JP2010226672A - Image dividing device, divided image encoder and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image dividing device and divided image encoder where all pixels to be an integer multiple of the divided image are encoded, without having to increase the bit rate. <P>SOLUTION: An ultra-high definition image is divided by the image-dividing device 102, and a padding image between a boundary line of the divided image and that of a macro block is transmitted to the divided image encoder 104 during blanking thereof, if the boundary line of the divided image is at a position different from that of the macro block. Furthermore, the padding image is extracted by the divided image encoder 104 so as to be combined into the divided image, and the image generated after the combination is encoded. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image dividing device that divides an image, a divided image encoding device that compresses and encodes a divided image, and a program.

  In recent years, ultra-high definition images (UHDTV) such as 4K × 2K images and Super Hi-Vision have been developed that greatly exceed the number of pixels of HDTV. These ultra-high-definition images have an extremely large amount of information and a remarkably high pixel rate. Therefore, it is difficult to provide an apparatus that compresses and encodes the entire ultra-high definition image at a time. Therefore, when compressing and encoding an ultra-high-definition image, the ultra-high-definition image is often divided into HDTV sizes by an image dividing device, and each divided image is encoded by an HDTV encoding device.

  Dividing a high-definition image in this way and compression-coding it with a device having a lower resolution than the original image has been performed in the past. For example, an initial encoding device for HDTV divides an HDTV image into a size smaller than SDTV or SDTV, and compresses and encodes an HDTV image by combining an encoding device for a size smaller than SDTV or SDTV. Similarly, when realizing a high-definition image encoding device, a technique is known in which an image is simply divided and encoded without providing an overlap region (see, for example, Non-Patent Document 1).

  In addition, when a high-definition image is divided and encoded, a technique of providing an overlap region in a divided image is known as a method for making a division boundary inconspicuous (see, for example, Patent Document 1). For example, on the transmission side, a high-definition image is divided after providing an overlap region of at least 32 pixels in the horizontal direction and 32 lines in the vertical direction, and each divided image is compressed and transmitted. On the other hand, the reception side receives each compression-encoded divided image and combines the overlap area after decoding, thereby reproducing a high-definition image and making the division boundary inconspicuous.

Japanese Patent Laid-Open No. 10-23403

Kurosumi, “Super Hi-Vision MPEG-2 Coding,” Video Information Media Society, vol. 61, no. 5, 2007, p. 608-611

  However, when a high-definition image is divided by providing an overlap region, there is a problem that the bit rate increases by the amount of overlap.

  In order to solve the above problem, an object of the present invention is to divide an encoding target image (for example, an ultra-high-definition image to be encoded) in a manner that does not increase the bit rate and makes the division boundary inconspicuous. It is to provide an image dividing device, a divided image encoding device, and a program.

  In order to solve the above problem, the present invention is applied when the size of an image (simple divided image) obtained by dividing the image to be encoded by the number of divisions for encoding is smaller than the encoding area of the encoding device. This technique is characterized in that an image of an adjacent simple divided image is used as a padding image added to the simple divided image.

  That is, an image dividing device according to the present invention is an image dividing device used when an encoding target image is divided and encoded, and when the encoding target image is divided, the divided image is converted into effective pixels. A divided image area determining means for determining whether to be a simple divided image having the same size as the area or an extended divided image obtained by adding a padding image to the simple divided image; and for each of the divided images determined by the divided image area determining means Whether the simple divided image or the extended divided image, and if it is the simple divided image, the simple divided image is transmitted within an image signal period, and if it is the extended divided image, A simple divided image included in the extended divided image is transmitted within an image signal period, and a padding image included in the extended divided image is transmitted within a blanking period other than the image signal period. A divided image allocating unit that separates and transmits the extended divided image, wherein the divided image region determining unit determines the size of the extended divided image, and the image width of the extended divided image is the image width of the effective pixel region ≦ expanded It is determined that the image width of the divided image <(image width of the effective pixel region + macro block width) and an integral multiple of the macro block width, and the image height of the extended divided image is the image of the effective pixel region. The height ≦ the image height of the extended divided image <(the image height of the effective pixel region + the macro block height), and an integral multiple of the macro block height is determined.

  In the image dividing apparatus according to the present invention, the blanking period is set as a horizontal blanking period and / or a vertical blanking period in the image signal.

