JP2017098898A - Image encoding device, control method of the same, program, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve compatibility among quantification of tile-division image data, reduction of a tile boundary distortion, and high controllability of a code amount when encoding.SOLUTION: An image encoding device includes: a setting part that sets a bit rate of image data of an encoding target; a frequency conversion part 105 that generates a plurality of sub-bands by performing the frequency conversion to a target tile; a quantification control part 106 that determines which one of a first quantification method of quantifying the same sub band in the tiles with the same quantification parameter and a second quantification method of quantifying the same sub band of tiles with a variable quantification parameter, in a plurality of sub bands of the target tile obtained by the frequency conversion part, in accordance with the bit rate; a quantification part 107 that quantifies the plurality of sub bands obtained by performing the frequency conversion of the target tile on the basis of the quantification parameter determined by the quantification control part; and an encoding part 108 that encodes conversion coefficient data after the quantification generated in the quantification part.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image encoding technique.

  Currently, an encoding method for compressing and encoding image data with high efficiency has been proposed. A representative coding method among them is JPEG (Joint Photographic Coding Experts Group). In this JPEG system, DCT (Discrete Cosine Transform) is used for frequency conversion. However, since DCT in JPEG is processed in units of fixed blocks of 8 × 8 pixels, it is known that block distortion tends to occur at a low bit rate.

  On the other hand, in a coding method called JPEG2000, which is standardized as a successor to JPEG, DWT (Discrete Wavelet Transform) is used in frequency conversion. DWT divides image data into a plurality of frequency band components by a filter called a filter bank that combines a high-pass filter and a low-pass filter. The DCT performs frequency conversion in units of fixed blocks, whereas the DWT has virtually no limitation on the size, and can perform frequency conversion in units of screens, for example, and block distortion due to quantization does not occur. Then, after the image data is divided into a plurality of frequency bands (hereinafter referred to as subbands), the data is quantized for each subband and then encoded. In quantization, in consideration of human visual characteristics, a method for reducing the code amount without losing the subjective image quality by assigning a large amount of code to the low frequency subband is generally used for still images and moving images. It is a natural idea.

  Here, DWT and quantization in JPEG2000 can be applied to an arbitrary rectangular block (hereinafter referred to as a tile). By dividing and encoding the tiles, it is possible to increase the parallelism of processing and reduce the RAM (line buffer) capacity held inside. For example, consider a case where image data is divided into left and right parts. In this case, the number of pixels in one horizontal line of each of the left and right tiles is half that of the original, and the capacity of the line buffer held when encoding can be reduced to half that of the original. However, since each tile performs independent quantization, a difference occurs in the quantization error of each tile. In particular, distortion occurs at the tile boundary portion at the time of high compression.

  As a prior art, in an image coding system that performs DWT-based tile division, a method of reducing image quality distortion caused by discarding lower bits for code amount control is disclosed in Patent Document 1 below.

JP 2013-175870 A

  If the tiles can use a common quantization parameter, tile boundary distortion will not occur. However, if the same quantization parameter is applied to each tile, the controllability of the code amount deteriorates, and there may occur a case where the tile does not converge to the target bit rate.

  Here, Patent Document 1 described above describes a method for reducing boundary distortion that occurs due to different lower bit discard positions between tiles in code amount control using bit plane quantization by embedded coding (EBCOT). Has been. Specifically, error data to be discarded is encoded and stored separately from tile data, and at the time of decoding, the tile data is decoded using the stored error data, thereby reducing boundary distortion. ing. However, this method requires an error data encoding processing unit, an error data decoding processing unit, and a tile data and error data combining processing unit at the time of decoding. .

  The present invention has been made in view of the above problems, and when quantizing and encoding tile-divided image data, it is possible to reduce tile boundary distortion and code amount with an easy control method while suppressing an increase in mounting scale. It is intended to provide a technology for achieving both high controllability and high controllability.

