4K to 8K video splicing method, device and computer-readable storage medium

Abstract

A 4K to 8K video stitching method, the method comprising: arranging the images of 4 channels of video in the order of top, bottom, left and right, namely, the upper left video area, the upper right video area, the lower left video area, and the lower right video area, the resolution of the 4 channels of video being 4K; establishing a coordinate system with the center point of the screen of the 4 channels of video as the origin, matching R, G, B values, and forming multiple groups of feature blocks; extracting coordinate points within the feature blocks; judging whether there is a repeated area according to the coordinate points according to preset rules; when there is a repeated area, calculating the boundary line of the repeated area; performing edge processing on the boundary of the repeated area; calculating the proportion of the repeated area, and using the interpolation method to compensate the pixels of the 4 channels of video images; stitching the 4 channels of 4K videos after compensating the pixels into an 8K video. The present invention also provides a device and a computer-readable storage medium. The video stitching method of the present invention has simple steps and is easier to operate.

Description

Technical Field

[0001] Embodiments of the present invention relate to a 4K to 8K video stitching technology, and more particularly to a 4K to 8K video stitching method, device, and computer-readable storage medium. Background Art

[0002] Currently, 8K TVs are not widely available due to the high price, large video source storage, high source pixel, and difficulty in supporting photographic equipment. Now, 4K video signal sources can be processed and converted into 8K video signal output for display to meet consumers' visual enjoyment of ultra-high-definition TVs. However, the existing splicing technology has complex steps and redundant algorithms. Summary of the invention

[0003] In view of the above, it is necessary to provide a 4K to 8K video stitching method, device and computer-readable storage medium with simple steps and easy operation.

[0004] An embodiment of the present invention provides a 4K to 8K video stitching method, the method comprising: arranging images of 4 channels of video in up, down, left and right order, namely an upper left video area, an upper right video area, a lower left video area, and a lower right video area, wherein the resolution of the 4 channels of video is 4K; establishing a coordinate system with the center point of the screen of the 4 channels of video as the origin, matching R, G, and B values, and forming multiple groups of feature blocks; extracting coordinate points within the feature blocks; judging whether there is a repeated area according to the coordinate points according to preset rules; when there is a repeated area, calculating the boundary line of the repeated area; performing edge processing on the boundary of the repeated area; calculating the proportion of the repeated area, and compensating pixels for the 4 channels of video images by interpolation; and stitching the 4 channels of 4K videos after compensating the pixels into an 8K video.

[0005] Optionally, judging whether there is a repeated area according to the coordinate points according to preset rules includes: matching left and right video areas, screening reference points of the left and right video areas, and using continuous linear vector matching to prove the existence of the repeated area.

[0006] Optionally, judging whether there is a repeated area according to the coordinate points according to preset rules further includes: matching upper and lower video areas, screening reference points of the upper and lower video areas, and using continuous linear vector matching to prove the existence of the repeated area.

[0007] Optionally, the calculation of the proportion of the repeated area and the use of interpolation method to compensate pixels for the four video images include: calculating the pixels to be compensated according to Formula 1, wherein Formula 1 is: (Z1 / 2+Z2 / 2-Z0 / 2)*8K / (A+B+C+D); Z1 is the total pixel points of the repeated area of ​​the upper and lower videos, Z2 is the total pixel points of the repeated area of ​​the left and right videos, Z0 is the total pixel points of the repeated area of ​​the four videos, and K is the resolution; the upper video includes the upper left video area and the upper right video area, the lower video includes the lower left video area and the lower right video area, the left video includes the upper left video area and the lower left video area, the right-left video includes the upper right video area and the lower right video area, and A, B, C, and D are the upper right video area, the upper left video area, the lower left video area, and the lower right video area, respectively.

