process. For example, when shooting fast-changing scenes such as the moment of the explosion of an object, the image data needs to be transmitted to the back-end receiving equipment over a long distance. For reception and data processing. The image formats output by these high-speed cameras are usually non-standard, not common SD, HD or UHD standard formats, and cannot be transmitted by common digital video transmitters and receivers.

When the digital camera outputs images, the commonly used interface forms include USB, 1394 (fire wire), GE (Ethernet), LVDS, CameraLink, etc. Among them, the USB interface is particularly suitable for connecting the camera directly to the computer display, which can support various resolutions and frame rates, and is often used in various standard-definition and high-definition cameras. The 1394 interface has a high transmission rate, occupies less CPU resources, and works stably, and is also widely used in digital cameras. Gigabit Ethernet (GE) has high transfer rates, is inexpensive, and is used in a variety of resolutions and frame rates. The CameraLink interface is the fastest bus type in current industrial cameras, and is generally used in high-resolution, high-speed area scan cameras or line scan cameras.

The speed of the USB2.0 interface is low, and the CPU needs to be involved in the management during the transmission process, which occupies and consumes a large amount of resources. The transmission distance is short, and the signal is easily attenuated. The speed of USB 3.0 has been greatly improved, but the shortcomings of requiring CPU processing and short transmission distance remain unsolved.

The 1394 interface technology is relatively complex, and the popularity of the interface is not high, and it has been gradually eliminated by the market.

Gigabit Ethernet requires that both the sending end and the receiving end of the image require the CPU to participate in the high-speed encapsulation/decapsulation processing of the TCP/IP protocol, and the corresponding PHY devices, transformer chips, and RJ45 interfaces need to be connected, and the volume is relatively large.

The LVDS bus interface cannot support high-speed data transmission, and the effective transmission distance is short.

The Camera Link interface is bulky, expensive, and has a short transmission distance.

Therefore, the technical problem that needs to be solved urgently now is how to use a simple and low-cost device to transmit high-speed video images from a remote high-speed camera to a back-end receiving device for processing.

SUMMARY OF THE INVENTION

In order to solve the problems in the background technology, the present invention relates to a high-speed video acquisition and transmission method, which realizes long-distance lossless high-speed transmission of non-standard high-speed video images through format conversion.

The concrete principle of the present invention is:

The present invention aims to provide a high-speed video acquisition and transmission method that can support long-distance transmission. The method of mapping non-standard format video to standard video format and then transmitting, converts 640*480@K ₁ fps high-speed video Map to the 1920*1080p@K ₂ fps high-definition video format that conforms to the BT.1120 standard, and then demap the data in the high-definition video format to restore 640*480@K ₁ fps high-speed video data and then encode and output.

The concrete technical scheme of the present invention is:

The present invention provides a high-speed video acquisition and transmission method, which is characterized in that it includes the following steps:

1) Collect high-speed images in non-standard formats;

The high-speed image in the non-standard format is 640*480@K ₁ fps, and the high-speed image format is YCbCr; wherein, K ₁ =200 or 400;

2) the high-speed image of non-standard format is mapped into the high-definition video data of standard format according to the mapping rule of setting; The high-definition video data of described standard format is 1920*1080@K ₂ fps, and the image format is YCbCr; Wherein: K ₂ =30 or 60;

3) restore the high-definition video data of the standard format to the high-speed image data of the non-standard format according to the set demapping rule;

4) performing HEVC/H265 compression coding on the recovered high-speed image data in a non-standard format;

5) Output the compressed code stream of high-speed image in non-standard format.

The mapping rules set in the above step 2) are specifically:

A1. Define the transmission format of K ₂ frames of high-definition video images conforming to the BT.1120 standard. In the transmission format of each frame of standard high-definition video images, an effective image interval is set, and the effective image interval is used to fill high-speed image data. The valid image interval has a total of 1920 columns and 1080 rows, that is, 2073600 bytes;

B1. Insert the sequence number byte 0 into the first byte of the valid image interval of the first frame of high-definition video format, and then continuously embed the 1-6 frames of high-speed images, numbered 1 to 6 respectively, and finally embed the seventh frame The first 360 lines of high-speed images are numbered 7;

