JP6950248B2 - Image transmitter, image receiver, image transmission method and image transmission program - Google Patents

Description

The present invention relates to an image transmission / reception technique for transmitting / receiving an image on a transmission line.

One of the video signal transmission standards is SDI (Serial Digital Interface), and 3 Gbps SDI (“3G-SDI”) is used for transmission of HDTV (High Definition Television) video. A 12 Gbps transmission path is required to transmit a video signal (“4K video”) with four times the number of pixels of an HDTV, but if the 4K video is divided into four and output, the existing 3G-SDI can be converted to four. This can be used for transmission.

In Patent Document 1, in an AV device that transmits and receives a video signal of a video to be displayed on one display panel by dividing it into a plurality of cables, a video indicated by each video signal transmitted by the plurality of cables is specified on the display panel. The technique of sorting so as to be the arrangement of is described.

Japanese Unexamined Patent Publication No. 2013-153410

When transmitting a video signal by dividing it into a plurality of cables, it is necessary to correctly connect a plurality of transmission cables connecting between the output terminal of the transmitting device and the input terminal of the receiving device. If there is a mistake in the connection of the transmission cable, a part of the video will be replaced when the video is combined by the receiving device, and the correct video will not be displayed. On the other hand, when transmitting a video signal on a predetermined transmission line, it is necessary to convert the video into an image size suitable for the transmission line before transmitting the video signal. Therefore, even if the transmitted video is confirmed on the monitor, the image is converted, so that there is a problem that it cannot be visually determined whether or not there is an error in the connection state of the transmission cable.

The present invention has been made in view of such a situation, and an object of the present invention is to provide an image transmission / reception technique capable of determining and correcting a connection state of a transmission cable when transmitting video on a predetermined transmission line. It is in.

In order to solve the above problems, the image transmission device of an embodiment of the present invention converts an input image into an output image having a larger number of pixels than the number of pixels of the input image, and the input image is not arranged in the output image. to place the invalid area is an area to be displayed, a conversion unit to output by dividing the output image into a plurality of divided images each having the invalid area (34), test each of the invalid area of the divided image A test image setting unit (33) for setting an image, a signal processing unit (50) for signal processing the plurality of divided images to generate an output signal, and an output signal of the plurality of divided images on a plurality of transmission paths. A transmission unit (60) for transmission is included. The test image is a pattern in which any sort of the plurality of divided images can be visually recognized from each other.

Another aspect of the present invention is an image receiving device. This device includes receiving units (210, 310) that receive input signals of a plurality of divided images in which an invalid area, which is an area in which the input image is not arranged in the output image, is arranged from a plurality of transmission lines. , the signal processing unit that generates image data by an input signal of the plurality of divided images signal processing (220,322a~322d), pixels than the number of pixels the entering force image by combining the plurality of divided images converter for generating a large number said output image and (234,326), the plurality of divided images test image in each of the invalid area is provided, a predetermined pattern of the test image extracted from the output image When the pattern of the test image is different from the predetermined pattern, each of the input signals of the plurality of divided images should be input to the detection unit (233, 325) that detects the difference by comparing with the patterns. It includes a switching control unit (235, 327) that generates a switching signal that specifies an input terminal of the signal processing unit. Each of the plurality of divided images has the invalid region.

Yet another aspect of the present invention is an image transmission method. In this method, an input image is converted into an output image having a larger number of pixels than the number of pixels of the input image, an invalid area which is an area displayed without arranging the input image in the output image is arranged, and the output is described. a converting step of outputting in a plurality of divided images each having the invalid area image, a test image setting step of setting a test image in each of the invalid area of the divided image, signal processing the plurality of divided images This includes a signal processing step of generating an output signal and a transmission step of transmitting the output signals of the plurality of divided images to a plurality of transmission paths. The test image is a pattern in which any sort of the plurality of divided images can be visually recognized from each other.

It should be noted that any combination of the above components and the conversion of the expression of the present invention between methods, devices, systems, recording media, computer programs and the like are also effective as aspects of the present invention.

According to the present invention, it is possible to determine and correct the connection state of the transmission cable when transmitting video on a predetermined transmission line.

It is a block diagram of the image transmission device which concerns on embodiment. 2 (a) to 2 (c) explain the image data written in the memory by the writing unit of the conversion processing unit of FIG. 1 and the image data read from the memory by the reading unit of the conversion processing unit of FIG. It is a figure. 3A to 3C are diagrams illustrating a conventional method of transmitting video by scaling according to a transmission line for comparison. It is a figure explaining the synchronization signal when the writing part of FIG. 1 writes an input image to a memory. 5 (a) and 5 (b) are timing charts when the writing unit of FIG. 1 writes the input image to the memory. It is a figure explaining the synchronization signal when the reading part of FIG. 1 reads an output image from a memory. 7 (a) and 7 (b) are timing charts when the read unit of FIG. 1 reads an output image from the memory. It is a block diagram of the image receiving apparatus which concerns on embodiment. 9 (a) and 9 (b) explain the image data written in the memory by the writing unit of the conversion processing unit of FIG. 8 and the image data read from the memory by the reading unit of the conversion processing unit of FIG. It is a figure. It is a block diagram of the image receiving apparatus which concerns on another embodiment. It is a flowchart explaining the image conversion procedure by the image transmission device of FIG. It is a flowchart explaining the image restoration procedure by the image receiving apparatus of FIG. 8 or FIG. 13 (a) to 13 (c) are diagrams for explaining a test image set in the invalid area of the output image by the test image setting unit of FIG. 14 (a) to 14 (x) are diagrams for explaining the pattern of the test image in the invalid region. 15 (a) to 15 (d) are diagrams for explaining a switching signal for switching the connection state of the transmission cable to the correct state.

FIG. 1 is a configuration diagram of an image transmission device 100 according to an embodiment. The image transmission device 100 includes an image pickup unit 10, a first signal processing unit 20, a conversion processing unit 30, a memory 40, a second signal processing unit 50, a switch 55, and a transmission unit 60.

As an example, the imaging unit 10 captures a 360-degree omnidirectional image with a dome-shaped photographing camera equipped with a fisheye lens or the like, and outputs image data of 2992 pixels in the horizontal direction and 2976 pixels in the vertical direction from the image sensor. The output format is, for example, a differential output such as LVDS (Low Voltage Differential Signaling).

The first signal processing unit 20 is an FPGA (Field Programmable Gate Array) as an example, and can input the image data output from the image sensor of the imaging unit 10 to the conversion processing unit 30 without changing the image size. ) Convert to a format such as RGB data of an array and output. For example, bus type conversion such as converting a SubLVDS input signal to an LVDS output signal, or frequency conversion such as converting a 600 MHz input signal to a 300 MHz output signal (the reverse frequency conversion is also possible). Format conversion is performed.

The conversion processing unit 30 is, for example, a SoC (System-on-a-Chip), and converts the video input signal from the first signal processing unit 20 into a video output signal in a format suitable for the transmission path of the transmission unit 60. Then, it is supplied to the second signal processing unit 50. Since the data input side synchronization signal and the data output side synchronization signal are different, the conversion processing unit 30 has a writing unit 32 that writes data to the memory 40 according to the input side synchronization signal, and data according to the output side synchronization signal from the memory 40. It is provided with a reading unit 34 for reading data, and different synchronization signals are used for writing to the memory 40 and reading from the memory 40.

The conversion processing unit 30 further includes a test image setting unit 33, and the test image setting unit 33 sets a test image in an invalid area of the image held in the memory 40. The number of horizontal pixels differs between the input image and the output image, and an invalid area in which pixel data does not exist occurs in the output image. By embedding the test image in this invalid area, it becomes possible to detect a connection error of the transmission cable on the side receiving the image data.