  In the image segmentation device according to the present invention, the image size of the encoding target image is 3840 pixels × 2160 lines or 7680 pixels × 4320 lines, and the image size of the effective pixel area is 1920 pixels × 1080 lines, The size of the macroblock is 16 pixels × 16 lines.

  Moreover, in order to solve the said subject, the division | segmentation image encoding apparatus which concerns on this invention inputs the division | segmentation image from the image division | segmentation apparatus as described in any one of Claims 1-3, and the said input division | segmentation image is When the padding image is included, the padding image transmitted during the blanking period is extracted and combined to form the extended divided image, and the combined divided image is output. If not, the padding image extraction / synthesis unit that outputs the input divided image and the divided image output from the padding image extraction / synthesis unit use the padding image if it is the extended divided image. A divided image encoding unit that encodes the divided image and encodes only the information of the simple divided image when the divided image is the simple divided image. That.

  In the divided image encoding device according to the present invention, the blanking period is set as a horizontal blanking period and / or a vertical blanking period in an image signal, and the padding image extraction and synthesis unit Is extracted from the horizontal blanking period and / or the vertical blanking period.

  Furthermore, according to the present invention, the image dividing device and / or the divided image encoding device according to the present invention can be configured as a computer, and the computer can function as the image dividing device and / or the divided image encoding device according to the present invention.

  According to the present invention, it is possible to divide and encode an encoding target image (for example, an encoding target ultra-high-definition image) in a manner that does not increase the bit rate and makes the division boundary inconspicuous.

1 is an overall configuration diagram of an image dividing device and a divided image encoding device according to Embodiment 1 of the present invention. FIG. 1 is a configuration diagram of an image dividing device according to a first embodiment of the present invention. It is a block diagram of the division | segmentation image coding apparatus of Example 1 by this invention. It is a figure explaining the flow of a process of the image division | segmentation apparatus and division | segmentation image encoding apparatus of Example 1 by this invention. It is a figure showing 4 division of an image. It is a figure showing the division | segmentation of the image of Example 1 by this invention. It is a figure showing the internal process of the division | segmentation and division | segmentation image coding apparatus of Example 1 by this invention. It is a figure showing the process per macroblock. It is the figure which divided | segmented the input image into (NxM) pieces. It is a figure explaining the flow of a process of the determination method of the image height of the divided image of Example 2 by this invention. It is a figure explaining the flow of a process of the determination method of the image width of the divided image of Example 2 by this invention. It is a figure showing the division | segmentation of the image of a prior art example. It is a figure showing the internal process of the image division | segmentation of a prior art example, and a division | segmentation image coding apparatus.

  First, terms used to describe the present invention are defined.

[Definition of terms]
The number of pixels in the horizontal direction of the image is defined as “image width”, and the number of lines in the vertical direction of the image is defined as “image height”.

  The number of pixels in the horizontal direction of the macroblock used when compressing the image is defined as “macroblock width”, and the number of lines in the vertical direction of the macroblock is defined as “macroblock height”. MPEG-2 and H.264. In the H.264 system, the macroblock width is 16 pixels and the macroblock height is 16 lines.

  A high-definition image to be divided and encoded is defined as an “encoding target image”, and this size is expressed as (org_Width, org_Height). For example, a case of org_Width = 3840 and org_Height = 2160, or a case of org_Width = 7680 and org_Height = 4320 is assumed as the high-definition video.

  The effective pixel area of the encoding device used to encode the divided image (divided image) is defined as “effective pixel region of the encoding device” or simply “effective pixel region”, and the size of the effective pixel region. Is expressed as (valid_Width, valid_Height). For example, in an HDTV image encoding device, valid_Width = 1920 and valid_Height = 1080.

  A region actually encoded by the encoding device is defined as “encoding region of the encoding device” or simply “encoding region”, and the size of the encoding region is expressed as (enc_Width, enc_Height). MPEG-2 and H.264. In many encoding methods such as H.264, encoding is possible only when the image width and image height are multiples of the macroblock width and macroblock height, so that an encoding region having a size larger than the effective pixel region is encoded. . For example, MPEG-2 system or H.264 format. In an encoding apparatus for HDTV images of the H.264 system, enc_Width = 1920 and enc_Height = 1088.