[0008] An embodiment of the present invention also provides a device, which includes a memory, a processor, and a 4K to 8K video stitching program stored in the memory and executable on the processor, wherein the 4K to 8K video stitching program implements the following steps when executed by the processor: arranging the images of 4 channels of video in up, down, left, and right directions, namely, an upper left video area, an upper right video area, a lower left video area, and a lower right video area, wherein the resolution of the 4 channels of video is 4K; establishing a coordinate system with the center point of the screen of the 4 channels of video as the origin, matching R, G, and B values, and forming multiple groups of feature blocks; extracting coordinate points within the feature blocks; judging whether there is a repeated area according to the coordinate points according to preset rules; when there is a repeated area, calculating the boundary line of the repeated area; performing edge processing on the boundary of the repeated area; calculating the proportion of the repeated area, and using interpolation to compensate for the pixel points of the 4 channels of video images; and stitching the 4 channels of 4K videos after the pixel points are compensated into an 8K video.

[0009] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the 4K to 8K video splicing method as described above are implemented.

[0010] Compared with the prior art, the 4K to 8K video stitching method, device and computer-readable storage medium can utilize the concept of defining reference points in feature blocks, improve the efficiency of searching for overlapping areas by optimizing the selection of reference points, optimize the range of searching for duplicate areas by comparing the vector relationship of coordinate points in similar feature blocks of two adjacent video areas, and use a continuous linear vector comparison method to determine the image duplicate areas of adjacent video areas, so as to improve the efficiency of image stitching processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a diagram of the operating environment of the device of the preferred embodiment of the present invention.

[0012] Figure 2It is a program module diagram of a preferred embodiment of the 4K to 8K video splicing system of the device of the present invention.

[0013] Figure 3 It is a feature block extraction diagram of each video in the coordinate system of a preferred embodiment of the present invention.

[0014] Figure 4 It is a left and right video vector matching diagram of a preferred embodiment of the present invention.

[0015] Figure 5 It is an upper and lower video vector matching diagram of a preferred embodiment of the present invention.

[0016] Figure 6 It is a left and right video continuous linear vector matching diagram of a preferred embodiment of the present invention.

[0017] Figure 7 It is an upper and lower video continuous linear vector matching diagram of a preferred embodiment of the present invention.

[0018] Figure 8 This is a flowchart of a 4K to 8K video stitching method according to a preferred embodiment of the present invention.

[0019] Main component symbols

[0020] Device 1 4K to 8K video splicing system 10 Memory 20 processor 30 Arrange Module 101 Coordinate creation module 102 Extraction module 103 Judgment module 104 Computing Module 105 Processing Module 106 Splicing module 107 step S300~S314 DETAILED DESCRIPTION

[0021] See also Figure 1 , which is a diagram of the operating environment of a preferred embodiment of the device 1 according to the embodiment of the present invention. The device 1 includes a running 4K to 8K video splicing system 10. The device 1 also includes a memory 20 and a processor 30, etc.

[0022] The memory 20 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. The processor 30 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips, etc.

[0023] See also Figure 2 As shown, it is a program module diagram of a preferred embodiment of the 4K to 8K video splicing system 10 of the present invention.

[0024] The 4K to 8K video stitching system 10 includes an arrangement module 101, a coordinate establishment module 102, an extraction module 103, a judgment module 104, a calculation module 105, a processing module 106 and a stitching module 107. The modules are configured to be executed by one or more processors (a processor 30 in this embodiment) to complete the present invention. The module referred to in the present invention is a computer program segment that completes a specific instruction. The memory 20 is used to store program code and other information of the 4K to 8K video stitching system 10. The processor 30 is used to execute the program code stored in the memory 20.

[0025] The arrangement module 101 is used to arrange the images of the four channels of video in up, down, left and right order, namely an upper left video area, an upper right video area, a lower left video area, and a lower right video area. The resolution of the four channels of video is 4K.

[0026] The coordinate establishment module 102 is used to establish a coordinate system with the center point of the four video images as the origin, match the R, G, B values, and form multiple sets of feature blocks.

[0027] like Figure 3 As shown, there are consistent feature blocks (A / A1 / A2 / A3, B / B1 / B2 / B3, C / C1 / C2 / C3, etc.). The features corresponding to the feature blocks are set by the developer. For example, it can be flowers. If blocks containing flowers are found in the 4 videos, they are consistent feature blocks.