2 (a) to 2 (c) explain the image data written in the memory 40 by the writing unit 32 of the conversion processing unit 30 and the image data read from the memory 40 by the reading unit 34 of the conversion processing unit 30. It is a figure. The operations of the writing unit 32 and the reading unit 34 of the conversion processing unit 30 will be described with reference to FIGS. 2A to 2C.

The writing unit 32 of the conversion processing unit 30 performs image processing on the input signals of 2992 pixels in the horizontal direction and 2976 pixels in the vertical direction received from the first signal processing unit 20, and uses the effective areas of the 2960 pixels in the horizontal direction and the 2960 pixels in the vertical direction as the input image. Write to memory 40. According to the horizontal synchronization signal that matches the number of horizontal pixels of the input image, pixel data with 2960 horizontal pixels per line is written to the memory 40, and 2960 lines according to the vertical synchronization signal that matches the number of vertical pixels of the input image. Is written to form one frame of image data in the memory 40. The image processing includes a process of color-converting the RGB data of the Bayer array into YC data such as color difference subsampling 4: 2: 2, and the memory 40 contains YC data having 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction. Is formed, and in the case of color difference subsampling 4: 2: 2, the amount of input data is 2960 × 2960 × 2 = 17,523,200 bytes per frame. As image processing, noise removal, color adjustment, white balance, and the like may be performed as necessary.

FIG. 2A shows image data written in the memory 40 by the writing unit 32. Pixels Y1 and Y2 to Y2960 are written in the first line, pixels Y2961 to Y5920 are written in the second line, and the last pixel is Y8761600 of the 2960th line. At the time of reading, the input image of 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction formed in the memory 40 is divided into upper and lower parts, and the pixel data is read out from the upper half input image and the lower half input image at the same time. The upper half input image is from the first line to the 1480th line, the first pixel is Y1 and the last pixel is Y4380800. The lower half input image is from the 1481th line to the 2960th line, the first pixel is Y438801, and the last pixel is Y8761600. The effective data amount of the Y (luminance) image is 2980 × 1480 = 4,380,800 bytes in the upper half and 2980 × 1480 = 4,380,800 bytes in the lower half. In the case of color difference subsampling 4: 2: 2, the C (color difference) image is composed of two types of data, Cb and Cr, and the C (color difference) image that combines Cb and Cr is the same as the Y (luminance) image. The amount of data. The color difference subsampling 4: 2: 2 is a format in which the horizontal resolutions of the color difference images Cb and Cr are reduced by half with respect to the luminance image Y, and Cb and Cb are used for the two pixels of Y. Each 1 pixel of Cr corresponds to each.

The reading unit 34 of the conversion processing unit 30 is different from the horizontal size (number of pixels) and vertical size (number of pixels) of the input image from each of the upper half input image and the lower half input image of 2960 pixels in the horizontal direction and 1480 pixels in the vertical direction. It is read from the memory 40 as an upper half output image and a lower half output image of 4096 pixels and 1080 pixels in the vertical direction. By reading pixel data with a horizontal pixel count of 4096 pixels per line according to a horizontal sync signal that matches the number of horizontal pixels in the output image, and reading 1080 lines according to a vertical sync signal that matches the number of vertical pixels in the output image. The upper half output image and the lower half output image of one frame are read from the memory 40. When reading out the 1080 line, pixel data exists up to the middle of the 1069th line, but since there is no pixel data after that, it is filled with data that is invalid as an image. When the writing unit 32 of the conversion processing unit 30 writes to the memory 40, data that is invalid as an image may be filled in the area where the pixel data of the memory 40 does not exist.

FIG. 2B shows image data read from the memory 40 by the reading unit 34. By reading out 4096 pixels per line from the upper half input image of 2960 horizontal pixels and 1480 vertical pixels in FIG. 2A, as shown in the upper part of FIG. 2B, 4096 horizontal pixels and 1080 vertical pixels The upper half output image is formed, and similarly, by reading out 4096 pixels per line from the lower half input image of 2960 pixels in the horizontal direction and 1480 pixels in the vertical direction in FIG. 2 (a), as shown in the lower part of FIG. 2 (b). The lower half output image of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction is formed.

As shown in FIG. 2B, the first line of the upper half output image includes pixels Y1 to Y2960 in the first line of the upper half input image in FIG. 2A and Y2961 to Y4096 in the second line. Pixels up to are arranged. Similarly, the pixels of the upper half input image are sequentially arranged from the second line to the 1069th line of the upper half output image, but the pixels of the upper half input image are insufficient in the 1070 line. The 2176 pixels in the first half of the 1070 line are arranged with the pixels of the 1480 line of the upper half input image, but the 1920 pixels in the latter half of the 1070 line do not have effective pixels, so that the image is as an image. Fills in invalid data. Further, 10 lines from the 1071st line to the 1080th line of the upper half output image are invalid lines filled with data that is invalid as an image. Therefore, the effective data amount of the upper half output image of FIG. 2B is 4096 × 1069 + 2176 = 4,380,800 bytes, and the invalid data amount of the image is 4096 × 10 + 1920 = 42,880 bytes. Become. The lower half output image is the same as the upper half output image.

Further, when reading the upper half output image of FIG. 2B, the reading unit 34 divides the upper half output image into two left and right as shown in the upper part of FIG. 2C, and the horizontal size (pixels) of the input image. It can also be read from the memory 40 as an upper left output image and an upper right output image of 2048 pixels in the horizontal direction and 1080 pixels in the vertical direction, which are different from the number) and the vertical size (number of pixels). According to the horizontal synchronization signal that matches the number of horizontal pixels in the output image, pixel data with 2048 horizontal pixels per line is read out at the same time on the left and right, and 1080 lines are displayed according to the vertical synchronization signal that matches the number of vertical pixels in the output image. By reading, the upper left output image and the upper right output image of one frame are read from the memory 40. Similarly, when reading the lower half output image of FIG. 2B, the reading unit 34 divides the lower half output image into two left and right as shown in the lower part of FIG. 2C, and the number of horizontal pixels of the output image. According to the horizontal synchronization signal that matches the number of horizontal sync signals, the left and right pixel data with 2048 horizontal pixels per line are read out at the same time, and 1080 lines are read out according to the vertical synchronization signal that matches the number of vertical pixels in the output image. The lower left output image and the lower right output image are read from the memory 40. As a result, an upper left output image, an upper right output image, a lower left output image, and a lower right output image having 2048 pixels in the horizontal direction and 1080 pixels in the vertical direction are generated.

The upper left output image, the upper right output image, the lower left output image, and the lower right output image are four divisions of a monitor image output with 4096 pixels in the horizontal direction and 2160 pixels in the vertical direction. The reason for dividing into four in this way is that the divided image is simultaneously transmitted by the transmission unit 60 through the four transmission lines SDI1, SDI2, SDI3, and SDI4. The conversion processing unit 30 supplies the upper left output image, the upper right output image, the lower left output image, and the lower right output image read from the memory 40 into four upper, lower, left, and right sides to the second signal processing unit 50.

In this way, the writing unit 32 writes the input image of 3K × 3K (width 2960 pixels, length 2960 pixels) to the memory 40, and the reading unit 34 writes the upper half input image of the width 2960 pixels and the length 1480 pixels formed in the memory 40. , The upper half output image and the lower half output image of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction, in other words, the output image of 4K × 2K (4096 pixels in the horizontal direction and 2160 pixels in the vertical direction) is read out from each of the lower half input images. In the process of this image conversion, the test image setting unit 33 sets the test image in the invalid area of the output image. There are several methods to set the test image in the invalid area of the output image as follows.