In order to solve this problem, for example, an image encoding device of the present invention has the following configuration. That is,
An image encoding device that divides image data to be encoded into a plurality of tiles and encodes the tiles as a unit,
Setting means for setting a bit rate of the image data to be encoded;
A frequency conversion means for generating a plurality of subbands by frequency converting the tile of interest;
For each of the plurality of subbands of the tile of interest obtained by the frequency conversion means,
(1) A first quantization method in which the same subband between tiles is fixed and is quantized with the same quantization parameter.
(2) Quantization control means for determining which of the second quantization methods for quantizing the same subband between tiles with a variable quantization parameter according to the bit rate set by the setting means When,
Quantization means for quantizing a plurality of subbands obtained by frequency-transforming the tile of interest based on the quantization parameter determined by the quantization control means;
Encoding means for encoding the transform coefficient data after quantization generated by the quantization means.

  According to the present invention, when quantizing and encoding tile-divided image data, the reduction in tile boundary distortion and the controllability with a high code amount are achieved with easy control while suppressing an increase in mounting scale. It becomes possible.

1 is a block configuration diagram of an imaging apparatus in a first embodiment. FIG. The figure for demonstrating the two quantization methods in embodiment. The flowchart which shows the determination process of the quantization method in 1st Embodiment. The figure which shows the example of the division | segmentation of the subband in 1st Embodiment, and the relationship between the bit rate and the quantization method applied to each subband. The figure which shows transition of the quantization parameter and the relationship between the bit rate in 2nd Embodiment, and the tolerance | permissible_range of a quantization parameter. The figure which shows transition of the code amount at the time of encoding of a subband. The flowchart which shows an encoding process procedure.

  Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. In the following description, an imaging apparatus represented by a digital camera will be described as an example of an apparatus to which the image encoding apparatus is applied. Therefore, the generation source of the image data to be encoded is an imaging unit included in the imaging device. However, the present invention is not applied only to the imaging apparatus. The generation source of the image data to be encoded is not limited to the imaging unit, and may be a storage medium storing the image data to be encoded, and the type thereof is not limited. It should be understood that this is for ease of understanding.

[First Embodiment]
FIG. 1 is a block configuration diagram of an imaging apparatus 100 to which the first embodiment is applied. The imaging apparatus 100 is configured to divide the captured image information into a plurality of tiles and then record it. Further, the tile size is variable. For example, the entire image data may be set as one tile. The control unit 110 controls each component of reference numerals 101 to 109 and controls the entire apparatus. Typically, the control unit 110 includes a CPU, a ROM storing a program and data indicating a processing procedure of the CPU, and a RAM used as a work area of the CPU. Note that some or all of the reference numerals 101 to 109 may be realized by the processing of the control unit 110.

  The operation unit 101 receives a user command and generates a control signal according to the command. For example, an imaging instruction, tile division number (tile size), recording bit rate setting, and the like are also input from the operation unit 101. The input tile division number is supplied to the tile division unit 103. The recording bit rate is supplied to the quantization control unit 106.

  Based on the control signal from the operation unit 101, the imaging unit 102 converts the light intensity signal transmitted through the red, green, and blue (RGB) color filters arranged for each pixel of the imaging sensor into image information and tiles it. The data is output to the dividing unit 103. In the first embodiment, the image data output from the imaging unit 102 is referred to as RAW image data meaning a raw (undeveloped) image. An example of a color filter arranged in the imaging unit 102 is a Bayer array. In the Bayer array, red (R), green (G), and blue (B) are arranged in a mosaic pattern for each pixel, and red (R) 1 pixel and blue (B) 1 for every 2 × 2 4 pixels. It has a structure in which pixels and two green (G1, G2) pixels are regularly arranged in one set.

  The tile dividing unit 103 temporarily stores the RAW image input from the imaging unit 102 in the internal memory, and divides the RAW image data into one or more tiles based on the number of tile divisions supplied from the operation unit 101. Note that the number of tile divisions from the operation unit 101 is in an M × N format, and the divided tiles are basically the same size. However, although the number of horizontal pixels of the original RAW image data is not necessarily an integer multiple of M, the remaining pixels are assigned to appropriate tiles according to preset conditions. The same applies to the number of divisions N in the vertical direction. The tile dividing unit 103 sequentially supplies the tiles obtained by the division to the plane forming unit 104. Instead of using the number of tile divisions input from the operation unit 101, the number of divisions so that the internal memory resource does not fail is automatically calculated according to the image size input from the imaging unit 102. Also good.

