JP6272689B2 - Bit depth decomposition allocation apparatus, bit depth combination restoration apparatus, image encoding apparatus, image decoding apparatus, and programs thereof - Google Patents

Description

  The present invention relates to a technique for converting one high bit depth image into a plurality of low bit depth images and transmitting them, and in particular, bit depth decomposition for decomposing and assigning a high bit depth image signal for a plurality of low bit depth image signals. Allocation device, bit depth combination restoration device that combines multiple low bit depth image signals for high bit depth image signal, decomposes and transforms bit depth of high bit depth image signal to generate multiple low bit depth image signals The present invention relates to an image encoding device that performs encoding, an image decoding device that decodes and combines a plurality of encoded low bit depth image signals to restore a high bit depth image, and a program thereof.

  In a current general image encoding apparatus, the bit depth of an image to be encoded is 8 bits or 10 bits, and in particular, only 8 bits are often targeted. On the other hand, 10-bit, 12-bit, or 16-bit images are handled as high bit depth images. The techniques for encoding these high bit depth images by the general image encoding apparatus are roughly divided into a technique for mapping and encoding a high bit depth image to a low bit depth image, and a high bit depth. Two types of techniques for separating and encoding an image into a plurality of low bit depth images are known.

(A technique for mapping a high bit depth image to a low bit depth image)
A technique for mapping and encoding a high bit depth image to a low bit depth image is to map a high bit depth image to a low bit depth image and to encode the low bit depth image by the general image encoding device. It is basic. On the decoding side, a low bit depth image is decoded and returned to a high bit depth image by inverse mapping. There are various techniques for a mapping method and a technique for transmitting auxiliary information regarding a difference between a high bit depth image of an original picture and a decoded high bit depth image from the encoding side to the decoding side.

  For example, for a high bit depth image, a low bit depth image is obtained by the Lloyd-Max quantization method based on the histogram, the obtained low bit depth image signal is encoded, and a codebook of quantization representative values is encoded. There is known a technique for transmitting data after conversion (see, for example, Patent Document 1). In this Patent Document 1, a typical bit depth transform coding technique is disclosed as a conventional technique.

(Technique that separates and encodes high bit depth images into multiple low bit depth images)
On the other hand, as an example of a technique for separating and encoding a high bit depth image into a plurality of low bit depth images, a high bit depth image signal is separated into a plurality of low bit depth image signals, and the general image A technique for encoding using a plurality of encoding apparatuses is known (see, for example, Non-Patent Document 1). In the technique of Non-Patent Document 1, a monochrome 16-bit depth image signal is separated into 8 bits of the most significant bit (MSB) side and 8 bits of the least significant bit (LSB) side, and the MSB side 8 bits are used as a luminance signal. The 8 bits on the LSB side are assigned to the color difference signal and encoded as a color signal. On the decoding side, the low bit depth image signal encoded by the luminance signal / color difference signal is decoded, and the MSB side 8 bits are set to the upper side and the LSB side 8 bits are set to the lower side, thereby simply combining, Restore a monochrome 16-bit depth image.

Japanese Patent No. 5027171

E. Francois, et al., “AHG18: On 16-bits support for Range Extensions”, JCTVC-N0142, JCT-VC, MPEG-H, July 23, 2013, [online], [November 21, 2013 Search], Internet <URL: http://phenix.int-evry.fr/jct/doc_end_user/current_document.php?id=7858>

  In the prior art, in the technique of mapping a high bit depth image to a low bit depth image and encoding, in addition to the general encoding device, a mapping / inverse mapping device and a residual signal of a high bit depth image are used. There is a problem that a device for encoding is required and the configuration becomes complicated.

  On the other hand, in the technique of separating and encoding a high bit depth image into a plurality of low bit depth images, in addition to the general encoding device, it is possible to add a very simple device such as bit decomposition / combination. It is simple. However, this technique has a problem that the image quality of a high bit depth image at the time of restoration is likely to deteriorate.

  For example, the principle and problems related to the technique of Non-Patent Document 1 are shown in FIG. In the technique of Non-Patent Document 1, an original image (16-bit image signal) of a 16-bit depth monochrome image (FIG. 6A) is “separated” into 8 bits on the MSB side and 8 bits on the LSB side (FIG. 6B )), Each of the two 8-bit image signals obtained (FIG. 6C) is subjected to image coding processing (FIG. 6D) to obtain a coded stream (FIG. 6E). On the decoding side, image decoding processing is performed on the encoded stream (FIG. 6F), and two 8-bit decoded image signals are obtained (FIG. 6G). In general, in an encoded image, deterioration is more likely to appear in the lower bits than in the upper bits. Therefore, in FIG. 6G, coding degradation is schematically shown as the color density (the darker the color, the greater the degradation). Next, with respect to two 8-bit decoded image signals, the MSB side 8 bits are set to the upper side, the LSB side 8 bits are set to the lower side, and “simple combination” is performed (FIG. 6 (H)). The monochrome image having a 16-bit depth is obtained as an output (FIG. 6 (I)).

  Generally, the encoded image quality decreases as the higher order bits deteriorate. For this reason, the restored high bit depth image obtained by the technique of Non-Patent Document 1 is higher in the upper 8 bits than the lower bits in the lower 8 bits, as represented by the color shading in FIG. There is a problem that deterioration of the lower bits becomes large (is likely to become large).

  The object of the present invention has been made in view of the above-mentioned problems, and performs bit depth conversion between a high bit depth image and a plurality of low bit depth images, and image quality degradation at the time of restoration of the high bit depth image. As a device for reducing image quality, a bit depth decomposition allocation device that decomposes and allocates a high bit depth image signal for a plurality of low bit depth image signals, and combines a plurality of low bit depth image signals for a high bit depth image signal Bit depth combination restoration apparatus, image coding apparatus that decomposes and converts bit depth of high bit depth image signal to generate and encode a plurality of low bit depth image signals, and a plurality of encoded low bit depth image signals Are to decode and combine them to restore a high bit depth image and to provide these programs.