The reading unit 34 does not have pixel data when reading the upper half output image and the lower half output image of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction from the upper half input image and the lower half input image of 2960 pixels in the horizontal direction and 1480 pixels in the vertical direction. The invalid area is filled with invalid data, and the test image supplied from the test image setting unit 33 at that time can be set in the invalid area. Alternatively, the upper half output image and the lower half output image read by the reading unit 34 are temporarily stored in the memory 40, and the test image setting unit 33 overwrites the invalid area of the upper half output image and the lower half output image with the test image. You may set a test image by doing so. As yet another method, when the writing unit 32 writes the input image to the memory 40, the pixel data is written to the memory area assuming the upper half input image and the lower half input image of 2960 pixels in the horizontal direction and 1480 pixels in the vertical direction, and the pixel data is written. Data that is invalid as an image may be embedded in an invalid area in which is not present. In that case, the writing unit 32 may set the test image supplied from the test image setting unit 33 in the invalid area.

A test image set in the invalid area of the output image by the test image setting unit 33 will be described with reference to FIGS. 13 (a) to 13 (c). As an example, the test image is set using at least a part of invalid lines of the four divided images of the upper left output image, the upper right output image, the lower left output image, and the lower right output image.

As shown in FIGS. 13 (a) and 13 (b), the upper left output image has a red area 501, the upper right output image has a green area 502, the lower left output image has a blue area 503, and the lower right output image has yellow. A region 504 is set, a black region 511 in the lower right corner of the red region 501, a black region 512 in the lower left corner of the green region 502, a black region 513 in the upper right corner of the blue region 503, and an upper left corner of the yellow region 504. The black area 514 is set. The red region 501, the green region 502, the blue region 503, and the yellow region 504 are referred to as "background regions" of the test image, and the arrangement of the four black regions 511, 512, 513, and 514 is referred to as the "pattern" of the test image.

In the examples of FIGS. 13 (a) and 13 (b), a part of the invalid line is used as the area of the test image, but the invalid line is used in the upper left output image, the upper right output image, the lower left output image, and the lower right output image. The red region 501, the green region 502, the blue region 503, and the yellow region 504 may be set by using all of the above.

The background area of the test image is a four-color image that can identify each of the upper left output image, upper right output image, lower left output image, and lower right output image. This is to improve visibility when determining the pattern of the test image in the output image. When the pattern of the test image is automatically detected, the background area does not necessarily have to be distinguished by four colors, and the same color may be used. Further, the colors of the four black regions 511, 512, 513, 514 constituting the pattern of the test image are arbitrary, and may be one color other than black, or may be distinguished by four different colors.

A test image is obtained by collecting only the invalid lines of the upper left output image, the upper right output image, the lower left output image, and the lower right output image of FIGS. 13 (a) and 13 (b) and configuring them as shown in FIG. 13 (c). You can grasp the whole pattern. When explaining the pattern of the test image, for convenience of explanation, the figure configured as shown in FIG. 13 (c) is used, but when it is displayed on the monitor compatible with 4K, the upper half of FIG. 13 (a) is used. Note that a 4K × 2K output image is displayed, which is a combination of the output image and the lower half output image of FIG. 13 (b).

As shown in FIGS. 13 (a) and 13 (b), since the input image is converted into an output image having a different number of horizontal pixels, the effective area of the output image is a fragmented image, and the output image is compatible with 4K. The effective area is meaningless as an image when displayed as it is on the monitor. On the other hand, since the test image in the invalid area is an image having a predetermined arrangement pattern of the black area in the background area, the pattern of the test image can be visually recognized when displayed on a monitor compatible with 4K. If there is a connection error in the transmission cable, the pattern of the test image will be different. Therefore, the correctness of the connection state of the transmission cable can be visually confirmed by the difference in the pattern of the test image.

The second signal processing unit 50 converts the upper left output image, upper right output image, lower left output image, and lower right output image of 2048 pixels in the horizontal direction and 1080 pixels in the vertical direction supplied from the reading unit 34 of the conversion processing unit 30 into the SDI signal format. Then, it is supplied to the transmission lines SDI1, SDI2, SDI3, and SDI4 of the transmission unit 60, respectively. The output signal supplied from the second signal processing unit 50 to the transmission unit 60 is a 3G-SDI signal standard (SMPTE 424M) for each channel, and the data transfer rate is about 3 Gbit / sec, respectively, for a total of about 12 Gbit / sec. Is.

The switch 55 receives a switching signal from the image receiving device 200 that specifies a transmission path for transmitting each of the output signals of the four divided images. The switch 55 switches the connection state between the output terminals of the output signals of the four divided images of the second signal processing unit 50 and the connection terminals of the four transmission lines of the transmission unit 60 based on the switching signal.

Unless there is a connection error in the transmission cable, the upper left output image, upper right output image, lower left output image, and lower right output image from the second signal processing unit 50 are supplied to the transmission lines SDI1, SDI2, SDI3, and SDI4 of the transmission unit 60, respectively. do it. However, when the image receiving device 200 detects a connection error of the transmission cable based on the test image, instead of reconnecting the transmission cable physically correctly, the switch 55 moves the second signal processing unit 50 based on the changeover signal. By switching the connection state between the output terminal of the above and the connection terminal of the transmission unit 60, a connection error of the transmission cable can be repaired.

The transmission unit 60 simultaneously transmits the upper left output image, the upper right output image, the lower left output image, and the lower right output image supplied from the second signal processing unit 50 on the transmission lines SDI1, SDI2, SDI3, and SDI4, respectively. As a result, the video of one frame is transmitted in four channels as a 3G-SDI signal format, and the amount of output data is 2048 × 1080 × 4 × 2 = 17,694,720 bytes per frame.

As is clear from FIGS. 2A to 2C, the conversion processing unit 30 converts the synchronization signal between the input and output to obtain an input image of 2960 horizontal pixels and 2960 vertical pixels, and 2048 horizontal pixels. Even if the image is converted into an upper left output image, an upper right output image, a lower left output image, and a lower right output image of 1080 vertical pixels, only the vertical and horizontal sizes of the image are changed, and the effective data amount of the image is not changed at all. Therefore, transmission can be performed on the 4-channel transmission lines SDI1 to SDI4 without degrading the image quality at all.

Conventionally, when a video signal is transmitted on a predetermined transmission line, it is necessary to convert the image size to match the synchronization signal of the transmission line, and it is unavoidable that the quality of the video is deteriorated due to scaling.

3A to 3C are diagrams illustrating a conventional method of transmitting an image by scaling according to a synchronization signal of a transmission line for comparison. Here, a case of transmitting in a transmission mode of 3840 pixels per line will be described. As shown in FIG. 3A, 440 pixels on the left and right are left open for alignment with 3840 pixels in one line with respect to an input image of 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction, and 2960 pixels are written in the horizontal direction and in the vertical direction. Reduces the image data so that it becomes 1080 lines and writes it to the memory. The amount of effective data is 2960 × 1080 = 3,196,800 bytes. Here, the image of the upper half is shown, but the image data of the lower half is similarly reduced to generate an image of 3840 pixels in the horizontal direction and 1080 pixels in the vertical direction.

As shown in FIG. 3B, it is read from the memory with an alignment of 3840 pixels per line. The amount of data to be transferred in one horizontal synchronization signal is 3840 pixels. As shown in FIG. 3C, the read image data is transmitted as it is in the SDI signal standard, and can be reproduced as an image by outputting it as it is on the monitor, but it is scaled (reduced) in the vertical direction. , Image quality deteriorates. In the scaling transmission method of FIGS. 3A to 3C, the left and right 440 pixels are not effectively used.