  Filling the coding area outside the effective pixel area with a black signal or the like when the coding area is wider than the effective pixel area is called “padding”. An area to be filled with padding is defined as “padding area”, and an image used for padding is defined as “padding image”. For example, MPEG-2 system or H.264 format. In the H.264 encoding apparatus for HDTV, the coding area is 1920 pixels × 1088 lines, whereas the effective pixel area is 1920 pixels × 1080 lines, so that 8 lines are padded. In the decoding apparatus, after decoding the coding area, the padding area is not used, and only the effective pixel area is output. Therefore, what kind of image is used for the padding is freely determined on the coding apparatus side. be able to.

  The number of divisions for encoding the encoding target image is defined as “division number”. When the number of divisions in the horizontal direction is expressed as div_X and the number of divisions in the vertical direction is expressed as div_Y, div_X = org_Width / valid_Width and div_Y = org_Heigh / valid_Height are obtained. In order to encode the encoding target image, (div_X × div_Y) encoding devices are required.

  An image obtained by dividing the encoding target image by the number of divisions is defined as a “simple divided image”. The size of the simple divided image is equal to or smaller than the size (valid_Width, valid_Height) of the effective pixel area of the encoding device. For example, in the above-described high-definition video example (in the case of org_Width = 3840, org_Height = 2160, or org_Width = 7680, org_Height = 4320), the image is divided into 4 or 16 and encoded by the HDTV encoding device. In this case, the size of the simple divided image is equal to the size of the effective pixel area of the encoding device.

  When enc_Width> valid_Width and / or enc_Height> valid_Height, an image obtained by adding a padding image to a simple divided image is defined as an “expanded divided image”. The size of the extended divided image is equal to the size (enc_Width, enc_Height) of the encoding area of the encoding device.

  A collective term for a simple divided image and an extended divided image is defined as a “divided image”. The number of divided images is (div_X × div_Y).

  In the embodiment of the present invention, first, the configuration of an image dividing device and a divided image encoding device will be described.

[Configuration of Image Dividing Device and Divided Image Encoding Device]
FIG. 1 is an overall configuration diagram of an image dividing device 102 and a divided image encoding device 104 according to the first embodiment of the present invention. The image dividing device 102 and the divided image encoding device 104 are an encoding target image input terminal 101, an image dividing device 102, n divided image transmission signals 103-1 to 103-n, and n divided image encoding devices 104. -1 to 104-n, a control unit 105, and n code data output terminals 106-1 to 106-n.

  An encoding target image to be encoded is input from the encoding target image input terminal 101. The image dividing device 102 divides the input encoding target image into a simple divided image or an extended divided image and transmits it to the encoding device 104.

  The divided image encoding device 104 encodes the divided image transmitted from the image dividing device 102. When the divided image includes a padding image, the padding image is extracted and combined, and the combined image is encoded. Then, the code data is output from the code data output terminal 106.

  The code data is decoded for each divided image by a decoding device (not shown), and the decoded images are combined into one image.

  The control unit 105 controls the operations of the image dividing device 102 and the divided image encoding device 104.

  FIG. 2 is a configuration diagram of the image dividing device 102. The image dividing device 102 includes a divided image region determining unit 201 and a divided image assigning unit 202. Since the control unit 105 controls the operations of the divided image region determining unit 201 and the divided image assigning unit 202, a control signal from the control unit 105 is input to the divided image region determining unit 201 and the divided image assigning unit 202. .

  The divided image area determination unit 201 determines whether an image transmitted from the image dividing apparatus 102 to the divided image encoding apparatus 104 is a simple divided image or an extended divided image, and determines the size of the divided image. Whether to be a simple divided image or an extended divided image is determined by the size of the divided image and the position of the divided image, as will be described later.

  The divided image area determination unit 201 has an image width of the extended divided image, that is, an image width of the effective pixel area ≦ an image width of the extended divided image <(image width of the effective pixel area + macroblock width), and an integer of the macroblock width. The image height of the extended divided image is determined so that the image height of the effective pixel area ≦ the image height of the extended divided image <(image height of the effective pixel area + macroblock height), and It is determined to be an integral multiple of the macroblock height. On the other hand, the divided image area determination unit 201 determines the size of the simple divided image as the size of the effective pixel area.

  A padding image is added to a simple divided image to obtain an extended divided image. As described later, an image of an adjacent simple divided image is used as the padding image, and the divided images are overlapped by the amount of the padding image.