The bit depth decomposition allocation apparatus of the present invention decomposes an image signal of a high bit depth image having a predetermined bit depth into image signals of a plurality of low bit depth images having a bit depth lower than the predetermined bit depth. A bit depth decomposition allocation apparatus to allocate, wherein each bit in the image signal of the high bit depth image is decomposed into a plurality of types in a predetermined bit unit, and each of the plurality of types of decomposed bits corresponds to the plurality of types Bit depth decomposition assigning means for assigning the image signal of the low bit depth image, and maintaining the depth order from the most significant bit side in the image signal of the high bit depth image for each assigned bit, Low bit depth signal generation means for generating each of the image signals of the plurality of low bit depth images by arranging in order from the position; With respect to the plurality of low bit depth image, in the image coding process in the plurality of types of compression ratio is expected, the bit depth decomposition assigning means, bit image signal of the high bit depth image For the depth, each bit of the bit area classified into three types of bit areas, upper bit, middle bit, and lower bit, is decomposed into a plurality of types according to the number of the plurality of types, and the plurality of types of the decomposed bits Are assigned to image signals of a plurality of low bit depth images corresponding to the plurality of types, giving priority to the bit order from the higher order, and the bits belonging to the bit region of the upper bits are the plurality of types of compression. Allocated for the image signal of the low bit depth image for which the image compression processing with the lowest compression rate is scheduled, and the bits belonging to the bit area of the lower bits are the plurality of types Allotted to the image signal of the low bit depth image for which the image coding process with the highest compression rate is scheduled, and the bit belonging to the bit region of the middle bit is the highest among the plurality of types of compression rates in order of image encoding processing low compression ratio is scheduled, it characterized that you have been configured to sequentially allocated to the image signal of the plurality of low bit depth image in bits.

  In the bit depth decomposition allocation apparatus of the present invention, the bit depth decomposition allocation unit decomposes a bit position with a remainder modulo n when the image signal of the high bit depth image is decomposed into n types. And means for allocating the decomposed n types of bits for image signals of a plurality of low bit depth images corresponding to the n types.

Furthermore, the bit depth combination restoration apparatus of the present invention uses a plurality of low bit depth image signals having a predetermined bit depth as image signals of high bit depth images having a bit depth higher than the predetermined bit depth. A bit depth combining / reconstructing apparatus for combining, a low bit depth signal receiving means for receiving the image signals of the plurality of low bit depth images, and decomposing each bit of the received image signals of the plurality of low bit depth images Bit depth combination for reconstructing the image signal of the high bit depth image and restoring the high bit depth image by maintaining and combining the depth order from the most significant bit side in the image signal of the high bit depth image comprising a restoring means, wherein the plurality of low bit-depth image is intended image decoding process in the plurality of types of compression ratio is performed, the plurality of low bit depth image Each image signal is a bit depth of an image signal of a high bit depth image. Each bit of the bit area classified into three types of bit areas, upper bit, middle bit, and lower bit, depends on the number of the plurality of types. In the case of being decomposed into a plurality of types, the bit depth combination restoration unit performs the high bit depth image on the image signals of the plurality of low bit depth images according to the image decoding processing of the plurality of types of compression rates. In order to reconstruct the image signal, a bit belonging to the bit region of the upper bit is extracted from the image signal subjected to the image decoding process with the lowest compression rate among the plurality of types of compression rates, and the plurality of types A bit belonging to the bit region of the lower bit is extracted from an image signal subjected to image decoding processing having the highest compression rate among the compression rates of the plurality of compression rates. The image signal of the high bit depth image is restored by extracting bits belonging to the bit area of the middle bit in order from the image signal subjected to the image decoding process with a low reduction ratio. and wherein the Tei Rukoto.

  Further, in the bit depth combination restoration apparatus of the present invention, each of the plurality of low bit depth images is obtained by decomposing bit positions into n types with a remainder modulo n for the image signals of the high bit depth images. In this case, the bit depth combination restoration unit decomposes the image signals of the plurality of low bit depth images so as to correspond to the n types of decomposed bits, and uses the image signals for the high bit depth images. It has the means to couple | bond, It is characterized by the above-mentioned.

Furthermore, it is a program for causing a program according to an aspect of the present invention to function as the bit depth decomposition allocation apparatus of the present invention .

Furthermore, it is a program for causing a program according to another aspect of the present invention to function as the bit depth combination restoration apparatus of the present invention .

  According to the present invention, when one high bit depth image is converted into a plurality of low bit depth images and transmitted, it is possible to suppress image quality deterioration of the restored high bit depth image with a simple configuration. .

It is a block diagram which shows the outline of an image coding system provided with the bit depth decomposition | disassembly allocation apparatus of one Embodiment by this invention, and two low bit depth signal encoding apparatuses. It is a block diagram which shows the outline of an image decoding system provided with the bit depth joint decompression | restoration apparatus of one Embodiment by this invention, and two low bit depth signal decoding apparatuses. (A)-(I) is explanatory drawing which shows schematic operation | movement of one Example in the bit depth decomposition | disassembly allocation apparatus and bit depth joint decompression | restoration apparatus of one Embodiment by this invention. It is a block diagram which shows the outline of the image coding apparatus of one Embodiment by this invention. It is a block diagram which shows the outline of the image decoding apparatus of one Embodiment by this invention. (A)-(I) is explanatory drawing which shows the principle and subject in a prior art.

First, in order to facilitate understanding of the present invention, terms are defined as follows.
・ High bit depth image: Input high bit depth image
(Image composed of M-bit depth image signal)
Low bit depth image A: One low bit depth when a high bit depth image is decomposed into two
Images (images composed of A-bit depth image signals)
Low bit depth image B: the other low bit depth when the high bit depth image is decomposed into two
Images (images composed of B-bit depth image signals)
Each bit is called M1, M2,... In order from the MSB of the high bit depth image.
Each bit is called a1, a2,... In order from the MSB of the low bit depth image A.
Each bit is called b1, b2,... Sequentially from the MSB of the low bit depth image B.

  In each embodiment described below, an example in which a 16-bit depth high bit depth image is mainly decomposed into two 8-bit depth low bit depth images A and B is shown.

  First, a bit depth decomposition allocation apparatus and a bit depth combination restoration apparatus according to the first embodiment of the present invention will be described.

<First Embodiment>
[Example 1]
In the bit depth decomposition allocation apparatus and the bit depth combination restoration apparatus according to the first embodiment, a plurality of examples are assumed. First, Example 1 will be described. In the first embodiment, out of an image signal of a high bit depth image (M bit depth image signal), an odd bit and an even bit are two low bit depth image signals (A bit depth image signal and B bit depth image signal). This is an example of disassembling alternately.