  The plane forming unit 104 decomposes the RAW image data of the tile supplied from the tile dividing unit 103 for each color component to form a color plane. Since the RAW image is a Bayer array, the color plane forming unit 104 extracts each color component of red (R), green (G1), green (G2), and blue (B) to form a color plane. Then, the plane forming unit 104 supplies the formed R plane, G1 plane, G2 plane, and B plane to the frequency conversion unit 105 in this order, for example.

  The frequency transform unit 105 performs discrete wavelet transform on the R plane received from the plane forming unit 104 and supplies transform coefficient data obtained as a result to the quantization unit 107. In this discrete wavelet transform, for example, an integer type 5 × 3 filter specified by JPEG2000 may be used, or a real type 9 × 7 filter may be used. Then, the frequency transform unit 105 also performs discrete wavelet transform on other planes (G1, G2, B planes) received from the “plane forming unit” 104, and the resulting transform coefficient data is quantized. 107 is supplied.

  The quantization control unit 106 determines a quantization parameter used by the quantization unit 107 based on the recording bit rate input from the operation unit 101, and supplies the quantization parameter to the quantization unit 107. Here, the quantization control unit 106 in the embodiment is configured to determine a quantization parameter from two quantization methods. Details of these two quantization methods will be described later.

  The quantization unit 107 performs quantization on the transform coefficient data received from the frequency transform unit 105 by using the quantization parameter received from the quantization control unit 106, and the quantized coefficients are entropy coding units. 108.

  The entropy encoding unit 108 compresses and encodes the quantized transform coefficient data received from the quantization unit 107 to generate encoded data. Then, the entropy encoding unit 108 supplies the generated encoded data to the storage medium 109. This compression coding is performed using entropy coding such as Golomb coding, for example.

  The storage medium 109 is a non-volatile memory (for example, a removable storage medium (SD card)) that stores encoded data received from the entropy encoding unit 108.

  Next, the first quantization method and the second quantization method in the quantization control unit 106 will be described with reference to FIGS. 2 (a) and 2 (b).

  FIG. 2A is a conceptual diagram illustrating the first quantization method, and FIG. 2B is a conceptual diagram illustrating the second quantization method. Hereinafter, these will be described in more detail. Note that the frequency conversion unit 105 in the embodiment will be described as performing discrete wavelet conversion twice for one tile. As a result, as shown in FIG. 4A, seven subbands SB0 to SB6 are generated from one tile. Note that the number of executions of this discrete wavelet transform is not limited to two, and there is no limit to the number of times as long as it is one or more. In some cases, the number of times may be set from the operation unit 101.

<First quantization method>
As shown in FIG. 2A, this is a method of uniformly quantizing one subband on the surface. In this quantization method, the relationship of one parameter per subband is fixed, and the quantization parameter (quantization step) of the same subband in each tile takes the same value between adjacent tiles. . Since the same quantization parameter is set for each subband of each tile, distortion of the tile boundary due to quantization does not occur. Note that the quantization parameter in this embodiment will be described by a method in which statistical values that converge near the bit rate input from the operation unit 101 are prepared in advance and applied to quantization.

  In the embodiment, seven subbands are generated from one tile from the frequency conversion unit 105. Therefore, a preset quantization parameter is set for each subband.

<Second quantization method>
As shown in FIG. 2B, a method of dividing one subband into lines and quantizing each line using different quantization parameters (Q # x) (x = 0, 1, 2,...). It is. Since this quantization method can change the quantization parameter for each line, the convergence to the set bit rate is high.

  A basic concept of the determination process of the first quantization method and the second quantization method will be described. When the bit rate is low (when the target compression rate is high), the quantization parameter is set to be large in order to suppress the code amount, and the quantization error becomes large. That is, the distortion at the tile boundary tends to be visually recognized. In accordance with this point, the first quantization method giving priority to the image quality of the tile boundary is determined.