[0028] The extraction module 103 is used to extract coordinate points in the feature block.

[0029] For example, select the coordinate points in a specific block (a / a1 / a2 / a3, b / b1 / b2 / b3, c / c1 / c2 / c3), etc. Take point a as an example, the coordinates are (Xa, Ya).

[0030] The judgment module 104 is used to judge whether there is a repeated area according to the coordinate points according to a preset rule.

[0031] Specifically, the judgment module 104 matches the left and right video areas, selects the reference points of the left and right video areas, uses continuous linear vector matching to prove the existence of the repeated area, matches the upper and lower video areas, selects the reference points of the upper and lower video areas, and uses continuous linear vector matching to prove the existence of the repeated area. The upper video includes the upper left video area and the upper right video area, and the lower video includes the lower left video area and the lower right video area.

[0032] For example, the judgment module 104 matches the left and right video areas, and marks the points of the special blocks that have only unique corresponding relationships, and the set of these points is recorded as Xin (i = a, b, c, d, ...; n = 1, 2). At the same time, the points of the special blocks in the right video area that have two corresponding relationships in the left video area are marked, and the set of these points is recorded as Xin (i = g, h, k, m ...; n = 1, 2, 3). And so on Xin (i = g, h, k, m ...; n = 1, 2, 3 ...). First, compare the points with the smallest absolute values ​​of the left and right video areas Xin (i = a, b, c, d, e, f ...; n = 1, 2), such as Figure 4 It can be seen that for the right video area, |Xa1|<|Xb1|<|Xc1|<|Xd1|<|Xe1|<|Xf1|, for the left video area |Xd2|<|Xc2|<|Xb2|<|Xa2|<|Xf2|<|Xe2|, select the right video area |Xin| minimum value as the vector matching point, and compare it to point a1 from the above; select the left video area |Xin| minimum value as the vector matching point, and compare it to point d2 from the above, and compare whether the vectors of the two selected points and the corresponding relationship points are equal. If they are equal (for example, ), continuous linear vector matching is used to prove the existence of the duplicate region. If they are not equal, the existence of the duplicate region is proved by the points of two corresponding special blocks, then Xin (i = g, h, k, m...; n = 1, 2, 3) is selected and compared and the minimum point of |Xin| in the left and right video regions is selected as the vector matching point.

[0033] Similarly, the judgment module 104 matches the upper and lower video areas, and selects the points of the special blocks that have only unique corresponding relationships, and the set of these points is recorded as Yin (i = a, b, c, d...; n = 1, 2). The points of the special blocks in the lower video area that have two corresponding relationships in the upper video area are selected, and the set of these points is recorded as Yin (i = g, h, k, m...; n = 1, 2, 3). And so on Yin (i = g, h, k, m...; n = 1, 2, 3...); compare the points with the smallest absolute values ​​of the upper and lower video areas Yin (i = a, b, c, d...; n = 1, 2), as shown below Figure 5 As shown, for the lower video area, |Ya1|<|Yb1|<|Yc1|<|Yd1|<|Ye1|<|Yf1|; for the upper video area |Yd2|<|Yc2|<|Yb2|<|Ya2|<|Yf2|<|Ye2|, the minimum value of |Yin| in the lower video area is selected as the vector matching point, and the point a1 is compared from the top; the minimum value of |Yin| in the upper video area is selected as the vector matching point, and the point d2 is compared from the top, and the vectors of the two selected points and the corresponding relationship points are compared to see if they are equal. If they are equal, The existence of the repeated area is proved by continuous linear vector matching. If they are not equal, the existence of the repeated area is proved by the points of two corresponding special blocks, then Yin (i = g, h, k, m...; n = 1, 2, 3) is selected and the minimum point of |Yin| in the upper and lower video areas is compared and selected as the vector matching point.

[0034] The calculation module 105 is used to calculate the boundary line of the repeated area when there is a repeated area.