(Image coding system provided with bit depth decomposition allocation apparatus)
FIG. 1 shows a block diagram of an image encoding system 10 including a bit depth decomposition allocation apparatus 20 and two low bit depth signal encoding apparatuses 30 and 40 according to an embodiment of the present invention.

  First, an image signal (M bit depth image signal) of a 16-bit high bit depth image is input to the bit depth decomposition allocation apparatus 20. The bit depth decomposition allocation apparatus 20 is an apparatus that decomposes and allocates a high bit depth image signal for a plurality of low bit depth image signals, and includes a bit depth decomposition allocation unit 21 and low bit depth signal generation units 22-1 and 22-. 2 and an auxiliary information generation unit 23.

  The bit depth decomposition allocation unit 21 decomposes each bit of the M bit depth image signal into two types in units of bits designated in advance, and converts the two types of decomposed bits into two low bit depth images corresponding to the two types. Allotted for image signals and output to low bit depth signal generation units 22-1 and 222-2, respectively. In the first embodiment, the bit depth decomposition allocation unit 21 decomposes the M-bit depth image signal with odd bits and even bits, and alternately separates the decomposed odd bits and even bits for the image signals of two low bit depth images. Allocate and output to low bit depth signal generators 22-1 and 222-2, respectively. For example, when a 16-bit depth image signal (M1, M2, M3, M4,..., M16) in a high bit-depth image is decomposed and assigned, odd bits M1, M3, M5,. .., A8 are allocated to the image signals (a1, a2,..., A8) of the low bit depth image A in the unit 22-1 and the even bits M2, M4, M6,. Assigned for the image signal (b1, b2,..., B8) of the low bit depth image B.

  The low bit depth signal generation unit 22-1 maintains the depth order from the MSB side for the bits (M1, M3, M5,..., M15) of the allocated M bit depth image signal, and sequentially from the upper bit position. By arranging, image signals (a1, a2,..., A8) of the low bit depth image A are generated and output to the low bit depth signal encoding device 30 as a bit stream.

  The low bit depth signal generation unit 22-2 maintains the depth order from the MSB side for the bits (M2, M4, M6,..., M16) of the allocated M bit depth image signal, and sequentially from the upper bit position. By arranging, image signals (b1, b2,..., B8) of the low bit depth image B are generated and output to the low bit depth signal encoding device 40 as a bit stream.

  The auxiliary information generation unit 23 generates decomposition allocation information indicating a decomposition allocation method when decomposing and allocating the image signal (A bit depth image signal and B bit depth image signal) from the M bit depth image signal to the low bit depth image signal. Obtained from the bit depth decomposition allocation unit 21 and output to the outside as an auxiliary information stream. The auxiliary information may be transmitted without being encoded, or may be configured to be transmitted after being encoded. However, the function of the auxiliary information generation unit 23 is unnecessary when the bit decomposition and allocation method is determined in advance between transmission and reception (between the encoding side and the decoding side).

  The low bit depth signal encoding device 30 receives the bit stream of the image signal of the low bit depth image A (A bit depth image signal) and performs an image encoding process for 8 bit depth to thereby generate the first encoded stream. Is generated and output to the outside. Here, the image coding process for 8-bit depth can be an arbitrary image coding process including a general image coding process.

  The low bit depth signal encoding device 40 receives the bit stream of the image signal of the low bit depth image B (B bit depth image signal) and performs an image encoding process for 8 bit depth to generate the second encoded stream. Is generated and output to the outside. Similarly to the above, the image coding process for 8-bit depth can be an arbitrary image coding process including a general image coding process.

  As described above, the bit depth decomposition / assignment apparatus 20 according to the present exemplary embodiment decomposes and assigns an M-bit depth image signal to an image signal (A-bit depth image signal and B-bit depth image signal) of a low-bit depth image. The upper bits in the M-bit depth image signal are assigned to the upper bits with low coding degradation of the low bit depth image A and the low bit depth image B (that is, the depth order from the MSB side in the M bit depth image signal is maintained) And assigning them by arranging them in order from the upper bit position).

(Image decoding system provided with bit depth combination restoration device)
FIG. 2 shows a block diagram of an image decoding system 50 including a bit depth joint restoration apparatus 80 and two low bit depth signal decoding apparatuses 60 and 70 according to an embodiment of the present invention. The image decoding system 50 is configured to input each encoded stream transmitted from the image encoding system 10, decode and combine the image signals in the respective encoded streams, and restore a high bit depth image. .

  The low bit depth signal decoding device 60 performs image decoding processing corresponding to the image encoding processing in the 8-bit low bit depth signal encoding device 30 on the input first encoded stream, thereby reducing the low bit depth. The bit stream of the image signal (A bit depth image signal) of the depth image A is decoded and output to the low bit depth signal receiving unit 81-1 in the bit depth combination restoration apparatus 80.

  The low bit depth signal decoding device 70 performs low-bit depth by performing image decoding processing corresponding to the image encoding processing in the 8-bit low bit depth signal encoding device 40 on the input second encoded stream. The bit stream of the image signal (B bit depth image signal) of the depth image B is decoded and output to the low bit depth signal receiving unit 81-2 in the bit depth combination restoration apparatus 80.

  The bit depth combination restoration apparatus 80 is an apparatus for combining a plurality of low bit depth image signals for a high bit depth image signal, and includes a low bit depth signal reception unit 81-1 and 81-2, a bit depth combination restoration unit 82, and An auxiliary information receiving unit 83 is provided.

  The low bit depth signal reception unit 81-1 receives the image signals (a1, a2,..., A8) of the low bit depth image A and outputs them to the bit depth combination restoration unit 82. Similarly, the low bit depth signal reception unit 81-2 receives the image signals (b1, b2,..., B8) of the low bit depth image B and outputs them to the bit depth combination restoration unit 82.