  On the other hand, when the bit rate is high (when the target compression ratio is low), the quantization parameter is set to be small, so that the quantization error is small. Since the distortion at the tile boundary is difficult to visually recognize, the second quantization method is given priority on the convergence of the bit rate.

  Next, the quantization method determination process performed by the quantization control unit 106 for the RAW image encoding process will be described based on the flowchart shown in FIG.

  In S301, the quantization control unit 106 acquires the bit rate R from the operation unit 101 set by the user. If the bit rate is held in a memory unless the user changes it, the bit rate may be acquired from the memory. In S <b> 302, the quantization control unit 106 compares the acquired bit rate R with a preset threshold value T, and performs size determination. When the bit rate R is lower than the threshold value T0 (when R <0T), the process proceeds to S303. If the bit rate R is greater than or equal to the threshold T0, the process proceeds to S304. An example of setting the threshold value T0 will be described later.

  In S303, the quantization control unit 106 determines that the first quantization method is used as the quantization method for each subband of the tile of interest. When the process proceeds to S304, the quantization control unit 106 determines that a combination of the first quantization method and the second quantization method is used as the quantization method for each subband of the tile of interest. To do.

  Here, a setting example of the threshold value T and a quantization method in which the first quantization method and the second quantization method are combined will be described with reference to FIGS. 4 (a) and 4 (b).

FIG. 4A shows subbands generated when the discrete wavelet transform is executed twice for one tile. Each subband has the following meaning.
Subband SB0 (also expressed as 2LL):
This is a set of transform coefficient data obtained by subjecting a set (1LL) of low-frequency component transform coefficient data obtained by the first discrete wavelet transform to vertical low-pass filter processing and horizontal low-pass filter processing.
Subband SB1 (also expressed as 2HL):
A set of transform coefficient data obtained by subjecting a set of low-frequency component transform coefficient data (1LL) obtained by the first discrete wavelet transform (1LL) to vertical low-pass filter processing and horizontal high-pass filter processing. Subband SB2 (also expressed as 2LH):
This is a set of transform coefficient data obtained by subjecting a set (1LL) of low-frequency component transform coefficient data obtained by the first discrete wavelet transform to vertical high-pass filter processing and horizontal low-pass filter processing.
Subband SB3 (also expressed as 2HH):
This is a set of transform coefficient data obtained by subjecting a set (1LL) of low-frequency component transform coefficient data obtained by the first discrete wavelet transform to vertical high-pass filter processing and horizontal high-pass filter processing.
Subband SB4 (also expressed as 1HL):
It is a set of transform coefficient data obtained by performing vertical low-pass filter processing and horizontal high-pass filter processing on tiles.
Subband SB5 (also expressed as 1LH):
This is a set of transform coefficient data obtained by performing vertical high-pass filter processing and horizontal low-pass filter processing on tiles.
Subband SB6 (also expressed as 1HH):
This is a set of transform coefficient data obtained by performing vertical high-pass filter processing and horizontal high-pass filter processing on tiles.

Here, a specific example of the processing of S303 and S304 of the quantization control unit 304 will be described. FIG. 4B is a table showing a quantization method application pattern for each subband. The magnitude relation of the threshold is T0 <T1 <T2 <T3.
・ Pattern 1:
This is a pattern in which the first quantization method is applied to all subbands obtained from the target tile when the bit rate R is less than the threshold value T0. The process corresponding to this pattern 1 is also the process of S303. Patterns 2 to 5 described below are the processes in S304.
・ Pattern 2:
A pattern in which the second quantization method is applied to subband SB6 when bit rate R is equal to or greater than threshold value T0 and less than T1, and the first quantization method is applied to other subbands SB0 to SB5. It is.
・ Pattern 3:
When the bit rate R is greater than or equal to the threshold T1 and less than T2, the second quantization method is applied to the subbands SB4 to SB6, and the first quantization method is applied to the other subbands SB0 to SB3. Pattern.
Pattern 4:
When the bit rate R is greater than or equal to the threshold T2 and less than T3, the second quantization method is applied to the subbands SB3 to SB6, and the first quantization method is applied to the other subbands SB0 to SB2 Pattern.
Pattern 5:
When the bit rate R is greater than or equal to the threshold T3, the second quantization method is applied to the subbands SB1 to SB6, and the first quantization method is applied to the subband SB0.