  The divided image assignment unit 202 determines whether each divided image determined by the divided image region determination unit 201 is a simple divided image or an extended divided image. If the divided image allocation unit 202 determines that the divided divided image is an extended divided image, padding in the extended divided image is performed. The image can be separated from the simple divided image in the extended divided image and transmitted to the divided image encoding device 104 at different timings.

  FIG. 3 is a configuration diagram of the divided image encoding device 104. The divided image encoding device 104 includes a padding image extraction / synthesis unit 301, a padding image memory 302, and a divided image encoding unit 303. The control unit 105 controls the operations of the padding image extraction / combination unit 301, the padding image memory 302, and the divided image encoding unit 303, so that the padding image extraction / synthesis unit 301, the padding image memory 302, and the divided image encoding unit 303 are controlled. Is supplied with a control signal from the control unit 105.

  The padding image extraction / synthesis unit 301 inputs the divided image from the image dividing device 102. When an input image has a padding image (an extended divided image is transmitted separately as a simple divided image and a padded image), a padding image transmitted during the blanking period is extracted and a padding image memory is extracted. Write to 302. Then, after outputting an image (simple divided image) other than the padding image to the divided image encoding unit 303, the padding image is read out from the padding image memory 302 and output to the divided image encoding unit 303. When the divided image does not have a padding image (a simple divided image is transmitted), the input image (simple divided image) is output to the divided image encoding unit 303 as it is.

  The padding image memory 302 temporarily stores the padding image.

  The divided image encoding unit 303 encodes the divided image input from the padding image extraction / synthesis unit 301.

  Next, the processing flow of the image dividing device 102 and the divided image encoding device 104 according to the first embodiment of the present invention will be described in detail.

[Operation of Image Dividing Device and Divided Image Encoding Device]
Here, an encoding target image (org_Width = 3840, org_Height = 2160) having an image width of 3840 pixels × image height of 2160 lines, which is often used as a 4K × 2K image, is divided into two in the horizontal direction (div_X = 2). When the effective pixel area is 1920 pixels × 1080 lines (valid_Width = 1920, valid_Height = 1080), which is the HDTV frame size, divided into two in the vertical direction (div_Y = 2), and compression encoded using MPEG-2 Will be described as an example.

  FIG. 4 is a diagram illustrating the processing flow of the image dividing device 102 and the divided image encoding device 104 according to the first embodiment of the present invention. First, in step S <b> 401, an encoding target image of 3840 pixels × 2160 lines is input to the divided image allocation unit 202.

  In step S402, the divided image assignment unit 202 performs pre-filter processing for removing noise, conversion processing of the number of samples of the color difference signal from the 4: 2: 2 format to the 4: 2: 0 format, and the like. Then, the processed image is written in a frame memory (not shown). Note that the prefilter process, the sample number conversion process, and the like are steps performed by many encoding apparatuses for improving the encoded image quality and converting the image format, and are not essential steps in the present invention.

  In step S <b> 403, the divided image area determination unit 201 determines a divided image area. In MPEG-2 video, there are layers of sequence, GOP (Group Of Pictures), pictures, slices, macroblocks, and blocks. A macroblock consists of 16 pixels × 16 lines, and motion compensation prediction is performed in units of macroblocks. The block is composed of 8 pixels × 8 lines, and DCT (Discrete Cosine Transform) and variable length coding are performed in units of blocks. In order to execute processing in units of macroblocks, the image width and image height need to be multiples of 16. However, when the input encoding target image is divided, the image width or image height of the simple divided image may not be a multiple of 16, as shown in FIG. In this case, a padding image is added to make the image width and image height a multiple of 16. The padding area is a hatched portion in FIG.

  In this embodiment, since the entire image is divided into four, the number of divided image transmission signals 103, the number of divided image encoding devices 104, and the number of code data output terminals 106 shown in FIG.

  As shown in FIG. 5, when an image of 3840 pixels × 2160 lines is divided into two parts in the horizontal direction and two parts in the vertical direction, an image [1,1], an image [1,2], an image [2,1], an image Four simple divided images of [2, 2] are generated, and each image size is 1920 pixels × 1080 lines. Since 1920 pixels × 1080 lines are the same as the size of the effective pixel area of the HDTV encoding apparatus, encoding can be performed using a conventional HDTV encoding apparatus as a divided image encoding unit.