  The auxiliary information receiving unit 83 acquires auxiliary information including bit decomposition / allocation information related to the image signals of the low bit depth images A and B from the auxiliary information stream transmitted from the image encoding system 10 and combines the bit depths. The data is output to the restoration unit 82. The auxiliary information may be extracted from the auxiliary information stream only if it is transmitted without being encoded, or may be extracted from the auxiliary information stream by performing a corresponding decoding process if it is transmitted after being encoded. Can do. Further, when the bit decomposition and allocation method is determined in advance between transmission and reception (between the encoding side and the decoding side), the function of the auxiliary information receiving unit 83 is not necessary.

  The bit depth combining / restoring unit 82 decomposes each bit of the image signals of the low bit depth images A and B received by the low bit depth signal receiving units 81-1 and 81-2, and converts these bits into auxiliary information. Are combined according to the predetermined bit allocation shown in FIG. 1 (ie, the depth order from the MSB side in the M-bit depth image signal is maintained), and the M-bit depth image signal is reconstructed to obtain a 16-bit Restore high bit depth images.

  FIG. 3 is an explanatory diagram of a schematic operation in the bit depth decomposition allocation apparatus 20 and the bit depth combination restoration apparatus 80 in Example 1 of the first embodiment. In FIG. 3, on the transmitting side (encoding side), a 16-bit high bit depth image is decomposed into two 8-bit low bit depth images A and B, subjected to image encoding processing, and transmitted. An example is shown in which a high bit depth image is restored by performing image decoding processing on two low bit depth images on the side (decoding side).

  Referring to FIG. 3, as Example 1 in the first embodiment according to the present invention, for a 16-bit depth color or monochrome original image (16-bit image signal) (FIG. 3A), bits are sequentially transmitted from the MSB side. Then, it is “decomposed” into odd bits and even bits and assigned to two 8-bit image signals (FIG. 3B), and the resulting 8-bit image signal (FIG. 3C) is subjected to image coding processing. (FIG. 3D), an encoded stream (FIG. 3E) is obtained. On the decoding side, image decoding processing is performed on the encoded stream (FIG. 3F), and two 8-bit decoded image signals are obtained (FIG. 3G). As described above with reference to FIG. 6, in general, in the encoded image, deterioration is more likely to appear in the lower bits than in the upper bits. Therefore, also in FIG. 3G, coding deterioration is schematically shown as the color density (the darker the color, the greater the deterioration). Next, for the two 8-bit decoded image signals, the bits are sequentially “decomposed” into odd bits and even bits from the MSB side, and “alternately combined” (FIG. 3 (H)), so that the 16-bit depth The image signal is reconstructed and a 16-bit depth image is restored and obtained as an output (FIG. 3 (I)).

  In the present embodiment, with respect to decomposition allocation information indicating a decomposition allocation method used when decomposing and allocating an image signal of an M bit depth to an image signal of a low bit depth image, transmission and reception (between the encoding side and decoding side) are performed in advance. In the case of setting, it is not always necessary to provide the auxiliary information generating unit 23 and the auxiliary information receiving unit 83, and it is not necessary to transmit or receive auxiliary information.

  As described above, according to Example 1 of the first embodiment of the present invention, the configuration is simple, and the encoding of the upper bits in the lower 8 bits in the 16-bit image signal as in the technique of Non-Patent Document 1 is performed. It is possible to solve the problem that the encoding deterioration of the lower bits in the upper 8 bits becomes larger (is likely to become larger) than the deterioration. Accordingly, image quality degradation is concentrated on the LSB side in a 16-bit image signal, and therefore, even when a high bit depth image signal is encoded using a plurality of low bit depth image signals, the image quality of the restored high bit depth image is determined. Can be improved.

[Example 2]
Next, Example 2 will be described in the bit depth decomposition allocation apparatus 20 and the bit depth combination restoration apparatus 80 of the first embodiment. In the second embodiment, when the two low bit depth signal encoding devices 30 and 40 perform image encoding processing at different compression rates, the upper bit, the middle bit, and the lower bit of the bit depth of the M bit depth image signal are described. In this example, the bits are classified into three types of bit areas, and each bit in the three types of bit areas is divided into a plurality of types according to the number of the plurality of types and assigned.

  In this example, the low bit depth signal encoding device 30 has an image encoding process for a luminance signal (a signal having a compression rate lower than that of the color difference signal), and the low bit depth signal encoding device 40 has a color difference signal (based on the luminance signal). In the following description, it is assumed that an image encoding process for a signal having a high compression rate is provided. It should be noted that in a general image coding apparatus, control is often performed so that the coding deterioration of the color difference signal is larger than the coding deterioration of the luminance signal. This is because the deterioration of the color difference signal is less visually noticeable to humans than the deterioration of the luminance signal.

(Image coding system provided with bit depth decomposition allocation apparatus)
Also in the second embodiment, as illustrated in FIG. 1, the bit depth decomposition allocation apparatus 20 is provided in an image encoding system 10 including two low bit depth signal encoding apparatuses 30 and 40. However, in the second embodiment, the operation of the bit depth decomposition allocation unit 21 is different from that in the first embodiment.

  Referring to FIG. 1, first, an image signal of a 16-bit high bit depth image (M bit depth image signal) is input to the bit depth decomposition allocation apparatus 20.

  The bit depth decomposition allocation unit 21 classifies the bit depth of the M-bit depth image signal into three types of bit areas, upper bits, middle bits, and lower bits, and each bit of the bit areas classified into the three types. The data is decomposed into a plurality of types according to the number of types related to the compression rate, and the bits are assigned to the low bit depth signal generation units 22-1 and 222-2 and output. More specifically, the upper bits of the M bit depth image signal are assigned to the low bit depth image A for the luminance signal (for the low bit depth signal encoding device 30), and the middle bit of the M bit depth image signal is the luminance signal. And low bit depth images A and B for color difference signals (for low bit depth signal encoding devices 30 and 40) are assigned alternately, and the lower bits of the M bit depth image signal are low bit depth images B for color difference signals ( Low bit depth signal encoding device 40).

  For example, when a 16-bit depth image signal (M1, M2, M3, M4,..., M16) in a high bit depth image is decomposed and assigned, M1 to M4 as upper bits, M5 to M12 as middle bits, and lower bits Classify into three types of bit areas M13 to M16, and M1 to M4 (the upper bits of the M bit depth image signal) and M5, M7, M9, and M11 (the odd bits of the middle bits of the M bit depth image signal) are low M6, M8, M10, M12 (even bits of the middle bit of the M-bit depth image signal) and M13 to M16 (M-bit depth image signal) assigned to the image signal (a1, a2,..., A8) of the bit depth image A Are assigned to the image signals (b1, b2,..., B8) of the low bit depth image B.