[0035] like Figure 6 As shown in the figure, continuous linear vector matching is used to prove the existence of the left and right video repetition area: assuming a as the reference point, a continuous linear scan is performed with the distance from the reference point a to each point as the radius to match the corresponding vector relationship of the left and right video areas. If the vectors are equal, it proves that the scanned left and right video areas are repetition areas. When the scanned vectors are not equal, it proves that the straight line where the vector is located is the boundary of the repetition area. Among them, the scanning range of each point: the corresponding point closest to the Y axis in the left and right video areas is used as the scanning starting point, and only the points in this area are scanned.

[0036] like Figure 7 As shown in the figure, continuous linear vector matching is used to prove the existence of the repetitive area of ​​the upper and lower videos: assuming a as the reference point, a continuous linear scan is performed with the distance from the reference point a to each point as the radius to match the corresponding vector relationship of the upper and lower video areas. If the vectors are equal, it proves that the scanned left and right video areas are repetitive areas. When the scanned vectors are not equal, it proves that the straight line where the vector is located is the boundary of the repetitive area. Among them, the scanning range of each point: the corresponding point closest to the X-axis in the upper and lower video areas is used as the scanning starting point, and only the points in this area are scanned.

[0037] The processing module 106 is configured to perform edge processing on the boundary of the repeated region.

[0038] In order to eliminate the edge distortion of the adjacent video areas, the edge pixels of the duplicate area to be retained are replaced by the pixels of the duplicate area not to be retained, thereby eliminating the edge distortion problem of image fusion.

[0039] The calculation module 105 is also used to calculate the proportion of the repeated area and compensate the pixels of the four video images by using the interpolation method.

[0040] The pixels that need to be compensated are calculated according to Formula 1, where Formula 1 is:

[0041] (Z1 / 2+Z2 / 2-Z0 / 2)*8K / (A+B+C+D)

[0042] Z1 is the total pixel points of the repeated area of ​​the upper and lower videos, Z2 is the total pixel points of the repeated area of ​​the left and right videos, Z0 is the total pixel points of the repeated area of ​​the 4 videos, and K is the resolution; the upper video includes the upper left video area and the upper right video area, the lower video includes the lower left video area and the lower right video area, the left video includes the upper left video area and the lower left video area, the right-left video includes the upper right video area and the lower right video area, and A, B, C, and D are the pixel points of the upper right video area, the upper left video area, the lower left video area, and the lower right video area respectively.

[0043] The stitching module 107 is used to stitch the four channels of 4K videos after pixel compensation into an 8K video.

[0044] In this embodiment, the concept of defining reference points within feature blocks can be utilized to improve the efficiency of searching for overlapping areas by optimizing the selection of reference points. The scope of searching for repeated areas is optimized by comparing the vector relationship between coordinate points within similar feature blocks of two adjacent video areas. The image repeated areas of adjacent video areas are determined using a continuous linear vector comparison method to improve the efficiency of image stitching processing. In order to eliminate edge distortion of adjacent video areas, the corresponding edge pixel points of the repeated area to be retained are replaced by pixel points of the repeated area that are not retained, thereby eliminating the edge distortion problem of image fusion.

[0045] See also Figure 8 , which is a flow chart of a 4K to 8K video stitching method according to a preferred embodiment of the present invention. The 4K to 8K video stitching method is applied to the device 1. The 4K to 8K video stitching method can be executed by the processor 30 Figure 2 The module 101-107 shown is implemented.

[0046] Step S300, arranging the images of the four channels of video in the order of top, bottom, left and right, namely an upper left video area, an upper right video area, a lower left video area, and a lower right video area, and the resolution of the four channels of video is 4K.

[0047] Step S302 , establishing a coordinate system with the center point of the four video images as the origin, matching R, G, B values, and forming multiple groups of feature blocks.

[0048] like Figure 3 As shown, there are consistent feature blocks (A / A1 / A2 / A3, B / B1 / B2 / B3, C / C1 / C2 / C3, etc.). The features corresponding to the feature blocks are set by the developer. For example, it can be flowers. If blocks containing flowers are found in the 4 videos, they are consistent feature blocks.