C1. Insert the serial number byte 1 into the first byte of the valid image interval of the second frame of high-definition video format, then embed the last 120 lines of the seventh frame of high-speed image, and then continuously embed the 8-13th frame of high-speed image, numbered as 8 to 13, and finally the first 240 lines of the 14th high-speed image are embedded and numbered 14;

D1. Insert the serial number byte 2 into the first byte of the valid image interval of the 3rd frame of high-definition video format, then embed the last 240 lines of the 14th frame of high-speed image, and then continuously embed the 15th-20th frame of high-speed image, numbered as From 15 to 20, the valid image interval of the 3rd frame of high-definition video format accommodates the redundant bytes after the 20th frame of 640*480 format image. In the effective image range of high-speed images in high-definition video format;

E1. Repeat steps B1 to D1 until the remaining 21st to K1 frames of high _- speed images are loaded into the 4th to K2 columns of high _- definition video format valid image intervals.

The demapping rules set in the above-mentioned step 3) are specifically:

A2. Obtain the data in the valid image interval of the first frame of high-definition video format, and determine the first byte. If the first byte is 0, execute step B2; if the first byte is 1, execute step C2, If the first byte is 2, execute step D2;

B2. Continuously take out the 1st to 6th frames of high-speed images in the valid image interval of the first frame of high-definition video format and directly compress and encode them, and the numbers are 1 to 6 respectively, and then cache the first 360 lines of the 7th frame of high-speed images, and then Return to step A2;

C2. First take out the last 120 lines of the seventh frame of high-speed image in the valid image interval of the second frame of high-definition video format, splicing it with the first 360 lines of the image of the seventh frame of high-speed image cached in step B2, and then perform compression coding, Then continuously take out the 8-13 frames of high-speed images for compression coding, the frame numbers are 8 to 13, and finally cache the first 240 lines of the 14th frame of high-speed images, and finally return to step A2;

D2. First take out the last 240 lines of the 14th high-speed image in the valid image interval of the third frame of high-definition video format, splicing with the first 240 lines of the 14th high-speed image cached in step C2, and then compressing and encoding, and then continuously Take out the 15th to 20th frames of high-speed images to prepare for compression coding, the frame numbers are 15 to 20, discard the remaining bytes, and finally return to step A2;

E2. Repeat steps A2 to D2 until all the data in the valid image interval of the K ₂ frame high-definition video format is received.

The beneficial effects of the present invention are:

The high-speed video transmission method adopted in the present invention can not only maintain the integrity of the original video, but also correctly and completely record the output original data after collecting, mapping, demapping, transmitting and compressing the high-speed images in non-standard formats in the whole process. , while realizing the long-distance lossless transmission of high-speed video data.

Description of drawings

1 is a system frame diagram of the present invention;

2 is a schematic diagram of the first frame high-definition video image mapping;

3 is a schematic diagram of the second frame high-definition video image mapping;

4 is a schematic diagram of the third frame high-definition video image mapping;

Figure 5 is a flow chart of data demapping.

Detailed ways

Referring to FIG. 1, a high-speed video capture and transmission method is provided in this embodiment to describe in more detail the transmission method that needs to be protected in the present invention:

The system includes a data sending end and a receiving end, both of which are equipped with SDI interfaces; the sending end and the receiving end transmit data through SDI coaxial cables;

Among them, the sending end includes a high-speed image sensor, a video acquisition module and a data mapping module integrated on the FPGA chip;

The receiving end includes a data demapping module, a video encoding module and a video output module integrated on the ARM chip;

Wherein: the high-speed image sensor is used to output a high-speed image in a non-standard format; the high-speed image in the non-standard format is 640*480@K ₁ fps, and the high-speed image format is YCbCr; wherein, K ₁ =200 or 400;

The video acquisition module collects the non-standard format high-speed image output by the high-speed image sensor, and then sends it to the data mapping module;

The data mapping module is responsible for mapping the high-speed image of the non-standard format into the high-definition video data of the standard format according to the set mapping rule; The high-definition video data of the standard format is 1920*1080@K ₂ fps, and the image format is YCbCr; Wherein: K ₂ =30 or 60;

The data demapping module restores the standard format high-definition video data to 640*480@K ₁ fps video according to the set demapping rules;

The video encoding module performs HEVC/H265 compression encoding on the recovered 640*480@K ₁ fps video;

The video output module is used to output 640*480@K ₁ fps compressed stream.