  The operations of the low bit depth signal generation units 22-1 and 222-2 and the auxiliary information generation unit 23 are the same as those in the first embodiment.

  Further, the low bit depth signal encoding devices 30 and 40 differ only in the compression rate, and their operations are the same as those in the first embodiment.

  As described above, also in the bit depth decomposition allocation apparatus 20 of Example 2 in the first embodiment, the image signal of the low bit depth image from the M bit depth image signal (A bit depth image signal and B When assigning to the bit depth image signal), the upper bits in the M bit depth image signal are assigned to the upper bits with less coding deterioration of the low bit depth image A and the low bit depth image B (that is, M The bit depth image signal is assigned by maintaining the depth order from the MSB side and arranging from the upper bits).

(Image decoding system provided with bit depth combination restoration device)
Also in the second embodiment, as illustrated in FIG. 2, the bit depth combination restoration apparatus 80 is provided in an image decoding system 50 including two low bit depth signal decoding apparatuses 60 and 70. However, the second embodiment is different from the first embodiment in the operation of the bit depth combination restoration unit 82.

  Referring to FIG. 2, the low bit depth signal decoding devices 60 and 70 perform the image encoding process in the low bit depth signal encoding devices 30 and 40 on the input encoded stream, as in the first embodiment. The image decoding process corresponding to each of the image signals is performed, the bit streams of the image signals of the low bit depth images A and B (A bit depth image signal and B bit depth image signal) are decoded, It outputs to the depth signal receivers 81-1 and 81-2.

  The low bit depth signal reception unit 81-1 receives the image signals (a1, a2,..., A8) of the low bit depth image A and outputs them to the bit depth combination restoration unit 82. Similarly, the low bit depth signal reception unit 81-2 receives the image signals (b1, b2,..., B8) of the low bit depth image B and outputs them to the bit depth combination restoration unit 82.

  The bit depth combining / restoring unit 82 decomposes each bit of the image signals of the low bit depth images A and B received by the low bit depth signal receiving units 81-1 and 81-2, and converts these bits into auxiliary information. Are combined according to the predetermined bit allocation shown in FIG. 5 (ie, each bit decomposed according to the image encoding process of the plurality of types of compression rates is divided into M bit depths for each of the three types of bit regions. The depth order from the MSB side in the image signal is maintained and combined), and the M-bit depth image signal is reconstructed to restore a 16-bit high bit depth image.

  More specifically, in the second embodiment, the bit depth combining / restoring unit 82 uses the M bit depth image signal to reconstruct a1 to a4 in the image signals (a1, a2,..., A8) of the low bit depth image A. Allocation is made to the upper bits M1 to M4, and a5 to a8 are assigned to the odd bits M5, M7, M9, and M11 of the middle bits of the M-bit depth image signal to be reconfigured. Further, the bit depth combining / restoring unit 82 converts the b1 to b4 in the image signals (b1, b2,..., B8) of the low bit depth image B into even bits M6 of the middle bits of the M bit depth image signal to be reconstructed. M8, M10, and M12 are assigned, and b5 to b8 are assigned to lower bits M13 to M16 of the M-bit depth image signal to be reconfigured. In this manner, the bit depth combination restoration unit 82 alternately couples the bits of the image signals of the low bit depth images A and B to restore a 16-bit high bit depth image.

  Also in the present embodiment, between the transmission and reception (between the encoding side and the decoding side) with respect to the decomposition allocation information indicating the decomposition allocation method used when decomposing and allocating the image signal from the M bit depth image signal to the image signal of the low bit depth image. When predetermined, it is not always necessary to provide the auxiliary information generating unit 23 and the auxiliary information receiving unit 83, and it is not necessary to transmit or receive auxiliary information.

  As described above, according to Example 2 of the first embodiment of the present invention, the same effect as Example 1 is obtained, and a high bit depth image is decomposed according to image encoding processing with different compression rates. When image coding processing with different compression rates is performed on a high bit depth image signal using a plurality of low bit depth image signals, the image quality of the restored high bit depth image can be improved.

Example 3
Next, Example 3 will be described in the bit depth decomposition allocation apparatus 20 and the bit depth combination restoration apparatus 80 of the first embodiment. In the second embodiment described above, the example in which the upper bits of the image signal with a high bit depth (M bit depth image signal) are determined as M1 to M4 has been described, but it is preferable to classify the upper bits, middle bits, and lower bits. The number of bits varies depending on the target bit rate, the setting of an encoding device for low bit depth, and its encoding software. Thus, as a third embodiment, an example will be described in which disassembly / assignment information indicating how many bits from the MSB are designated as upper bits is transmitted as auxiliary information.

(Image coding system provided with bit depth decomposition allocation apparatus)
The third embodiment is configured in the same manner as the second embodiment. As shown in FIG. 1, the bit depth decomposition allocation apparatus 20 is provided in an image coding system 10 including two low bit depth signal coding apparatuses 30 and 40. It is done. However, the third embodiment is different from the second embodiment in that the auxiliary information generation unit 23 is provided in the bit depth decomposition allocation apparatus 20 as an indispensable component.

  The auxiliary information generation unit 23 acquires, from the bit depth decomposition allocation unit 21, decomposition allocation information indicating a decomposition allocation method used when decomposing and allocating an M-bit depth image signal to an image signal of a low bit depth image, as auxiliary information. It is configured to output externally. However, when the bit depth of the image signal of the high bit depth image (M bit depth image signal) is classified into three types of bit areas, the auxiliary information generation unit 23 according to the third embodiment has one or more bit areas. The decomposition allocation information regarding the number of bits is transmitted as auxiliary information. For example, decomposition allocation information indicating how many bits from the MSB of the M-bit depth image signal are designated as upper bits is transmitted as auxiliary information. The data to be transmitted at this time may take any form as long as the number of bits handled as the upper bits is known.