On the other hand, if the image conversion methods described in FIGS. 2A to 2C are used, scaling is not performed, so that the image quality does not deteriorate. However, when the quadrant image of FIG. 2C is displayed on the monitor as it is, it is meaningless as an image because a plurality of lines of the original input image are mixed in one line. As will be described later, in the image receiving device, it is necessary to take out an image of 2960 pixels at a time, reconstruct it, and restore the original image.

FIG. 4 is a diagram illustrating a synchronization signal on the input side when the writing unit 32 writes the input image to the memory 40. The writing unit 32 writes an input image of 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction to the memory 40 at a frame rate of 60 / 1.001 fps (frame per second) under a clock frequency of 720.72 / 1.001 MHz. The writing unit 32 samples the pixels by giving the horizontal synchronization signal at the timing of 3900 samples and the vertical synchronization signal at the timing of 3080 lines so that the number of effective samples to be input is 2960 samples in the horizontal direction and 2960 lines in the vertical direction. The dotted arrow in the figure indicates the writing order.

5 (a) and 5 (b) are timing charts when the writing unit 32 writes the input image to the memory 40. FIG. 5A shows a horizontal synchronization signal (Hsync) given every 3900 samples and a vertical synchronization signal (Vsync) given every 3080 lines.

FIG. 5B is an enlarged view of the vertical synchronization signal (Hsync) of FIG. 5A, in which 2960 pixels per line are sampled as valid data in the horizontal direction during one horizontal synchronization period. The timing chart until the last 4380800th pixel of the 1480th line of the upper half is sampled is shown. After this, the lower half is sampled in the same manner.

FIG. 6 is a diagram illustrating a synchronization signal on the output side when the reading unit 34 reads an output image from the memory 40. The reading unit 34 has a frame rate of 60/1 for the upper half of the input image stored in the memory 40 to the upper half of the output image having 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction under a clock frequency of 148.5 / 1.001 MHz. It is read from the memory 40 at 001 fps, and the lower half of the output image of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction is read out from the memory 40 at a frame rate of 60 / 1.001 fps from the lower half of the input image stored in the memory 40. Here, the upper half of the input / output image will be mainly described, and the lower half is the same as the upper half, so the description thereof will be omitted as appropriate. The reading unit 34 gives a horizontal synchronization signal of 2200 samples and a vertical synchronization signal at a timing of 1125 lines so that the number of effective samples to be output is 2048 samples in the horizontal direction and 1080 lines in the vertical direction, and pixels are provided for each line. Sampling. The dotted arrow in the figure indicates the reading order. It is divided into two parts on the left and right and read out at the same time on the left and right. Further, the C image is also divided into two and read out at the same time on the left and right as in the Y image.

7 (a) and 7 (b) are timing charts when the reading unit 34 reads the output image from the memory 40. FIG. 7A shows a horizontal sync signal (Hsync) given every 2200 samples and a vertical sync signal (Vsync) given every 1125 lines.

FIG. 7B is an enlarged view of the vertical synchronization signal (Hsync) of FIG. 7A, and the upper half of the output image is the data for SDI1 (upper left output image) in the horizontal direction during one horizontal synchronization period. ) And SDI2 data (upper right output image), 2048 pixels per line are sampled as valid data, and a timing chart is shown until the 1070th line is sampled. In the 1070th line, the pixels from the 4388624th to the 4380672th as the SDI1 data are read as effective samples, the pixels from the 4380673th to the 4380800th as the SDI2 data are read out as the effective sample, and the rest are as images. The invalid data is filled. From the 1071st line to the 1080th line, only data that is invalid as an image is filled.

The lower half of the output image is the same as the upper half, and the SDI3 data (lower left output image) and the SDI4 data (lower right output image) are simultaneously read out according to the same synchronization signal. That is, an image in which the Y1 pixel of the SDI1 data, the Y2049 pixel of the SDI2 data, the Y438801 pixel of the SDI3 data, and the Y43882849 pixel of the SDI4 data are used as the first read pixels, and the output timing is aligned thereafter to divide the image into four parts vertically and horizontally. Are read out at the same time according to the synchronization signal. Also, the C image is read out in the same manner as the Y image.

FIG. 8 is a configuration diagram of the image receiving device 200 according to the embodiment. The image receiving device 200 includes a receiving unit 210, a switch 215, a third signal processing unit 220, a conversion processing unit 230, a memory 240, an SD (Secure Digital) card 250, and a display control unit 260.

The image receiving device 200 receives the image data transmitted from the image transmitting device 100, restores the image, and displays or saves the image. Since the output image produced by the image transmitting device 100 does not make sense as an image even if it is displayed on the monitor as it is, it is necessary to restore the original image in the image receiving device 200.

The receiving unit 210 receives the image data of 2048 pixels in the horizontal direction and 1080 pixels in the vertical direction transmitted by the transmitting unit 60 of the image transmitting device 100 through the four transmission lines SDI1, SDI2, SDI3, and SDI4. The amount of input data is 2048 × 1080 × 4 × 2 = 17,694,720 bytes per frame. The upper left image, upper right image, lower left image, and lower right image received by SDI1, SDI2, SDI3, and SDI4 are supplied to the third signal processing unit 220.

The third signal processing unit 220 is an FPGA as an example, and converts the input image data into a format that can be input to the conversion processing unit 230 without changing the image size and outputs the data.

As an example, the conversion processing unit 230 is SoC, and restores and outputs the original image from the video input signal from the third signal processing unit 220. Since the data input side synchronization signal and the data output side synchronization signal are different, the conversion processing unit 230 reads data from the memory 240 according to the output side synchronization signal and the writing unit 232 that writes data to the memory 240 according to the input side synchronization signal. A reading unit 234 is provided to convert a synchronization signal between input and output.

The conversion processing unit 230 further includes a detection unit 233 and a switching control unit 235. The detection unit 233 compares the pattern of the test image in the invalid area of the input image held in the memory 240 with the predetermined pattern to detect the difference. When the pattern of the test image in the invalid area of the input image is different from the predetermined pattern, the switching control unit 235 receives each divided image received by the four transmission lines SDI1, SDI2, SDI3, and SDI4 in the receiving unit 210. Generates a switching signal that specifies the input terminal of the third signal processing unit 220 to which the signal of is to be input, and supplies the switching signal to the switch 215.

The switch 215 switches the connection state between the connection terminals of the four transmission lines of the receiving unit 210 and the input terminals of the input signals of the four divided images of the third signal processing unit 220 based on the switching signal.

As long as there is no connection error of the transmission cable, the receiving unit 210 may supply each divided image received by the four transmission lines SDI1, SDI2, SDI3, and SDI4 to the corresponding input terminals of the third signal processing unit 220, respectively. However, when a transmission cable connection error is detected based on the test image, instead of reconnecting the transmission cable physically correctly, the switch 215 performs the connection terminal of the receiver 210 and the third signal processing based on the changeover signal. By switching the connection state between the input terminals of the unit 220, it is possible to repair the connection error of the transmission cable.

When the image receiving device 200 is not provided with the switch 215, the switching control unit 235 transmits the switching signal to the image transmitting device 100, and the switch 55 of the image transmitting device 100 is based on the switching signal of the second signal processing unit 50. By switching the connection state between the output terminal and the connection terminal of the transmission unit 60, it is possible to repair the connection error of the transmission cable.

9 (a) and 9 (b) explain the image data written in the memory 240 by the writing unit 232 of the conversion processing unit 230 and the image data read from the memory 240 by the reading unit 234 of the conversion processing unit 230. It is a figure. The operations of the writing unit 232 and the reading unit 234 of the conversion processing unit 230 will be described with reference to FIGS. 9A and 9B.