[0049] Step S304, extracting coordinate points within the feature block.

[0050] For example, select the coordinate points in a specific block (a / a1 / a2 / a3, b / b1 / b2 / b3, c / c1 / c2 / c3), etc. Take point a as an example, the coordinates are (Xa, Ya).

[0051] Step S306: determining whether there is a repeated area according to the coordinate points and a preset rule.

[0052] Specifically, the device 1 matches the left and right video areas, selects the reference points of the left and right video areas, uses continuous linear vector matching to prove the existence of the repeated area, matches the upper and lower video areas, selects the reference points of the upper and lower video areas, and uses continuous linear vector matching to prove the existence of the repeated area. The upper video includes the upper left video area and the upper right video area, and the lower video includes the lower left video area and the lower right video area.

[0053] For example, device 1 matches the left and right video areas, marks the points of the special blocks that have only unique corresponding relationships, and the set of these points is recorded as Xin (i = a, b, c, d, ...; n = 1, 2). At the same time, the points of the special blocks in the right video area that have two corresponding relationships in the left video area are marked, and the set of these points is recorded as Xin (i = g, h, k, m ...; n = 1, 2, 3). And so on Xin (i = g, h, k, m ...; n = 1, 2, 3 ...). First, compare the points with the smallest absolute values ​​of the left and right video areas Xin (i = a, b, c, d, e, f ...; n = 1, 2), such as Figure 4 It can be seen that for the right video area, |Xa1|<|Xb1|<|Xc1|<|Xd1|<|Xe1|<|Xf1|, for the left video area |Xd2|<|Xc2|<|Xb2|<|Xa2|<|Xf2|<|Xe2|, select the right video area |Xin| minimum value as the vector matching point, and compare it to point a1 from the above; select the left video area |Xin| minimum value as the vector matching point, and compare it to point d2 from the above, and compare whether the vectors of the two selected points and the corresponding relationship points are equal. If they are equal (for example, ), continuous linear vector matching is used to prove the existence of the duplicate region. If they are not equal, the existence of the duplicate region is proved by the points of two corresponding special blocks, then Xin (i = g, h, k, m...; n = 1, 2, 3) is selected and compared and the minimum point of |Xin| in the left and right video regions is selected as the vector matching point.

[0054] Similarly, device 1 matches the upper and lower video areas, and selects the points of the special blocks that have only unique corresponding relationships, and the set of these points is recorded as Yin (i = a, b, c, d...; n = 1, 2). Select the points of the special blocks in the lower video area that have two corresponding relationships in the upper video area, and the set of these points is recorded as Yin (i = g, h, k, m...; n = 1, 2, 3). And so on Yin (i = g, h, k, m...; n = 1, 2, 3...); compare the points with the smallest absolute values ​​of the upper and lower video areas Yin (i = a, b, c, d...; n = 1, 2), as shown below Figure 5 As shown, for the lower video area, |Ya1|<|Yb1|<|Yc1|<|Yd1|<|Ye1|<|Yf1|; for the upper video area |Yd2|<|Yc2|<|Yb2|<|Ya2|<|Yf2|<|Ye2|, the minimum value of |Yin| in the lower video area is selected as the vector matching point, and the point a1 is compared from the top; the minimum value of |Yin| in the upper video area is selected as the vector matching point, and the point d2 is compared from the top, and the vectors of the two selected points and the corresponding relationship points are compared to see if they are equal. If they are equal, The existence of the repeated area is proved by continuous linear vector matching. If they are not equal, the existence of the repeated area is proved by the points of two corresponding special blocks, then Yin (i = g, h, k, m...; n = 1, 2, 3) is selected and the minimum point of |Yin| in the upper and lower video areas is compared and selected as the vector matching point.

[0055] Step S308: When there is a repeated area, calculate the boundary line of the repeated area.