For example,
Transmitting “the number of upper bits (for example, 4)” or “the code word representing the number of upper bits (for example, 4)”;
Transmitting "a code word representing the number of upper bits 1 (for example, 3)" or "the number of upper bits 1 (for example, 3)",
Transmitting “the number of upper bits / 2 (for example, 2)” or “the codeword representing the number of upper bits / 2 (for example, 2)”;
Transmit “the number of upper bits / 2-1 (for example, 1)” or “the number of upper bits / 2-1 (for example, 1)”.
Etc. are assumed.

  The number of middle bits and lower bits may be transmitted together with the upper bits, but the remaining bits based on the number of bits in one type of bit area, for example, the number of lower bits = upper bits If the number of bits in two bit areas is determined between transmission and reception (between encoding and decoding side) so as to be obtained by calculation, only the number of bits in one type of bit area can be transmitted.

  In addition, when the number of bits in the remaining two types of bit areas is calculated from the number of bits in one type of bit area, the number of bits in the middle or lower bits is transmitted instead of the number of bits in the upper bits. Alternatively, a table number of a predetermined number of bits may be transmitted.

(Image decoding system provided with bit depth combination restoration device)
Since the third embodiment can be configured in the same manner as the second embodiment, as illustrated in FIG. 2, the bit depth combination restoration apparatus 80 includes an image decoding system 50 including two low bit depth signal decoding apparatuses 60 and 70. Is provided. However, the third embodiment is different from the second embodiment in that the bit depth combination restoration apparatus 80 is configured to always refer to auxiliary information transmitted.

  The low bit depth signal receivers 81-1 and 81-2 respectively receive the image signals (a1, a2,..., A8) of the low bit depth image A and the image signals (b1, b2,..., B8) of the low bit depth image B. ) And output to the bit depth combination restoration unit 82.

  The auxiliary information receiving unit 83 acquires auxiliary information including bit decomposition / allocation information related to the image signal of the low bit depth image from the auxiliary information stream transmitted from the image encoding system 10, and the bit depth combination restoring unit 82. Output to. As this auxiliary information, when the bit depth of the M bit depth image signal in the high bit depth image is classified into three types of bit areas, decomposition allocation information relating to the number of bits of one or more bit areas is transmitted. For example, decomposition allocation information indicating how many bits from the MSB of the M-bit depth image signal are designated as upper bits is transmitted as auxiliary information. Based on this decomposition allocation information, according to a predetermined method between transmission and reception (for example, the number of bits in one bit area such as the number of bits in the lower bits = the upper bits) The number is also calculated) to determine the bit positions of the decomposed upper bit, middle bit, and lower bit.

  Thereby, the bit depth combination restoration apparatus 80 can appropriately discriminate the three types of bit regions designated by the transmission side, and can reconstruct the M bit depth image signal.

  According to the third embodiment, on the transmission side (bit depth decomposition allocation apparatus 20 side), for example, the number of bits classified and designated as upper bits is variably controlled according to the expected encoding degradation of the luminance signal and the color difference signal. Therefore, image quality deterioration can be further suppressed for the high bit depth image restored on the receiving side (bit depth combination restoration device 80).

  For example, as the control of the number of bits designated as the upper bits, when there is no difference in coding deterioration of the luminance signal and the color difference signal (that is, when the quantization characteristic value of the luminance signal and the color difference signal is the same), the highest 1 A bit that is designated as an upper bit when only the bit is an upper bit and the coding deterioration of the color difference signal is larger than that of the luminance signal (that is, the quantization characteristic value of the color difference signal is larger than the quantization characteristic value of the luminance signal). It is preferable to control to increase the number.

[Other Examples]
In each of the above-described embodiments, an example in which a high bit depth image signal is decomposed into two low bit depth image signals has been described, but the present invention is also applied to a case where the high bit depth image signal is decomposed into three or more low bit depth image signals. It is possible.

  Further, in each of the above-described embodiments, the example in which the bit position of the high bit depth image signal to be decomposed is decomposed by the odd bits and the even bits, but on the transmission side, the bit depth decomposition allocation unit 21 performs n types ( When decomposing into n> 1), the bit position is decomposed with a remainder modulo n, and each of the n types of decomposed bits is assigned to image signals of a plurality of low bit depth images corresponding to the n types. What is necessary is just to comprise. At this time, on the reception side, the bit depth combination restoration unit 82 decomposes the image signals of the plurality of low bit depth images so as to correspond to the decomposed n types of bits, and for the image signal of the high bit depth image Configure to join.

  In each of the above-described embodiments, it may be assumed that the number of bits of the high bit depth image signal is smaller than the total number of bits of the decomposed low bit depth image signal. In this case, the bit of the high bit depth image signal can be assigned only to the upper bits of the low bit depth image signal, and a dummy bit (for example, 0) can be used on the lower bit side.

For example, a mode assumed at the present time to decompose a high bit depth image signal into a plurality of low bit depth image signals is as follows. Here, a mode of decomposition is represented by “bit depth of high bit depth image signal” → “bit depth of a plurality of low bit depth image signals”.
・ 16 bits → 8 bits or 10 bits or a combination thereof ・ 12 bits → 8 bits or 10 bits or a combination thereof ・ 10 bits → 8 bits

As an example, a 16-bit depth image signal (M1, M2, M3, M4,..., M16) in a high-bit depth image is reduced to a low bit-depth image A (image signals a1, a2,. When decomposing and assigning to the image signal of the bit depth image B (image signals b1, b2,..., B10),
(M1, M3, M5,..., M15, 0, 0) → (a1, a2,..., A10)
(M2, M4, M6,..., M16, 0, 0) → (b1, b2,..., B10)
And it is sufficient.

In addition, when dividing and allocating to three or more, a dummy bit (for example, 0) can be used on the lower bit side. In this case as well, allocation is performed by maintaining the depth order from the MSB side in the M-bit depth image signal and arranging from the upper bits. For example, as an application example of the first embodiment, a 16-bit depth image signal (M1, M2, M3, M4,..., M16) in a high bit-depth image is converted into an 8-bit low-bit depth image A (image signal a1). , A2,..., A8), a low bit depth image B (image signals b1, b2,..., B8) and a low bit depth image C (image signals c1, c2,..., C8).
(M1, M4, M7, M10, M13, M16, 0, 0) → (a1, a2,..., A8)
(M2, M5, M8, M11, M14, 0, 0, 0) → (b1, b2,..., B8)
(M3, M6, M9, M12, M15, 0, 0, 0) → (c1, c2,..., C8)
And it is sufficient.