  The 1920 pixels × 1080 lines have an image width that is a multiple of 16, but the image height is not a multiple of 16. Therefore, the divided image region determination unit 201 sets the extended image as the image [1,1] and the image [1,2]. The image width of the extended divided image is determined to be 1920 pixels so that 1920 pixels ≦ image width of the extended divided image <(1920 pixels + 16 pixels) and an integer multiple of 16. The image height of the extended divided image is determined to be 1088 lines so that 1080 lines ≦ image height of the extended divided image <(1080 lines + 16 lines) and an integral multiple of 16.

  As described later, the image [2, 1] and the image [2, 2] are left as simple divided images because the divided image encoding unit 303 provides a padding area for eight lines.

  In step S <b> 404, the divided image assignment unit 202 reads out the divided image from the frame memory (not shown) based on the region determined by the divided image region determination unit 201.

  FIG. 6 shows the allocation of divided images. For the image [1, 1], eight lines shown by diagonal lines, that is, the upper eight lines of the image [2, 1] are used as padding images. As for the image [1, 2], the eight lines shown by diagonal lines, that is, the upper eight lines of the image [2, 2] are used as padding images. The present invention is different from the prior art in that 1081 lines to 1088 lines of the original image are used as padding images, and 1081 lines to 1088 lines are overlapped.

  The difference in effect between the case of padding using the original image according to the present invention and the case of padding the image inside the encoding apparatus will be described below. When padding an image inside the encoding device, the image dividing device outputs the divided image without padding the image. Then, the image is padded so that the image size becomes a multiple of 16 inside the encoding device at the subsequent stage. FIG. 12 shows the allocation of divided images in this case. FIG. 12A shows an encoding target image, and FIGS. 12B and 12C show a case where the image [1, 1] is padded inside the encoding apparatus. When an image of 1920 pixels × 1080 lines shown in FIG. 12B is input, the encoding apparatus pads 8 lines so that the image size is a multiple of 16 as shown in FIG. The shaded area indicates the padding area). Then, the padded image of 1920 pixels × 1088 lines is encoded. Unlike the present invention, since this padding is performed inside the encoding apparatus, it is performed independently for each divided screen, and other divided images are not referred to.

  The padding of an image performed inside the encoding apparatus includes a method of performing a single color such as black and gray and a method of repeating the final line. However, neither method can refer to other divided images. Accordingly, the 8 lines padded in the image [1, 1] are different from the upper 8 lines of the image [2, 1], and similarly, the 8 lines padded in the image [1, 2] 2, 2] is different from the upper 8 lines.

  When encoding is performed in this state, since a macroblock having a padding image is encoded in a state including an image different from the original image, a large amount of code distortion occurs. Therefore, if the upper and lower images are connected after decoding, the dividing boundary line will be noticeable. However, in this embodiment, since the original image is a padding image, it is possible to reduce distortion when connecting the upper and lower divided images after decoding than when padding the image inside the encoding device.

  For the image [2, 1] and the image [2, 2], there is no divided image on the lower side, and therefore, 8 lines are padded within the divided image encoding unit 303. For this reason, a large amount of encoding distortion occurs in the lower part of the image [2, 1] and the image [2, 2], but the lower part of the image [2, 1] and the image [2, 2] is connected to another image. Since there is no division boundary line, there is no problem.

  The assignment of the divided images of the image [1,1] and the image [2,1] will be described with reference to FIG. The shaded area is the padding area. As shown in FIG. 7B, the image of 1080 lines (effective pixel area) is transmitted to the divided image encoding device 104 during the image signal period as in the conventional case. The padding image is transmitted to the divided image encoding device 104 during the blanking period using the blanking period. That is, the effective pixel area and the padding area are separated and transmitted. Then, as shown in FIG. 7C, the divided image encoding device 104 synthesizes the images transmitted during the blanking period.

  When an image is transmitted from the image dividing device 102 to the divided image encoding device 104, a vertical blanking signal indicating the start of the frame is transmitted for each frame, and a horizontal blanking signal indicating the start of the line for each line. Is sent out. In the case of HD-SDI (High Definition Serial Digital Interface), the vertical blanking period in which the vertical blanking signal is transmitted is 45 lines, and the horizontal blanking period in which the horizontal blanking signal is transmitted is 280 pixels. Accordingly, when an image divided into HDTV image sizes is transmitted to the divided image encoding device 104 using HD-SDI, the padding image can be sufficiently transmitted in the blanking period.