  In the pattern 5 with the highest bit rate, since distortion at the tile boundary is difficult to visually recognize, there are the most subbands to which the second quantization method is applied, and the control is focused on the convergence of the bit rate.

The subband in the embodiment is expressed as SB # (# = 0, 1, 2,...). If such an expression is used even when the number of wavelet transforms is other than 2, it can be said that the frequency components are represented in order of decreasing frequency components. The processing of the quantization control unit 106 in the above patterns 1 to 5 can be summarized as (1) and (2) below.
(1) When a plurality of subbands obtained by frequency conversion are arranged in order from the lowest frequency, the first quantization method is applied to the M subbands from the lowest frequency to the higher frequency, and the M + 1th The second quantization method is applied to N subbands from 1 to the highest frequency.
(2) M is increased as the bit rate is lower (or N is increased as the bit rate is higher).
The lowest frequency subband SB0 (2LL in the embodiment) employs the first quantization method regardless of the bit rate.

  As described above, in coding of tiled images, image coding that can simultaneously suppress tile boundary distortion and control the amount of code by dynamically switching the quantization method according to the bit rate. A device can be provided.

  Note that the encoding process based on the first quantization method quantizes and encodes all the transform coefficients of the target subband based on the same quantization parameter, and therefore description thereof is unnecessary. Therefore, in the following, how the change of the quantization parameter of each line changes based on the second quantization method will be described together with the encoding process.

  When encoding is performed for lines 0, 1, 2,... Of the target subband, encoded data of each line is generated. That is, as long as encoding of each line continues, the total code amount increases monotonously.

  In addition to the bit rate R set by the user, the target code amount at the end of encoding all lines of the target subband is determined by the number of divided tiles, the number of wavelet transforms, and the type of target subband.

  FIG. 6 shows the transition of the code amount when the target subband is quantized and encoded in order from the line 0. The straight line that rises to the right in the figure indicates a straight line that represents an ideal transition connecting the target code amount and the origin (hereinafter referred to as an ideal straight line). The accumulated target code amount from the 0th line to the i-th line is represented as “TA (i)”. Actually, quantization and encoding do not always move on the ideal line, and the accumulated code amount increases while exceeding or falling below the ideal line. If it exceeds the ideal straight line, it means that it is necessary to suppress the generation of the code amount. Therefore, when the next line is quantized, the quantization parameter (quantization step) is increased by a predetermined value. On the other hand, when it falls below the ideal straight line, it means that the image quality may be increased by increasing the code amount. Therefore, when the next line is quantized, the quantization parameter (quantization step) is decreased by a predetermined value.

  Based on this point, encoding processing of a certain subband under the control of the control unit 110 in the embodiment will be described with reference to the flowchart of FIG. Note that the variable i used in the following description is used to specify the line number. The variable A is used for storing the accumulated code amount from the 0th line to the i-th line.

  First, in step S1, the control unit 110 controls the quantization control unit 106, and sets an initial value for the quantization parameter R prior to encoding the subband of interest. This initial value is determined by the bit rate R set by the user, the number of divided tiles, the number of wavelet transforms, and the type of the subband of interest.

    Next, in S2, control unit 110 clears variables i and A to 0, respectively. In step S3, the control unit 110 controls the quantization unit 107 to input the transform coefficient data of the i-th line from the target subband from the frequency conversion unit 105. In step S4, the control unit 110 performs quantization using the quantization parameter R. To do. In step S5, the control unit 110 controls the entropy encoding unit 108 to execute entropy encoding on the transform coefficient data for one line after quantization. In step S6, the control unit 110 updates the variable A by adding the encoded data amount of the i-th line to the variable A.