  In the second and third embodiments described above, when each of the two low bit depth images A and B is scheduled to be encoded with two different compression rates, the bit depth decomposition is performed. The allocating unit 21 assigns each bit of the bit area classified into the three types of bit areas of the high-order bit, the middle-order bit, and the low-order bit for the bit depth of the image signal of the high bit depth image to the two different compression rates. The example which decomposes | disassembles according to the image coding process was demonstrated. Here, the present invention can also be applied to a case where three or more low bit depth images obtained by decomposing and assigning a high bit depth image signal are scheduled to be encoded with a plurality of different compression rates. it can.

  That is, when three or more low-bit-depth images are scheduled to be encoded with a plurality of different compression rates, the bit order from the higher order is taken into consideration when the images are included with the same compression rate. Are assigned to an image signal of a low bit depth image for which image coding processing with the lowest compression rate is scheduled among the plurality of types of compression rates. The bit belonging to the bit region of the bit is assigned for an image signal of a low bit depth image for which image coding processing with the highest compression rate among the plurality of types of compression rates is scheduled, and belongs to the bit region of the middle bit Are sequentially used for image signals of the plurality of low bit depth images in bit units in order from the image compression process with the lowest compression rate among the plurality of types of compression rates. It may be configured to direct Ri. Also in this case, each of the plurality of low bit depth signal generation units maintains the depth order from the MSB side in the image signal of the high bit depth image for each assigned bit, and sequentially from the upper bit position. By arranging, an image signal of the low bit depth image is generated.

  On the receiving side, the bit depth combination restoration unit 82 reconstructs a high bit depth image signal according to the image decoding processing of the plurality of types of compression rates for the image signals of the plurality of low bit depth images. Extracting bits belonging to the upper bit area from the image signal that has been subjected to the image decoding process with the lowest compression rate among a plurality of types of compression rates, and image having the highest compression rate among the plurality of types of compression rates Bits belonging to the lower bit region are extracted from the image signal subjected to the decoding process, and sequentially from the image signal subjected to the image decoding process with the lowest compression rate among the plurality of types of compression rates, in units of bits. By extracting the bits belonging to the bit area of the middle bit, the image signal of the high bit depth image is restored.

For example, in view of the aspect of the third embodiment, dummy bits (for example, 0) are also used when each bit classified into three types of bit areas of upper bits, middle bits, and lower bits is divided into three or more and assigned. Can be used on the lower bit side. In this case as well, allocation is performed by maintaining the depth order from the MSB side in the M-bit depth image signal and arranging from the upper bits. As an example, a 16-bit depth image signal (M1, M2, M3, M4,..., M16) in a high bit depth image is converted into upper bits (M1 to M4), middle bits (M5 to M12), and lower bits (M13 to M16). ), An 8-bit depth low bit depth image A (image signals a1, a2,..., A8) that is scheduled to be encoded at a low compression rate for luminance signals, and high for the first color difference signal. Low bit depth image B (image signals b1, b2,..., B8) scheduled for encoding at a compression rate and a high compression rate (the same compression rate as the first color difference signal) for the second color difference signal Are divided into image signals of low bit depth images C (image signals c1, c2,..., C8) that are scheduled to be encoded, the middle bits (M5 to M12) and the lower bits (M13 to M16) Is the same pressure Considering the case, including those of the rate with priority bit order from the upper,
(M1, M2, M3, M4, M5, M8, M11, 0) → (a1, a2,..., A8)
(M6, M9, M12, M14, M16, 0, 0, 0) → (b1, b2,..., B8)
(M7, M10, M13, M15, 0, 0, 0, 0) → (c1, c2,..., C8)
And it is sufficient.
Further, in consideration of the case where there are a plurality of encoding processes prepared for luminance signals, the upper bit order (M1 to M4), middle order bits (M5 to M12) and lower order bits (M13 to M16) It may be configured so that the priority is assigned.

  Next, an image encoding device and an image decoding device will be described as a second embodiment according to the present invention.

Second Embodiment
(Image coding device)
FIG. 4 is a block diagram of an image encoding device 10a according to an embodiment of the present invention. In the first embodiment described above, the bit depth decomposition allocation apparatus 20 and the two low bit depth signal encoding apparatuses 30 and 40 have been described as separate apparatuses. However, in the second embodiment, the bit depth decomposition allocation apparatus 20 This is an example configured as a single image encoding device 10a having the functions of and two low bit depth signal encoding devices 30 and 40.

  The image encoding device 10a includes a bit depth decomposition / allocation unit 21a, low bit depth signal generation units 22a-1, 22a-2, an auxiliary information generation unit 23a, and low bit depth signal encoding units 30a, 40a.

  The operations of the bit depth decomposition allocation unit 21a, the low bit depth signal generation units 22a-1, 22a-2, the auxiliary information generation unit 23a, and the low bit depth signal encoding units 30a, 40a are the same as those of the first embodiment described above. This corresponds to the operations of the depth decomposition allocation unit 21, the low bit depth signal generation units 22-1 and 222-2, the auxiliary information generation unit 23, and the low bit depth signal encoding devices 30 and 40, respectively.

(Image decoding device)
FIG. 5 is a block diagram of an image decoding device 50a according to an embodiment of the present invention. In the first embodiment described above, the bit depth combination restoration apparatus 80 and the two low bit depth signal decoding apparatuses 60 and 70 have been described as separate apparatuses. However, in the second embodiment, the bit depth combination restoration apparatus 80 This is an example configured as one image decoding device 50a having a function and the functions of two low bit depth signal decoding devices 60 and 70.

  The image decoding device 50a includes low bit depth signal decoding units 60a and 70a, a bit depth combination restoration unit 82a, and an auxiliary information reception unit 83a.

  The operations of the low bit depth signal decoding units 60a and 70a, the bit depth combination restoration unit 82a, and the auxiliary information reception unit 83a are the same as the low bit depth signal decoding units 60 and 70 and the bit depth combination restoration unit 82 in the first embodiment described above. And corresponding to each operation of the auxiliary information receiving unit 83a.

  Therefore, also in 2nd Embodiment, it can comprise like each Example in 1st Embodiment mentioned above, and the effect similar to 1st Embodiment can be acquired.