  The allocation of the divided images of the image [1,1] and the image [2,1] will be described with reference to FIG. 13 in the case where the image is padded inside the encoding device without using the divided image device according to the present invention. To do. In this case, when assigning divided images, the divided image device transmits the divided images to the divided image encoding device without padding the images. Then, as shown in FIG. 13B, the divided image encoding device performs image padding.

  In step S405, the padding image extraction / combination unit 301 determines whether there is a padding image. If an image is inserted during the blanking period, the image is determined as a padding image (S405-Yes), and the process proceeds to step S406. If no image is inserted during the blanking period, it is determined that there is no padding image (S405-No), and the process proceeds to step S407.

  In step S <b> 406, the padding image extraction / combination unit 301 extracts a padding image and writes it to the padding image memory 302. Then, after outputting an image other than the padding image to the divided image encoding unit 303, the padding image is read out from the padding image memory 302 and output to the divided image encoding unit 303. Note that the padding image memory may be unnecessary depending on the position of the blanking period for transmitting the padding image.

  In step S407, the padding image extraction / synthesis unit 301 transmits the input image as it is to the divided image encoding unit 303. Then, padding for 8 lines is performed by the divided image encoding unit 303. The padding image is the same as that of the conventional encoding device, and a single color such as black or gray is used or the last line is used repeatedly.

  In step S408, the divided image encoding unit 303 performs encoding (DCT, quantization, motion compensation prediction, variable length encoding, etc.) on the input image, and outputs encoded data from the image output terminal 105. . As the divided image encoding unit 303, a conventional HDTV encoding apparatus can be used. At the time of encoding, the size of the padding image is the same whether it is the padding image extracted in step S406 (padding according to the present invention) or the padding image internally padded in step 407 (conventional method). Therefore, even if the present invention is applied, the bit rate after encoding is substantially equal to the conventional one.

  When the divided image is a simple divided image and an image of 1080 lines without a padding image is input, the HDTV encoding apparatus performs padding for 8 lines so as to be a multiple of 16, and performs encoding as 1088 lines. Therefore, even if the image dividing apparatus 102 performs padding for 8 lines and inputs a 1088 line divided image, the bit rate does not increase.

  Furthermore, since the overlap region remains only for the padding, there is no effect when the decoded images are combined. That is, the number of pixels corresponding to the number of division coding units can be transmitted as an effective image, and if four HDTV coding devices are used, all pixels of a 4K × 2K image (3840 pixels × 2160 lines) are coded. It is possible.

  In addition, the divided and encoded image can be decoded using a conventional decoding device. This is because the decoding device truncates the padding area and outputs only the effective image. Therefore, no matter how the padding portion is created, the decoding apparatus and the apparatus that combines the divided decoded images at the subsequent stages are not affected.

  Next, an encoding target image (org_Width = W, org_Height = H) having an image width of W pixels and an image height of H lines is divided into M (div_X = M) in the horizontal direction and N (div_Y = N) in the vertical direction. This case will be described as a second embodiment. In this case, since (N × M) divided images are generated, the number of divided image transmission signals 103, the number of divided image encoding devices 104, and the number of code data output terminals 106 shown in FIG. ) It becomes a piece.

  FIG. 9 shows a diagram in which the encoding target image is divided into (N × M) pieces. A divided image at an arbitrary position is represented as [x, y]. In order to divide the image height into N, 1 ≦ x ≦ N, and in order to divide the image width into M, 1 ≦ y ≦ M.

  The operation in the second embodiment is the same as that in FIG. Among these, the determination of the divided image area in step S403 will be described.

  FIG. 10 is a flowchart for explaining the processing flow of the method for determining the image height of a divided image. In step S1001, it is determined whether the image height (H / N) of the simple divided image is divisible by 16. If not divisible (step S1001-Yes), the process proceeds to step S1002. If divisible (step S1001-No), the process proceeds to step S1004.

  In step S1002, it is determined whether or not the value of x in the divided image [x, y] is N. If x is other than N (step S1002-Yes), the process proceeds to step S1003. If x is N (step S1002-No), the process proceeds to step S1004.

  In step S1003, a padding image is added to the simple divided image to obtain an extended divided image. The image height of the extended divided image is determined so that the image height of the effective pixel region ≦ the image height of the extended divided image <(image height of the effective pixel region + 16 lines) and an integer multiple of 16.