The writing unit 232 of the conversion processing unit 230 receives the upper left image and the upper right image from the third signal processing unit 220, and according to the horizontal synchronization signal, the pixel data of the upper left image and the upper right image having 2048 horizontal pixels per line. Is written to the corresponding positions on the left and right of the memory 240, and 1080 lines are written according to the vertical synchronization signal. As a result, the image data of the upper half of one frame is formed in the memory 240. Similarly, the writing unit 232 receives the lower left image and the lower right image from the third signal processing unit 220, and according to the horizontal synchronization signal, the pixel data of the lower left image and the lower right image having 2048 horizontal pixels per line. Is written to the corresponding positions on the left and right of the memory 240, and 1080 lines are written according to the vertical synchronization signal. As a result, the image data of the lower half of one frame is formed in the memory 240.

As shown in the upper part of FIG. 9A, the upper left image of 2048 pixels horizontally and 1080 pixels vertically transmitted by SDI1 is written in the left half as shown by the dotted arrow, and 2048 pixels horizontally and vertically transmitted by SDI2. The upper right image of 1080 pixels is written in the right half as shown by the dotted arrow. As a result, an image of the upper half of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction is formed in the memory 240. Similarly, as shown in the lower part of FIG. 9A, the upper left image of the horizontal 2048 pixels and the vertical 1080 pixels transmitted by SDI3 is written in the left half as shown by the dotted arrow, and the horizontal 2048 transmitted by SDI4. The upper right image of pixels and 1080 pixels in the vertical direction is written in the right half as shown by the dotted arrow. As a result, an image of the lower half of 4096 pixels in the horizontal direction and 1080 pixels in the vertical direction is formed in the memory 240. The image data divided into four in the vertical and horizontal directions is written to the memory 240 at the same time.

Regarding the horizontal synchronization signal and the vertical synchronization signal when the writing unit 232 writes the image data to the memory 240, the conversion processing unit 30 of the image transmitting device 100 described with reference to FIGS. 6, 7 (a) and 7 (b). The horizontal synchronization signal and the vertical synchronization signal similar to the timing chart when the reading unit 34 reads the image data divided into four from the memory 40 may be used.

As shown in FIG. 9B, the reading unit 234 of the conversion processing unit 230 follows the horizontal synchronization signal from the images of the upper half and the lower half of the horizontal 4096 pixels and the vertical 1080 pixels formed in the memory 240, respectively. The image of the upper half and the lower half of one frame is read from the memory 240 by reading the pixel data having 2960 horizontal pixels per line and reading 1480 lines according to the vertical synchronization signal. As a result, an image having 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction is generated. This is the original image captured by the image transmission device 100, and can be displayed as it is on the display device or stored in a recording medium.

Regarding the horizontal synchronization signal and the vertical synchronization signal when the reading unit 234 reads the image data from the memory 240, the conversion processing unit 30 of the image transmitting device 100 described with reference to FIGS. 4, 5 (a) and 5 (b). The writing unit 32 may use a horizontal synchronization signal and a vertical synchronization signal similar to the timing chart when the image data is written to the memory 40.

The reading unit 234 may convert the size of the read image as needed. For example, the image size may be changed according to the output destination of a display device or the like. As an example, an image having 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction is converted into 2048 pixels in the horizontal direction and 2048 pixels in the vertical direction. Alternatively, the reading unit 234 may perform a process of cutting out a part of the area or expanding the image area without converting the size of the read image. As an example, black regions may be added to the left and right of an image having 2960 pixels in the horizontal direction and 2960 pixels in the vertical direction to convert the image into 3840 pixels in the horizontal direction and 2960 pixels in the vertical direction.

In this way, the writing unit 232 writes the upper half input image and the lower half input image of the horizontal 4096 pixels and the vertical 1080 pixels to the memory 240, and the reading unit 234 writes the upper half of the horizontal 4096 pixels and the vertical 1080 pixels formed in the memory 240. The upper half output image and the lower half output image of 2960 pixels in the horizontal direction and 1480 pixels in the vertical direction are read out from the input image and the lower half input image, respectively. The writing unit 232 forms a 4K × 2K (horizontal 4096 pixels, vertical 2160 pixels) input image in the memory 240, and the reading unit 234 extracts a 3K × 3K (horizontal 2960 pixels, vertical 2960 pixels) output image. In the process of this image conversion, the detection unit 233 detects the pattern of the test image in the invalid region of the input image and compares it with a predetermined pattern.

If there is no connection error of the transmission cable, the upper half input image of the horizontal 4096 pixels and the vertical 1080 pixels formed in the memory 240 is the same as that shown in FIG. 13 (a), and the horizontal 4096 pixels formed in the memory 240. The lower half input image of 1080 pixels in the vertical direction is the same as that shown in FIG. 13 (b). The detection unit 233 extracts a test image by combining the invalid line of the upper half input image of 4096 pixels horizontally and 1080 pixels vertically and the invalid line of the lower half input image of 4096 pixels horizontally and 1080 pixels vertically formed in the memory 240. do. This is the same as that shown in FIG. 13 (c) if there is no connection error of the transmission cable. The detection unit 233 determines whether or not the pattern of the test image of the received input image is the predetermined pattern of FIG. 13C.

The display control unit 260 reads an input image of 4K × 2K (width 4096 pixels, height 2160 pixels) formed in the memory 240 and displays it on a 4K compatible monitor. As a result, the user can visually recognize the test image in the invalid area of the input image, and can visually determine whether or not the test image has a predetermined pattern. In this case, the user may manually set the switching signal to the switching control unit 235 by omitting the automatic determination by the detection unit 233.

The streaming unit 236 encodes the image data read by the reading unit 234, and AVC / H. 264 and HEVC / H. A coded stream is generated according to a moving image compression coding standard such as 265. The SD card interface 238 stores the encoded video stream on the SD card 250. A display unit (not shown) may display the image read by the reading unit 234 on the display device.

The pattern of the test image of the invalid region will be described with reference to FIGS. 14 (a) to 14 (x). FIG. 14A schematically shows the test image of FIG. 13C. Here, the entire region of the invalid line shown in FIG. 13C is used as the background region, and the central black region straddling SDI1 to SDI4 is used as the pattern. If the transmission cable is connected correctly, the black test image on the lower right of SDI1, the black test image on the lower left of SDI2, the black test image on the upper right of SDI3, and the black test image on the upper left of SDI4 are shown in FIG. 14A. Arranged as shown, the black areas are centered. Assuming that the connection terminals of the receiving unit 210 to which the transmission lines SDI1, SDI2, SDI3, and SDI4 should be connected are A, B, C, and D, respectively, {connection terminals A, SDI1}, {connection terminals B, SDI2}, and {connection The connection status of the transmission cables, terminals C, SDI3} and {connection terminals D, SDI4}, constitutes the correct pattern of the test image.

However, for example, if the transmission cable is mistakenly inserted between the connection terminal C and the connection terminal D, {connection terminal A, SDI1}, {connection terminal B, SDI2}, {connection terminal C, SDI4}, {connection terminal D, A transmission cable called SDI3} is connected, and the black test image on the upper left of SDI4 is input to the connection terminal C, and the black test image on the upper right of SDI3 is input to the connection terminal D. In this case, the pattern of the test image as shown in FIG. 14B is detected, and the black region does not gather in the center. As another example, if the transmission cable is mistakenly inserted between the connection terminal B and the connection terminal C, {connection terminal A, SDI1}, {connection terminal B, SDI3}, {connection terminal C, SDI2}, {connection terminal The transmission cable D, SDI4} is connected, and the black test image on the upper right of SDI3 is input to the connection terminal B, and the black test image on the lower left of SDI2 is input to the connection terminal C. In this case, the pattern of the test image as shown in FIG. 14C is detected, and the black region does not gather in the center. Similarly, in FIGS. 14 (d) to 14 (x), there is a mistake in inserting the transmission cable between any two or more terminals, and the pattern of the test image is such that the black region does not gather in the center.