[0056] like Figure 6 As shown in the figure, continuous linear vector matching is used to prove the existence of the left and right video repetition area: assuming a as the reference point, a continuous linear scan is performed with the distance from the reference point a to each point as the radius to match the corresponding vector relationship of the left and right video areas. If the vectors are equal, it proves that the scanned left and right video areas are repetition areas. When the scanned vectors are not equal, it proves that the straight line where the vector is located is the boundary of the repetition area. Among them, the scanning range of each point: the corresponding point closest to the Y axis in the left and right video areas is used as the scanning starting point, and only the points in this area are scanned.

[0057] like Figure 7 As shown in the figure, continuous linear vector matching is used to prove the existence of the repetitive area of ​​the upper and lower videos: assuming a as the reference point, a continuous linear scan is performed with the distance from the reference point a to each point as the radius to match the corresponding vector relationship of the upper and lower video areas. If the vectors are equal, it proves that the scanned left and right video areas are repetitive areas. When the scanned vectors are not equal, it proves that the straight line where the vector is located is the boundary of the repetitive area. Among them, the scanning range of each point: the corresponding point closest to the X-axis in the upper and lower video areas is used as the scanning starting point, and only the points in this area are scanned.

[0058] Step S310: performing edge processing on the boundary of the repeated area.

[0059] In order to eliminate the edge distortion of the adjacent video areas, the edge pixels of the duplicate area to be retained are replaced by the pixels of the duplicate area not to be retained, thereby eliminating the edge distortion problem of image fusion.

[0060] Step S312, calculating the proportion of the repeated area, and using the interpolation method to compensate the pixels of the four video images.

[0061] The pixels that need to be compensated are calculated according to Formula 1, where Formula 1 is:

[0062] (Z1 / 2+Z2 / 2-Z0 / 2)*8K / (A+B+C+D)

[0063] Z1 is the total pixel points of the repeated area of ​​the upper and lower videos, Z2 is the total pixel points of the repeated area of ​​the left and right videos, Z0 is the total pixel points of the repeated area of ​​the 4 videos, and K is the resolution; the upper video includes the upper left video area and the upper right video area, the lower video includes the lower left video area and the lower right video area, the left video includes the upper left video area and the lower left video area, the right-left video includes the upper right video area and the lower right video area, and A, B, C, and D are the pixel points of the upper right video area, the upper left video area, the lower left video area, and the lower right video area respectively.

[0064] Step S314, stitching the four channels of 4K videos after pixel compensation into an 8K video.

[0065] By applying the above method to the above device, the concept of defining reference points within the feature block can be utilized to improve the efficiency of searching for overlapping areas by optimizing the selection of reference points. The range of searching for repeated areas can be optimized by comparing the vector relationship of coordinate points within similar special blocks of two adjacent video areas. The image repeated areas of adjacent video areas are determined by a continuous linear vector comparison method to improve the efficiency of image stitching processing. In order to eliminate the edge distortion of the splicing of adjacent video areas, the corresponding edge pixel points of the repeated area to be retained are replaced by the pixel points of the repeated area that are not retained, thereby eliminating the edge distortion problem of image fusion.

[0066] It is worth noting that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A 4K to 8K video splicing method, characterized in that: The method comprises: The images of the four channels of video are arranged in up, down, left and right order, namely an upper left video area, an upper right video area, a lower left video area, and a lower right video area, and the resolution of the four channels of video is 4K; A coordinate system is established with the center point of the four video images as the origin, and R, G, and B values ​​are matched to form multiple groups of feature blocks; Extract the coordinate points within the feature block; Determine whether there is a repeated area according to the coordinate points according to a preset rule; When there is a repeated area, calculating the boundary line of the repeated area; Performing edge processing on the boundary of the repeated area; Calculate the proportion of repeated areas and use interpolation to compensate pixels for the 4-channel video images; After pixel compensation, four channels of 4K videos are stitched together into an 8K video. The step of calculating the proportion of the repeated area and compensating pixels for the four-channel video images by using an interpolation method includes: The pixels that need to be compensated are calculated according to Formula 1, where Formula 1 is: (Z1 / 2+Z2 / 2-Z0 / 2)*8K / (A+B+C+D); Among them, Z1 is the total pixel points of the repeated area of ​​the upper and lower videos, Z2 is the total pixel points of the repeated area of ​​the left and right videos, Z0 is the total pixel points of the repeated area of ​​the four videos, and K is the resolution; the upper video includes the upper left video area and the upper right video area, the lower video includes the lower left video area and the lower right video area, the left video includes the upper left video area and the lower left video area, and the right-left video includes the upper right video area and the lower right video area. A, B, C, and D are the pixel points of the upper right video area, the upper left video area, the lower left video area, and the lower right video area, respectively.