  Next, in S7, the control unit 110 determines whether or not the difference between the accumulated code amount A up to the i-th line and the accumulated target code amount TA (i) is equal to or less than a preset threshold value ε. If it is equal to or smaller than the threshold ε, the process proceeds to S11, and the control unit 110 determines whether or not all lines have been encoded. If not, the variable i is incremented by “1” in S12, and after S3 Repeat the process.

  When the difference between the accumulated code amount A up to the i-th line and the accumulated target code amount TA (i) exceeds the threshold ε, the control unit 110 advances the process to S8, and whether A <TA (i) is satisfied. That is, it is determined whether or not the accumulated code amount A is less than the accumulated target code amount TA (i). If A <TA (i), the code amount may be increased. Therefore, the control unit 110 increases the quantization parameter R by a predetermined positive value ΔR in S9. If A> TA (i), it is necessary to reduce the code amount. Therefore, the control unit 110 decreases the quantization parameter R by a predetermined value ΔR in S10.

  As a result, it is possible to realize encoding in which the quantization parameter of the target subband is changed for each line.

[Second Embodiment]
Since the apparatus configuration according to the second embodiment is the same as that of the first embodiment, description of the configuration is omitted. Also, the quantization method determination process performed by the quantization control unit 106 is the same as that in the first embodiment, and is therefore omitted. The difference is that in the second quantization method according to the second embodiment (FIG. 2B), the quantization parameter between the same subbands in adjacent tiles is set to be within a predetermined range. The quantization parameter is determined.

  In the second quantization method in the first embodiment, the processing content is always the same regardless of the bit rate, and the difference in the quantization parameter between tiles is not considered. Therefore, particularly at the low bit rate, there is a possibility that distortion of the tile boundary may occur when the amount of quantization parameter difference between adjacent tiles is large.

  Therefore, in the second quantization method of the second embodiment, the setting range is set for the quantization parameter for each line between adjacent tiles according to the bit rate, so that the image quality and the convergence property of the bit rate are more finely defined. The objective is to optimize the balance. The details will be described below.

  A quantization parameter setting method and a setting value range in the second quantization method according to the second embodiment will be described. In the second embodiment, it is assumed that the quantization parameter of the first line of the subband of the target tile of the divided tile has the same value between tiles.

FIG. 5A shows the setting range of the quantization parameter of a predetermined subband of the tile. FIG. 5B shows that the quantization parameter for each line changes so as to be within the set range shown in FIG. As for the setting method of the threshold values T0 and T1 in FIG. 5A, for example, a value obtained by equally dividing the recordable maximum bit rate may be set. However, it is assumed that the threshold value relationship is T0 <T1, and that there are three setting patterns based on the threshold value. Further, the size relationship between the allowable ranges A, B, and C of the quantization parameter difference is A <B <C.
・ Pattern 1:
When the bit rate R is less than the threshold value T0, the quantization parameter of each line used in the second quantization method is set to fall within a range of ± A with respect to the quantization parameter of the first line.
・ Pattern 2:
When the bit rate R is greater than or equal to the threshold value T0 and less than T1, the quantization parameter used in the second quantization method is set to fall within a range of ± B with respect to the quantization parameter of the first line.
・ Pattern 3:
When the bit rate R is equal to or higher than the threshold value T1, the quantization parameter used in the second quantization method is set so as to fall within a range of ± C with respect to the quantization parameter of the first line.

  As can be seen from FIGS. 5A and 5B, in the pattern 3 having the highest bit rate, the distortion at the tile boundary is difficult to visually recognize, so the quantization parameter is set in a wider range.

  On the other hand, in the pattern 1 with the lowest bit rate, when the bit rate decreases, the distortion at the tile boundary becomes easy to visually recognize. Therefore, the tile distortion is suppressed by setting the quantization parameter in a narrower range.