There are a total of 24 possible patterns in which four transmission cables are mistakenly inserted between the four connection terminals A, B, C, and D. That is, there may be 23 patterns of the test image of the incorrect connection state shown in FIGS. 14 (b) to 14 (x) with respect to the pattern of the test image of the correct connection state of FIG. 14 (a). By comparing the detected pattern of the test image with the correct pattern of FIG. 14A, it is possible to determine whether the connection state of the transmission cable to the connection terminals A, B, C, and D of the receiving unit 210 is correct or not. ..

If a easily visible pattern in which a black area is gathered in the center as shown in FIG. 14 (a) is used as a correct pattern, the user actually sees the test image on a 4K compatible monitor and views 14 (b) to 14 (Fig. 14). With the pattern of x), it can be easily grasped that there was a connection error of the transmission cable. However, if the detection unit 233 automatically detects the difference in the pattern of the test image, it is not always necessary to set the easily visible pattern as shown in FIG. 14 (a) as the correct pattern. For example, in FIG. 14 (c). The pattern may be used as the correct pattern, and the other patterns may be discriminated as the incorrect pattern. Further, if a pattern other than that shown in FIG. 14A is easy for the user to see, it can be used as a correct pattern. For example, if a pattern having black four corners as shown in FIG. 14 (x) is used as a correct pattern, the user can visually compare with other patterns to determine the difference between the patterns.

15 (a) to 15 (d) are diagrams for explaining a switching signal for switching the connection state of the transmission cable to the correct state. FIG. 15A is a diagram showing the arrangement of the four connection terminals A, B, C, and D of the receiving unit 210. FIG. 15B shows a pattern of a test image in a state where transmission lines SDI1, SDI2, SDI3, and SDI4 are correctly connected to the connection terminals A, B, C, and D of FIG. 15A, respectively. The lower right black test image of SDI1 is input to the connection terminal A, the lower left black test image of SDI2 is input to the connection terminal B, the upper right black test image of SDI3 is input to the connection terminal C, and the upper left black test image of SDI4 is input to the connection terminal D. Therefore, the black area is gathered in the center.

FIG. 15C shows a pattern of the test image detected by the detection unit 233. This is the same as the pattern shown in FIG. 14 (d), there is a mistake in inserting three transmission cables between the three connection terminals B, C, and D, and the SDI1 is correctly connected to the connection terminal A. However, SDI3 is connected to the connection terminal B, SDI4 is connected to the connection terminal C, and SDI2 is erroneously connected to the connection terminal D. As a result, the lower right black test image is input to the connection terminal A, the upper right black test image is input to the connection terminal B, the upper left black test image is input to the connection terminal C, and the lower left black test image is input to the connection terminal D. Areas do not gather.

The detection of the pattern of the test image of FIG. 15C is between the connection terminals A to D of the transmission unit 60 of the image transmission device 100 and the connection terminals A to D of the reception unit 210 of the image reception device 200. It means that the following transmission cable is misinserted. The connection terminal A of the transmission unit 60 is correctly connected to the connection terminal A of the reception unit 210, but the connection terminal B of the transmission unit 60 is connected to the connection terminal D of the reception unit 210, and the connection terminal C of the transmission unit 60 is the reception unit. The connection terminal D of the transmission unit 60 is connected to the connection terminal B of the 210, and the connection terminal D of the transmission unit 60 is connected to the connection terminal C of the reception unit 210. The erroneous connection relationship between the transmitting unit 60 and the receiving unit 210 can be described as (AtoA, BtoD, CtoB, DtoC). In order to return this erroneous connection relationship to the correct connection relationship, the input / output connection relationship is on the receiving unit 210 side or the transmitting unit 60 side using a switching signal that is inverted (AtoA, BtoC, CtoD, DtoB). Just switch.

On the left side of FIG. 15D, erroneous connection states of {connection terminal A, SDI1}, {connection terminal B, SDI3}, {connection terminal C, SDI4}, and {connection terminal D, SDI2} are shown. On the right side of FIG. 15D, the correct connection states of {connection terminal A, SDI1}, {connection terminal B, SDI2}, {connection terminal C, SDI3}, and {connection terminal D, SDI4} are shown. .. The arrows in FIG. 15 (d) indicate switching signals (AtoA, BtoC, CtoD, DtoB) for correcting the incorrect connection state in FIG. 15 (c) to the correct connection state in FIG. 15 (b). The switching control unit 235 supplies switching signals (AtoA, BtoC, CtoD, DtoB) to the switch 215.

The switch 215 outputs the input side terminal A of the switch 215 to the output side terminal A, the input side terminal B to the output side terminal C, and the input side terminal C based on the switching signals (AtoA, BtoC, CtoD, DtoB). The input side terminal D is switched to the side terminal B so as to be connected to the output side terminal B. That is, the switch 215 switches the connection state between the receiving unit 210 and the third signal processing unit 220, and connects the connection terminal A of the receiving unit 210 to the input terminal A of the third signal processing unit 220 and the connection terminal of the receiving unit 210. B is the input terminal C of the third signal processing unit 220, the connection terminal C of the receiving unit 210 is the input terminal D of the third signal processing unit 220, and the connection terminal D of the receiving unit 210 is the input of the third signal processing unit 220. Connect to terminal B.

As a result, the output of SDI1 from the connection terminal A of the receiving unit 210 is input to the input terminal A of the third signal processing unit 220, and the output of SDI3 from the connection terminal B of the receiving unit 210 is the output of the third signal processing unit 220. It is input to the input terminal C, the output of SDI4 from the connection terminal C of the receiving unit 210 is input to the input terminal D of the third signal processing unit 220, and the output of SDI2 from the connection terminal D of the receiving unit 210 is the third signal. It is input to the input terminal B of the processing unit 220. In this way, the operation of the switch 215 returns the incorrect connection state to the correct connection state, and eliminates the connection error of the transmission cable between the image transmitting device 100 and the image receiving device 200.

When the image receiving device 200 does not have the switch 215 and the image transmitting device 100 is provided with the switch 55, the switching control unit 235 transmits the same switching signal (AtoA, BtoC, CtoD, DtoB) to the image transmitting device 100. The switch 55 of the image transmission device 100 connects the output terminal A of the second signal processing unit 50 to the connection terminal A for SDI1 of the transmission unit 60 based on the switching signals (AtoA, BtoC, CtoD, DtoB). 2 The output terminal B of the signal processing unit 50 is connected to the connection terminal C for SDI 3 of the transmission unit 60, the output terminal C of the second signal processing unit 50 is connected to the connection terminal D for SDI 4 of the transmission unit 60, and the second signal processing unit 50 is connected. 2 The output terminal D of the signal processing unit 50 is connected to the connection terminal B for SDI2 of the transmission unit 60.

As a result, the output of SDI1 from the output terminal A of the second signal processing unit 50 is input to the connection terminal A of the transmission unit 60, and the output of SDI2 from the output terminal B of the second signal processing unit 50 is the output of the transmission unit 60. It is input to the connection terminal C, the output of SDI3 from the output terminal C of the second signal processing unit 50 is input to the connection terminal D of the transmission unit 60, and the output of SDI4 from the output terminal D of the second signal processing unit 50 is output. It is input to the connection terminal B of the transmission unit 60.