The specific steps of the transmission method of the present invention based on the system are as follows:

Step 1) The high-speed graphic sensor outputs non-standard format high-speed image data and sends it to the data mapping module through the video capture module;

The high-speed image in the non-standard format is 640*480@K ₁ fps, and the high-speed image source format is YCbCr; wherein, K ₁ =200 or 400;

Step 2) data mapping module maps the high-speed image of non-standard format into the high-definition video data of standard format according to the set mapping rule; The high-definition video data of described standard format is 1920*1080@K ₂ fps, and the image format is YCbCr; Wherein: K ₂ =30 or 60;

The specific mapping rules are as follows (in this embodiment, K ₁ =400, K ₂ =60):

A1. Define the transmission format of 60 frames of high-definition video images conforming to the BT.1120 standard in the data mapping module, and set a valid image interval in the transmission format of each frame of standard high-definition video images, and the valid image interval is used to fill High-speed image data, the effective image interval has a total of 1920 columns and 1080 rows, that is, 2073600 bytes;

As shown in Figure 2, B1, insert the sequence number byte 0 into the first byte of the valid image interval of the first frame of high-definition video format, and then insert the 1st to 6th frames of high-speed images consecutively, and number 1 to 6 respectively , and finally the first 360 lines of the 7th high-speed image are embedded and numbered 7;

As shown in Figure 3, C1, insert sequence number byte 1 into the first byte of the valid image interval of the second frame of high-definition video format, then embed the last 120 lines of the seventh frame of high-speed image, and then continuously embed the 8-13 Frame high-speed images, numbered 8 to 13, and finally embedded in the first 240 lines of the 14th high-speed image, numbered 14;

As shown in Figure 4, D1, insert the sequence number byte 2 into the first byte of the valid image interval of the 3rd frame of high-definition video format, then embed the last 240 lines of the 14th frame of high-speed image, and then embed the 15th-20th consecutively. The frames of high-speed images are numbered 15 to 20. The valid image interval of the third frame of high-definition video format accommodates the redundant bytes after the 20th frame of 640*480 format images. Load continuous 3 frames of high-speed images in the high-definition video format into the effective image interval;

E1. Repeat steps B1 to D1 until the remaining 21st to 400th frames of high-speed images are loaded into the effective image interval of the 4th to 60th columns of the high-definition video format.

Step 3) The transmitting end transmits the standard format high-definition video data to the receiving end through the coaxial cable;

Step 4) the data demapping module of the receiving end restores the high-speed image data of the non-standard format from the high-definition video data of the standard format according to the set demapping rule;

The demapping rules proposed in this paper will be introduced below with reference to FIG. 5 (K ₁ =400, K ₂ =60 in this embodiment):

A2. The data demapping module obtains the data in the valid image interval of the first frame of high-definition video format, judges the first byte, if the first byte is 0, execute step B2, if the first byte is 1, then Execute step C2, if the first byte is 2, execute step D2;

B2. Continuously take out the 1st to 6th frames of high-speed images in the valid image interval of the first frame of high-definition video format and directly send them to the video coding module, the numbers are 1 to 6 respectively, and then take out the first 360 lines of the 7th frame of high-speed images for caching , and then return to step A2;

C2. First take out the last 120 lines of the seventh frame of high-speed image in the valid image interval of the second frame of high-definition video format, splicing it with the first 360 lines of the image of the seventh frame of high-speed image cached in step B2, and then send it to the video encoding module, and then continuously take out the 8-13 frames of high-speed images and send them to the video coding module, the frame numbers are 8 to 13, and finally cache the first 240 lines of the 14th frame of high-speed images, and finally return to step A2;

D2. First take out the last 240 lines of the 14th frame of high-speed image in the valid image interval of the third frame of high-definition video format, splicing with the first 240 lines of the image of the 14th frame of high-speed image cached in step C2, and then send it to the video coding module, Then continuously take out the 15th to 20th frames of high-speed images and send them to the video coding module, the frame numbers are 15 to 20, discard the remaining bytes, and finally return to step A2;

E2. Repeat steps A2 to D2 until all data in the valid image interval of the 60-frame high-definition video format is received.