  Further, the function of each component of the bit depth decomposition allocation apparatus 20 and the bit depth combination restoration apparatus 80, and the image encoding apparatus 10a and the image decoding apparatus 50a of each embodiment can be realized by a computer. A program for causing a computer to realize each component according to the present invention is stored in a memory (not shown) provided inside or outside the computer. By appropriately reading a program describing processing contents for realizing the function of each component under the control of a central processing unit (CPU) provided in the computer and executing the program, each embodiment The function of each component of the apparatus can be realized by a computer. Here, the function of each component may be realized by a part of hardware.

  The present invention has been described with reference to specific embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the technical idea thereof. For example, an example in which auxiliary information is transmitted as an auxiliary information stream has been described. However, it may be configured to be multiplexed with any one or all of a bit stream and an encoded stream regarding transmission of an image signal. . Further, the bit stream, the encoded stream, and the auxiliary information stream can be realized in a form adapted to an existing transmission system.

  According to the present invention, when one high bit depth image is converted into a plurality of low bit depth images and transmitted, it is possible to suppress image quality deterioration of the restored high bit depth image with a simple configuration. Therefore, it is useful for the purpose of encoding and transmitting an image signal of a high bit depth image by an image encoding process for a plurality of low bit depth images.

DESCRIPTION OF SYMBOLS 10 Image coding system 10a Image coding apparatus 20 Bit depth decomposition | disassembly allocation apparatus 21, 21a Bit depth decomposition | disassembly allocation part 22-1, 22-2, 22a-1, 22a-2 Low bit depth signal generation part 23, 23a Auxiliary information Generation unit 30, 40 Low bit depth signal encoding device 30a, 40a Low bit depth signal encoding unit 50 Image decoding system 50a Image decoding device 60, 70 Low bit depth signal decoding device 60a, 70a Low bit depth signal decoding unit 80 bits Depth Combined Restoration Device 81-1 and 81-2 Low Bit Depth Signal Receiving Unit 82, 82a Bit Depth Combined Restoring Unit

Claims (6)

A bit depth decomposition assigning apparatus that decomposes and assigns an image signal of a high bit depth image having a predetermined bit depth to a plurality of low bit depth image signals having a bit depth lower than the predetermined bit depth,
Each bit in the image signal of the high bit depth image is decomposed into a plurality of types in a predetermined bit unit, and the decomposed plurality of types of bits are used for the image signals of a plurality of low bit depth images corresponding to the plurality of types. A bit depth decomposition assigning means to assign;
Maintaining the depth order from the most significant bit side in the image signal of the high bit depth image for each assigned bit, and arranging in order from the upper bit position, the image signals of the plurality of low bit depth images Low bit depth signal generating means for generating each of
Equipped with a,
When image encoding processing with a plurality of types of compression rates is scheduled for the plurality of low bit depth images,
The bit depth decomposition assigning means assigns each bit of the bit areas classified into three types of bit areas of upper bits, middle bits and lower bits with respect to the bit depth of the image signal of the high bit depth image. There is means for decomposing into a plurality of types according to the number and assigning each of the decomposed bits to image signals of a plurality of low bit depth images corresponding to the plurality of types with priority from the highest bit order. And
Bits belonging to the bit region of the upper bits are allocated for image signals of low bit depth images for which image encoding processing with the lowest compression rate among the plurality of types of compression rates is scheduled,
Bits belonging to the bit region of the lower bits are allocated for an image signal of a low bit depth image for which image encoding processing with the highest compression rate among the plurality of types of compression rates is scheduled,
The bits belonging to the bit area of the middle bit are the images of the plurality of low bit depth images in bit units in order from the image encoding process with the lowest compression rate among the plurality of types of compression rates. bit depth decomposition allocation apparatus characterized that you have been configured to sequentially assign the signal.   The bit depth decomposition assigning unit decomposes the bit position with a remainder modulo n when the image signal of the high bit depth image is decomposed into n types, and each of the n types of decomposed bits is converted into the n bits. 2. The bit depth decomposition allocation apparatus according to claim 1, further comprising means for allocating image signals of a plurality of low bit depth images corresponding to types. A bit depth combination restoration apparatus for combining a plurality of low bit depth image signals having a predetermined bit depth for a high bit depth image signal having a bit depth higher than the predetermined bit depth,
Low bit depth signal receiving means for receiving image signals of the plurality of low bit depth images;
By decomposing each bit of the received image signals of the plurality of low bit depth images and maintaining and combining the depth order from the most significant bit side in the image signal of the high bit depth image, the high bit depth image Bit depth combined restoration means for reconstructing an image signal and restoring the high bit depth image;
Equipped with a,
The plurality of low bit depth images are subjected to image decoding processing at a plurality of types of compression rates, and each of the plurality of low bit depth images has a bit depth of an image signal of a high bit depth image. When each bit of the bit area classified into the three types of bit areas of the upper bit, the middle bit, and the lower bit is decomposed into a plurality of types according to the number of the plurality of types,
In order to reconstruct the image signal of the high bit depth image according to the image decoding processing of the plurality of types of compression ratios for the image signals of the plurality of low bit depth images,
Extracting bits belonging to the bit region of the upper bits from an image signal subjected to image decoding processing with the lowest compression rate among the plurality of types of compression rates,
Extracting bits belonging to the bit region of the lower bits from an image signal subjected to image decoding processing with the highest compression rate among the plurality of types of compression rates,
By extracting the bits belonging to the bit region of the middle bit in bit units sequentially from the image signal subjected to the image decoding process with the lowest compression rate among the plurality of types of compression rates, the high bit depth image bit depth coupled restoration apparatus characterized that you have been configured to restore an image signal. When each image signal of the plurality of low bit depth images is a high bit depth image obtained by dividing the bit position into n types with a modulus modulo n.
The bit depth combining / restoring means includes means for decomposing the image signals of the plurality of low bit depth images so as to correspond to the n types of decomposed bits and combining the image signals of the high bit depth images. The bit depth combination restoration apparatus according to claim 3 , wherein: The program for functioning a computer as a bit depth decomposition | disassembly allocation apparatus of Claim 1 or 2 . The program for functioning a computer as a bit depth coupling | bonding decompression | restoration apparatus of Claim 3 or 4 .