Here, as described above, the connection terminal A of the transmission unit 60 is correctly connected to the connection terminal A of the reception unit 210, but the connection terminal B of the transmission unit 60 is connected to the connection terminal D of the reception unit 210. The connection terminal C of the 60 is connected to the connection terminal B of the reception unit 210, and the connection terminal D of the transmission unit 60 is connected to the connection terminal C of the reception unit 210. Therefore, as a result, SDI1, SDI2, SDI3, and SDI4 are input to the connection terminals A, B, C, and D of the receiving unit 210, respectively. In this way, the operation of the switch 55 returns the incorrect connection state to the correct connection state, and eliminates the connection error of the transmission cable between the image transmitting device 100 and the image receiving device 200.

FIG. 10 is a configuration diagram of an image receiving device 300 according to another embodiment. The image receiving device 300 includes a receiving unit 310, a switch 315, a computer 320, and a monitor 330. The receiving unit 310 receives the image data transmitted from the image transmitting device 100, and the computer 320 restores the received image data and displays it on the monitor 330.

The receiving unit 310 receives the image data of 2048 pixels in the horizontal direction and 1080 pixels in the vertical direction transmitted by the transmitting unit 60 of the image transmitting device 100 on the four transmission lines SDI1, SDI2, SDI3, and SDI4.

The computer 320 includes four SDI input capture boards 322a to 322d for receiving four channels of 3G-SDI signals. The upper left image, upper right image, lower left image, and lower right image received by SDI1, SDI2, SDI3, and SDI4 of the receiving unit 310 are input to the SDI input capture boards 322a, 322b, 322c, and 322d, respectively.

Similar to the writing unit 232 of FIG. 8, the memory interface 323 writes the upper left image, the upper right image, the lower left image, and the lower right image to the memory 324 according to the horizontal synchronization signal, and the upper half of the horizontal 4096 pixels and the vertical 1080 pixels and The image data of the lower half is formed in the memory 324.

The detection unit 325 compares the pattern of the test image in the invalid area of the input image held in the memory 324 with the predetermined pattern to detect the difference. When the pattern of the test image in the invalid area of the input image is different from the predetermined pattern, the switching control unit 327 receives each divided image received by the four transmission lines SDI1, SDI2, SDI3, and SDI4 in the receiving unit 310. Generates a switching signal that specifies the input terminals of the SDI input capture boards 322a to 322d to which the signal of 322a to 322d should be input, and supplies the switching signal to the switch 315.

The switch 315 switches the connection state between the connection terminals of the four transmission lines of the receiving unit 310 and the input terminals of the four SDI input capture boards 322a to 322d based on the switching signal.

As long as there is no connection error of the transmission cable, the receiving unit 310 may supply the divided images received by the four transmission lines SDI1, SDI2, SDI3, and SDI4 to the SDI input capture boards 322a to 322d, respectively. However, if a transmission cable connection error is detected based on the test image, instead of reconnecting the transmission cable physically correctly, the switch 315 transfers the connection terminal of the receiver 310 and the four SDI inputs based on the changeover signal. By switching the connection state between the input terminals of the capture boards 322a to 322d, it is possible to repair the connection error of the transmission cable.

The switch 315 may be provided outside the computer 320, or may be built in the computer 320. If it is difficult to provide the switch 315 on the computer 320 side, the switching control unit 327 transmits a switching signal to the image transmitting device 100, and the switch 55 of the image transmitting device 100 transmits the switching signal to the second signal processing unit 50 based on the switching signal. The connection state between the output terminal of the above and the connection terminal of the transmission unit 60 may be switched.

Similar to the reading unit 234 of FIG. 8, the image format conversion unit 326 reads the image data from the memory 324 with the alignment of 2960 pixels horizontally according to the horizontal synchronization signal, and reads the image of 2960 pixels horizontally and 2960 pixels vertically. It supplies to the graphic board 328. When reading the image data, the image format conversion unit 326 may perform image processing such as changing the image size, cutting out a part of the area without changing the image size, or expanding the image area. Further, the image format conversion unit 326 may change the color difference format of the image. For example, the 4: 2: 2 YUV format may be changed to the RGB format.

The graphic board 328 transfers the image data supplied from the image format conversion unit 326 to the monitor 330 and displays it.

The graphic board 328 may read an input image of 4K × 2K (horizontal 4096 pixels, vertical 2160 pixels) formed in the memory 324 and display it on a 4K compatible monitor. As a result, the user can visually recognize the test image in the invalid area of the input image, and can visually determine whether or not the test image has a predetermined pattern. In this case, the user may manually set the switching signal to the switching control unit 327 by omitting the automatic determination by the detection unit 325.

FIG. 11 is a flowchart illustrating an image conversion procedure by the image transmission device 100.

The imaging unit 10 captures an input image and gives the acquired input image to the first signal processing unit 20 (S10).

The first signal processing unit 20 converts the input image into a format suitable for the conversion processing unit 30 and supplies the input image to the conversion processing unit 30 (S12).

The writing unit 32 of the conversion processing unit 30 writes the input image to the memory 40 according to the synchronization signal A on the input side (S14). The synchronization signal A on the input side is a horizontal synchronization signal and a vertical synchronization signal that match the number of horizontal pixels and the number of vertical pixels of the input image, as described with reference to FIGS. 4, 5 (a), and 5 (b). ..

The reading unit 34 of the conversion processing unit 30 reads the output image from the memory 40 according to the synchronization signal B on the output side (S16). The synchronization signal B on the output side is a horizontal synchronization signal and a vertical synchronization signal that match the number of horizontal pixels and the number of vertical pixels of the output image, as described with reference to FIGS. 6, 7 (a) and 7 (b). ..

The test image setting unit 33 of the conversion processing unit 30 sets the test image in the invalid area of the output image read from the memory 40 (S17).

The second signal processing unit 50 converts the output image read from the memory 40 into a transmission format suitable for a transmission line such as 3G-SDI, and supplies the transmission signal to the transmission unit 60 (S18). The transmission unit 60 transmits image data on the transmission line (S20).

The size of the output image needs to be smaller than the image size that can be transmitted on a predetermined transmission line, and if transmission cannot be performed on one channel transmission line, the output image is divided and transmitted using a plurality of channel transmission lines. In that case, when the output image is read from the memory 40, the image data is read by dividing the output image according to the number of channels of the transmission line.

When a switching signal is received from the image receiving device 200 (Y in S22), the switch 55 switches the input / output connection between the second signal processing unit 50 and the transmitting unit 60 based on the switching signal, so that the connection state of the transmission line is changed. Is repaired (S24). After that, the process returns to step S20, and the transmission unit 60 transmits the image data again under the repaired connection state.

When the switching signal is not received from the image receiving device 200 (N in S22), the process ends.

FIG. 12 is a flowchart illustrating an image restoration procedure by the image receiving devices 200 and 300.

The receiving units 210 and 310 receive the image data transmitted from the image transmitting device 100 on a plurality of transmission lines (S30).

The third signal processing unit 220 and the SDI input capture boards 322a to 322d convert the received input image into a format suitable for the writing unit 232 and the memory interface 323 and supply the received input image to the writing unit 232 and the memory interface 323 (S32). ..

The writing unit 232 and the memory interface 323 write the input image to the memory 240 and 324 according to the synchronization signal B (S34). As the synchronization signal B, the same synchronization signal B as the synchronization signal B on the output side when the image transmission device 100 reads the output image from the memory 40 can be used.

The detection unit 233, 325 determines whether or not the test image is correct by comparing the pattern of the test image in the invalid region of the test image of the input image with a predetermined pattern (S35). If the test image is correct (Y in S35), the process proceeds to step S36.