  In step S1004, if (H / N) is divisible by 16, or x = N, that is, if there is no divided image on the lower side, the divided image assignment unit 202 is instructed not to pad the line.

  FIG. 11 is a flowchart for explaining the processing flow of the method for determining the image width of a divided image. In step S1101, it is determined whether the image width (W / M) of the simple divided image is divisible by 16. If not divisible (step S1101-Yes), the process proceeds to step S1102, and if divisible (step S1101-No), the process proceeds to step S1104.

  In step S1102, it is determined whether or not the y value of the divided image [x, y] is M. If y is other than M (step S1102-Yes), the process proceeds to step S1103. If y is M (step S1102-No), the process proceeds to step S1104.

  In step S1103, a padding image is added to the divided image to obtain an extended divided image. The image width of the extended divided image is determined so that the image width of the effective pixel area ≦ the image width of the extended divided image <(image width of the effective pixel area + 16 pixels) and an integer multiple of 16.

  In step S1104, when (W / M) is divisible by 16, or when y = M, that is, when there is no divided image on the right side, the divided image assignment unit 202 is instructed not to pad the pixel.

  Thus, in the present embodiment, a divided image having an arbitrary size can be generated and encoded from an arbitrary image size.

  In the above-described embodiment, the case where the size of the macroblock is 16 pixels × 16 lines has been described. However, when the size of the macroblock is a pixel × b lines, similarly, the image width of the divided image is a multiple of a. The padding is performed so that the image height of the divided image is a multiple of b.

  In the above-described embodiment, only the case where the division number div_X (org_Width / valid_Width) in the horizontal direction and the division number div_Y (org_Heigh / valid_Height) in the vertical direction are divisible is explained. It is also possible to apply the present invention by filling in. However, in such a case, better image quality can be obtained by overlapping without applying the present invention.

  Here, a computer can be preferably used to function as the image dividing device 102, and such a computer centralizes the control unit 105 for causing the divided image region determining unit 201 and the divided image assigning unit 202 to function. This can be realized by a storage unit including an arithmetic processing unit (CPU) and at least one memory.

  Furthermore, the functions of the divided image area determining unit 201 and the divided image assigning unit 202 can be realized by causing such a computer to execute a predetermined program by the CPU. Furthermore, a program for realizing the functions of the divided image area determination unit 201 and the divided image allocation unit 202 can be stored in a predetermined area of the storage unit (memory) described above. Such a storage unit can be constituted by a RAM or the like inside the computer, or can be constituted by an external storage device (for example, a hard disk). Further, such a program can be constituted by a part of software (stored in a ROM or an external storage device) on an OS used by a computer as the image dividing device 102.

  Furthermore, a program for causing a computer that functions as the image dividing device 102 to function as means as each component of the present invention can be recorded on a computer-readable recording medium.

  Similarly, in order to function as the divided image encoding device 104, a computer can be suitably used. Such a computer includes a padding image extraction / synthesis unit 301, a padding image memory 302, and a divided image encoding unit 303. The control unit 105 for functioning can be realized by a central processing unit (CPU) and a storage unit including at least one memory.

  Furthermore, the functions of the padding image extraction / synthesis unit 301, the padding image memory 302, and the divided image encoding unit 303 can be realized by causing such a computer to execute a predetermined program by the CPU. Furthermore, a program for realizing the functions of the padding image extraction / synthesis unit 301, the padding image memory 302, and the divided image encoding unit 303 can be stored in a predetermined area of the storage unit (memory). Such a storage unit can be constituted by a RAM or the like inside the computer, or can be constituted by an external storage device (for example, a hard disk). Further, such a program can be constituted by a part of software (stored in a ROM or an external storage device) on an OS used by a computer as the image dividing device 102.

  Furthermore, a program for causing a computer that functions as the divided image encoding device 104 to function as means as each component of the present invention can be recorded on a computer-readable recording medium.

  Each of the above-described embodiments has been described as a representative example. However, many modifications and replacements can be made within the spirit and scope of the present invention, and each embodiment can be combined to realize another embodiment. It will be apparent to those skilled in the art that Therefore, the present invention should not be construed as being limited by the above-described embodiments, and various modifications and changes can be made without departing from the scope of the claims.

  As described above, according to the present invention, the encoding target image (for example, the ultra-high definition image to be encoded) is divided and encoded in such a manner that the bit rate is not increased and the division boundary is inconspicuous. Therefore, it is useful for an arbitrary application in which an image is divided and encoded.