  By doing the above, in the encoding of tile-divided images, the balance between image quality and bit rate convergence can be adjusted more finely by dynamically switching the setting range of the quantization parameter according to the bit rate. It is possible to provide an image encoding apparatus capable of

  Note that the quantization parameter setting range of this embodiment has been described focusing on one subband, but the setting ranges A, B, and C have the same magnitude relationship in any subband. The setting ranges A, B, and C are used as limiters for quantization parameters. If applied to the flowchart of FIG. 7, when the set value A is applied, immediately after S9, S10, etc., if the range of the “quantization parameter initial value ± A” is exceeded, it is returned to the range. What is necessary is just to add a process.

(Other examples)
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

DESCRIPTION OF SYMBOLS 100 ... Imaging device 101 ... Operation part 102 ... Imaging part 103 ... Tile division part 104 ... Plane formation part 105 ... Frequency conversion part 106 ... Quantization control part 107 ... Quantization part 108 ... Entropy code 110, storage medium, 110, control unit

Claims (11)

An image encoding device that divides image data to be encoded into a plurality of tiles and encodes the tiles as a unit,
Setting means for setting a bit rate of the image data to be encoded;
A frequency conversion means for generating a plurality of subbands by frequency converting the tile of interest;
For each of the plurality of subbands of the tile of interest obtained by the frequency conversion means,
(1) A first quantization method in which the same subband between tiles is fixed and is quantized with the same quantization parameter.
(2) Quantization control means for determining which of the second quantization methods for quantizing the same subband between tiles with a variable quantization parameter according to the bit rate set by the setting means When,
Quantization means for quantizing a plurality of subbands obtained by frequency-transforming the tile of interest based on the quantization parameter determined by the quantization control means;
Encoding means for encoding transform coefficient data after quantization generated by the quantization means;
An image encoding device comprising: The quantization control means includes
If the bit rate set by the setting means is below a first threshold, determine that all subbands obtained from the tile of interest are encoded using the first quantization method;
If the bit rate set by the setting means does not fall below the first threshold, the first quantization method and the focus will be applied to preset subbands in all subbands obtained from the focus tile. The image coding apparatus according to claim 1, wherein the sub-bands other than the above in all sub-bands obtained from tiles are determined to be used in the second quantization method. The quantization control means includes
The first quantization method is applied to the M subbands from the lowest frequency to the higher frequency when the plurality of subbands obtained by the frequency conversion are arranged in order from the lowest frequency, and the M + 1th from the M + 1th most. Apply the second quantization method for N subbands up to high frequencies,
The image encoding apparatus according to claim 2, wherein the M is increased as the bit rate is lower. The quantization control means determines that the first quantization method is applied to a subband having the lowest frequency among all subbands obtained from the tile of interest regardless of the bit rate. The image encoding device according to claim 1, wherein The image coding apparatus according to any one of claims 1 to 4, wherein the second quantization method determines a quantization parameter in units of a line of a subband of interest. The quantization control means determines an allowable range of a quantization parameter when quantizing the target subband according to the second quantization method according to a bit rate. The image encoding device according to item 1.   The image encoding apparatus according to claim 6, wherein the allowable range is narrowed as the bit rate is lower.   The image coding apparatus according to claim 1, wherein the frequency conversion unit is a wavelet conversion unit. A control method of an image encoding device that divides image data to be encoded into a plurality of tiles and encodes the tiles as a unit,
A setting step for setting a bit rate of the image data to be encoded;
A frequency conversion step in which the frequency conversion means generates a plurality of subbands by frequency converting the tile of interest;
For each of the plurality of subbands of the tile of interest obtained by the frequency conversion step, the quantization control means
(1) A first quantization method in which the same subband between tiles is fixed and is quantized with the same quantization parameter.
(2) A quantization control step for determining which of the second quantization methods for quantizing the same subband between tiles with a variable quantization parameter according to the bit rate set in the setting step When,
A quantization step of quantizing a plurality of subbands obtained by frequency-converting the tile of interest based on a quantization parameter determined in the quantization control step;
An encoding step in which encoding means encodes the transformed transform coefficient data generated in the quantization step;
A method for controlling an image encoding device comprising:   A program for causing a computer to function as each unit included in the image encoding device according to any one of claims 1 to 8, when the computer reads and executes the computer.   A computer-readable storage medium storing the program according to claim 10.

Images