If the test image is incorrect (N in S35), the switching control units 235 and 327 generate a switching signal and supply it to the switches 215 and 315 (S42). The switch 215 switches the input / output connection between the receiving unit 210 of the image receiving device 200 and the third signal processing unit 220 based on the switching signal, and the switch 315 switches between the receiving unit 310 of the image receiving device 300 and the SDI input capture board. By switching the input / output connection between 322a and 322d based on the switching signal, the connection state of the transmission line is restored (S44). After step S44, the process returns to step S32 and the subsequent processes are executed.

The reading unit 234 and the image format conversion unit 326 read the output signal from the memory 240 and the memory 324 according to the synchronization signal A (S36). As the synchronization signal A, the same synchronization signal A on the input side when the image transmission device 100 writes the input image to the memory 40 can be used.

The reading unit 234 and the image format conversion unit 326 change the format of the read output image as necessary (S38).

The output image is displayed on the monitor 330 as it is, or is compressed and encoded and saved as a moving image stream on an SD card 250 or the like (S40).

If the image receiving device 200 is not provided with the switch 215 or the image receiving device 300 is not provided with the switch 315, the switching signal generated in step S42 is transmitted to the image transmitting device 100 to transmit the switching signal to the image transmitting device 100. The switch 55 of the above repairs the connection state of the transmission line by switching the input / output connection based on the changeover signal.

In the above embodiment, a method of converting a 3K × 3K input image into four 2K × 1K (horizontal 2048 pixels, vertical 1080 pixels) output images and transmitting them in a 4-channel 3G-SDI has been described. This is an example, and the sizes of the input image and the output image, the number of divisions of the output image, the operating speed and the number of transmission lines, and the like can be other than this. In the case of a transmission line capable of transmitting 4K x 2K on one channel, an input image of 3K x 3K (2960 pixels in width and 2960 pixels in height) is converted into an output image of 4K x 2K (4096 pixels in width and 2160 pixels in height). It may be transmitted on a 1-channel transmission line. Further, the input image of 3K × 3K (2960 × 2960 pixels) is converted into the upper half output image and the lower half output image of 4K × 1K (horizontal 4096 pixels, vertical 1080 pixels), and the upper half output image and the lower half output image are converted. May be combined and transmitted as a final 4K × 2K (horizontal 4096 pixels, vertical 2160 pixels) output image on a 1-channel transmission path. When the size of the input image is horizontal K pixel and vertical L pixel, and the size of the output image is horizontal M pixel and vertical N pixel, the number of pixels of the output image is equal to or larger than the number of pixels of the input image because scaling is not performed. That is, it is premised that K × L ≦ M × N. When L> N for the number of vertical pixels or K> M for the number of horizontal pixels, it is necessary to convert the horizontal synchronization signal or the vertical synchronization signal between the input and output to convert the image size. The image conversion method of the form can be applied to convert the output image into a transmission format without degrading the image quality. Further, when the reading unit 34 reads the pixel data from the memory in units of the number of horizontal pixels of the output image, it may be divided by a number other than four divisions. When the number of divisions is P, the P-divided image data can be output by dividing the P-division and switching the output destination, and the transmission unit can transmit the P-divided image data to the P transmission lines. can.

As described above, according to the image transmitting device 100 according to the embodiment, the image receiving devices 200 and 300 are used to convert the output image into an output image having a size suitable for a predetermined transmission line while maintaining the resolution of the input image. In, the original image can be restored and displayed or saved, and the image quality is not deteriorated. In addition, since high-quality video is transmitted using an existing transmission line, development costs can be reduced.

The present invention has been described above based on the embodiments. Embodiments are examples, and it will be understood by those skilled in the art that various modifications are possible for each of these components and combinations of each processing process, and that such modifications are also within the scope of the present invention. ..

The functional configuration of each device described in the embodiment can be realized by hardware resources or software resources, or by collaboration between hardware resources and software resources. Processors, ROMs, RAMs, and other LSIs can be used as hardware resources. Programs such as operating systems and applications can be used as software resources.

10 Imaging unit, 20 1st signal processing unit, 30 Conversion processing unit, 32 Writing unit, 33 Test image setting unit, 34 Reading unit, 40 memory, 50 2nd signal processing unit, 55 switch, 60 Transmitting unit, 100 Image transmission Device, 200 image receiving device, 210 receiving unit, 215 switch, 220 third signal processing unit, 230 conversion processing unit, 232 writing unit, 233 detecting unit, 234 reading unit, 235 switching control unit, 236 streaming unit, 238 SD Card interface, 240 memory, 250 SD card, 260 display control unit, 300 image receiver, 310 receiver, 315 switch, 320 computer, 322a, 322b, 322c, 322d SDI input capture board, 323 memory interface, 324 memory, 325 Detection unit, 326 image format conversion unit, 327 switching control unit, 328 graphic board, 330 monitor.

Claims (6)

The input image is converted into an output image having a larger number of pixels than the number of pixels of the input image, an invalid area which is an area where the input image is not arranged and displayed in the output image is arranged, and the output image is converted into the invalid area. A conversion unit that divides and outputs a plurality of divided images having regions, and
A test image setting unit that sets a test image in each of the invalid areas of the divided image,
A signal processing unit that generates an output signal by signal processing the plurality of divided images,
A transmission unit that transmits output signals of the plurality of divided images to a plurality of transmission lines is included.
The image transmission device is characterized in that the test image is a pattern in which arbitrary rearrangement of the plurality of divided images can be visually recognized from each other. The first aspect of claim 1, wherein the test image is a pattern composed of one color, and an arbitrary rearrangement of the plurality of divided images can be visually recognized by the arrangement of the one color. Image transmitter. When the pattern of the test image is different from the predetermined pattern, a switching signal for designating a transmission path to transmit each of the output signals of the plurality of divided images is received, and the signal is based on the switching signal. 1. The image transmitter described. A receiving unit that receives input signals of a plurality of divided images in which an invalid area, which is an area displayed without arranging the input image in the output image, is arranged from a plurality of transmission lines.
A signal processing unit that generates image data by processing the input signals of the plurality of divided images.
A conversion unit to generate the output image is also a large number of pixels than the number of pixels the entering force image by combining the plurality of divided images,
A detection unit that detects differences by comparing a pattern of the test image extracted from the output image with a predetermined pattern by providing a test image in the invalid region of each of the plurality of divided images.
When the pattern of the test image is different from the predetermined pattern, a switching control unit that generates a switching signal that specifies an input terminal of the signal processing unit to which each of the input signals of the plurality of divided images should be input is provided. Including
An image receiving device, wherein each of the plurality of divided images has the invalid area. The input image is converted into an output image having a larger number of pixels than the number of pixels of the input image, an invalid area which is an area where the input image is not arranged and displayed in the output image is arranged, and the output image is converted into the invalid area. A conversion step that divides and outputs a plurality of divided images each having an area, and
A test image setting step of setting a test image in each of the invalid areas of the divided image, and
A signal processing step of processing the plurality of divided images to generate an output signal, and
The transmission step of transmitting the output signals of the plurality of divided images to a plurality of transmission lines is included.
The image transmission method, wherein the test image is a pattern in which arbitrary rearrangement of the plurality of divided images can be visually recognized from each other. The input image is converted into an output image having a larger number of pixels than the number of pixels of the input image, an invalid area which is an area where the input image is not arranged and displayed in the output image is arranged, and the output image is converted into the invalid area. A conversion step that divides and outputs a plurality of divided images each having an area, and
A test image setting step of setting a test image in each of the invalid areas of the divided image, and
A signal processing step of processing the plurality of divided images to generate an output signal, and
A computer is made to execute a transmission step of transmitting the output signals of the plurality of divided images to a plurality of transmission lines.
The test image is an image transmission program characterized in that an arbitrary rearrangement of the plurality of divided images is a pattern that can be visually recognized from each other.

Images