DESCRIPTION OF SYMBOLS 101 Encoding target image input terminal 102 Image dividing device 103-1 to 103-n Division image transmission signal 104-1 to 104-n Division image encoding device 105 Control unit 106-1 to 106-n Code data output terminal 201 Division Image area determination unit 202 Divided image allocation unit 301 Padding image extraction / synthesis unit 302 Padding image memory 303 Divided image encoding unit

Claims (7)

An image dividing device used when dividing and encoding a coding target image,
Divided image area determination that determines whether the divided image is a simple divided image having the same size as the effective pixel area or an extended divided image obtained by adding a padding image to the simple divided image when the encoding target image is divided Means,
For each of the divided images determined by the divided image region determining means, it is determined whether the divided image is the simple divided image or the extended divided image. If the divided divided image is the simple divided image, the simple divided image is included in the image signal period. And when it is the extended divided image, the simple divided image included in the extended divided image is transmitted within the period for the image signal, and the padding image included in the extended divided image is transmitted outside the period for the image signal. Divided image assigning means for transmitting within the blanking period and separating and transmitting the extended divided image,
The divided image area determining means determines the size of the extended divided image, and the image width of the extended divided image, the image width of the effective pixel area ≦ the image width of the extended divided image <(image width of the effective pixel area + macro. Block width) and an integral multiple of the macroblock width, and the image height of the extended divided image is the image height of the effective pixel region ≦ the image height of the extended divided image <(the effective pixel region) Image height + macroblock height) and an integer multiple of the macroblock height.   The image dividing apparatus, wherein the blanking period is set as a horizontal blanking period and / or a vertical blanking period in an image signal. The image size of the encoding target image is 3840 pixels × 2160 lines or 7680 pixels × 4320 lines,
The image size of the effective pixel area is 1920 pixels × 1080 lines,
The image dividing apparatus according to claim 1 or 2, wherein the size of the macroblock is 16 pixels x 16 lines. A divided image encoding device used when an encoding target image is divided and encoded,
When the divided image from the image dividing device according to any one of claims 1 to 3 is input and the input divided image includes the padding image, the transmitted image is transmitted during the blanking period. A padding image extracting and synthesizing unit for outputting the input divided image when the input divided image does not have the padding image; ,
When the divided image output from the padding image extraction / synthesis unit is the extended divided image, the divided image is encoded using the padding image, and when the divided image is the simple divided image, the simple divided image is output. A divided image coding apparatus comprising: a divided image coding unit that performs coding using only image information. The blanking period is set as a horizontal blanking period and / or a vertical blanking period in the image signal,
5. The divided image encoding apparatus according to claim 4, wherein the padding image extraction / synthesis unit extracts the padding image from the horizontal blanking period and / or the vertical blanking period. In a computer that functions as an image dividing device used when dividing and encoding an image to be encoded,
(A) When dividing the encoding target image, a step of determining whether the divided image is a simple divided image having the same size as the effective pixel area or an extended divided image obtained by adding a padding image to the simple divided image When,
(B) For each of the divided images determined in step (a), it is determined whether the image is the simple divided image or the extended divided image. If the divided image is the simple divided image, the simple divided image is included in the image signal period. And the simple divided image included in the extended divided image is transmitted within the image signal period, and the padding image included in the extended divided image is transmitted to the image signal period. A program for performing transmission within a blanking period other than the step of separating and transmitting the extended divided image,
In the step (a), the size of the extended divided image is set. Regarding the image width of the extended divided image, the image width of the effective pixel region ≦ the image width of the extended divided image <(image width of the effective pixel region + macro block width). ), And the image height of the extended divided image is determined to be an integral multiple of the macroblock width. The image height of the effective pixel region ≦ the image height of the extended divided image <(the image of the effective pixel region) (Height + macroblock height) and a program that is determined to be an integral multiple of the macroblock height. In a computer functioning as a divided image encoding device used when dividing and encoding an image to be encoded,
(C) When the divided image sent in step (b) is input and the divided image is a divided image having the padding image, the padding image transmitted during the blanking period is Extract and combine and output to form the extended split image,
When other divided images are input, a step of outputting the divided images;
(D) If the divided image output by the step (c) is a divided image having the padding image, the divided image is encoded using the padding image; And a step of performing encoding using only image information.

Images