Step 5) Perform the restored high-speed image data in a non-standard format

HEVC/H265 compression encoding;

Step 6) Output the compressed code stream of the high-speed image in non-standard format.

Claims (1)

1. a kind of high-speed video collecting and transmitting method, which comprises the following steps:

1) high speed image of noncanonical format is acquired；

The high speed image of the noncanonical format is 640*480@K₁Fps, high speed image format are YCbCr；Wherein, K₁=200 or 400；

2) high speed image of noncanonical format is mapped to the HD video data of reference format by the mapping ruler of setting；

The mapping ruler is specifically:

A1, K is defined₂Frame meets the transformat of the high clear video image of BT.1120 standard, the HD video figure of each frame standard Effective image section is set in the transformat of picture, the effective image section is for filling high speed image data, this is effectively Total 1920 column of image interval and 1080 rows, i.e. 2073600 bytes；

B1, first character section is inserted into serial number byte 0 in the effective image section of the 1st clear video format of vertical frame dimension, then continuously It is embedded in 1-6 frame high speed image, and number is 1~6 respectively, is finally embedded in the preceding 360 row image of the 7th frame high speed image, and compiled Number be 7；

C1, first character section is inserted into serial number byte 1 in the effective image section of the 2nd clear video format of vertical frame dimension, is then inserted into the 7th Rear 120 row of frame high speed image, then it is consecutively embedded 8-13 frame high speed image, number is 8~13, is finally embedded in the 14th vertical frame dimension speed Preceding 240 row of image, and numbering is 14；

D1, first character section is inserted into serial number byte 2 in the effective image section of the 3rd clear video format of vertical frame dimension, is then inserted into the Rear 240 row of 14 frame high speed images, then it is consecutively embedded 15-20 frame high speed image, number is the 15~20, the 3rd frame HD video Extra byte after the 20th frame 640*480 format-pattern of effective image section receiving of format is uniformly filled with 00；So far, just will Continuous 20 vertical frame dimension speed loading images enter in the effective image section of the continuous clear video format of 3 vertical frame dimension；

E1, step B1~D1 is repeated, until by remaining 21~K₁Vertical frame dimension speed loading images are to 4~K₂Column HD video lattice In the effective image section of formula；

The HD video data of the reference format are 1920*1080@K₂Fps, high-definition image format are YCbCr；Wherein: K₂= 30 or 60；

3) the HD video data of reference format are reverted to the high speed image number of noncanonical format by the demapping rule of setting According to；

Demapping rule is specifically:

A2, obtain the 1st clear video format of vertical frame dimension effective image section in data, judge first character section, if first Byte thens follow the steps B2 for 0, if first character section thens follow the steps C2 for 1, if first character section thens follow the steps D2 for 2；

B2, it is continuously withdrawn 1-6 frame high speed image in the effective image section of the 1st clear video format of vertical frame dimension and directly carries out compression volume Code, number are respectively 1~6, are then cached 360 row images before the 7th frame high speed image, then return step A2；

C2, rear 120 row for first taking out the 7th frame high speed image in the effective image section of the 2nd clear video format of vertical frame dimension, by itself and step The preceding 360 row image of the 7th frame high speed image cached in rapid B2 carries out compressed encoding after carrying out sequential concatenation, then continuously takes again 8-13 frame high speed image carries out compressed encoding out, and frame number is 8~13, finally takes out preceding 240 row of the 14th frame high speed image It is cached, finally returns to step A2；

D2, rear 240 row for first taking out the 14th frame high speed image in the effective image section of the 3rd clear video format of vertical frame dimension, with step The preceding 240 row image of the 14th frame high speed image cached in C2 carries out compressed encoding after carrying out sequential concatenation, is then continuously withdrawn 15th~20 frame high speed image carries out compressed encoding, and frame number 15~20 remaining byte will abandon below, and finally return to step A2；

E2, step A2~D2 is repeated, until by K₂Data in the effective image section of the clear video format of vertical frame dimension have all received Finish；

4) high speed image data of the noncanonical format recovered is subjected to HEVC/H265 compressed encoding；

5) compressed bit stream of the high speed image of noncanonical format is exported.