2. The 4K to 8K video splicing method according to claim 1, characterized in that: The determining whether there is a repeated area according to the coordinate points according to a preset rule includes: Match the left and right video areas, select the reference points of the left and right video areas, and use continuous linear vector matching to determine the existence of repeated areas, wherein the left video includes the upper left video area and the lower left video area, and the right video includes the upper right video area and the lower right video area.

3. The 4K to 8K video splicing method according to claim 2, characterized in that: The step of determining whether there is a repeated area according to the coordinate points according to a preset rule also includes: Match the upper and lower video areas, filter the reference points of the upper and lower video areas, and use continuous linear vector matching to determine the existence of repeated areas, wherein the upper video includes the upper left video area and the upper right video area, and the lower video includes the lower left video area and the lower right video area.

4. A device, characterized in that: The device includes a memory, a processor, and a 4K to 8K video splicing program stored in the memory and executable on the processor. When the 4K to 8K video splicing program is executed by the processor, the following steps are implemented: The images of the four channels of video are arranged in up, down, left and right order, namely an upper left video area, an upper right video area, a lower left video area, and a lower right video area, and the resolution of the four channels of video is 4K; A coordinate system is established with the center point of the four video images as the origin, and R, G, and B values ​​are matched to form multiple groups of feature blocks; Extract the coordinate points within the feature block; Determine whether there is a repeated area according to the coordinate points according to a preset rule; When there is a repeated area, calculating the boundary line of the repeated area; Performing edge processing on the boundary of the repeated area; Calculate the proportion of repeated areas and use interpolation to compensate pixels for the 4-channel video images; After pixel compensation, four channels of 4K videos are stitched together into an 8K video. The step of calculating the proportion of the repeated area and compensating pixels for the four-channel video images by using an interpolation method includes: The pixels that need to be compensated are calculated according to Formula 1, where Formula 1 is: (Z1 / 2+Z2 / 2-Z0 / 2)*8K / (A+B+C+D); Among them, Z1 is the total pixel points of the repeated area of ​​the upper and lower videos, Z2 is the total pixel points of the repeated area of ​​the left and right videos, Z0 is the total pixel points of the repeated area of ​​the four videos, and K is the resolution; the upper video includes the upper left video area and the upper right video area, the lower video includes the lower left video area and the lower right video area, the left video includes the upper left video area and the lower left video area, and the right-left video includes the upper right video area and the lower right video area. A, B, C, and D are the pixel points of the upper right video area, the upper left video area, the lower left video area, and the lower right video area, respectively.

5. The device according to claim 4, characterized in that The determining whether there is a repeated area according to the coordinate points according to a preset rule includes: Match the left and right video areas, select the reference points of the left and right video areas, and use continuous linear vector matching to determine the existence of repeated areas, wherein the left video includes the upper left video area and the lower left video area, and the right video includes the upper right video area and the lower right video area.

6. The device according to claim 4, characterized in that The step of determining whether there is a repeated area according to the coordinate points according to a preset rule also includes: Match the upper and lower video areas, filter the reference points of the upper and lower video areas, and use continuous linear vector matching to determine the existence of repeated areas, wherein the upper video includes the upper left video area and the upper right video area, and the lower video includes the lower left video area and the lower right video area.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the 4K to 8K video stitching method according to any one of claims 1 to 3 are implemented.

Images