JP2009260747A - Image encoding device, and control method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To properly set an encoding speed and encoding efficiency by determining, in accordance with the size of an image, the number of pixel blocks to be encoded in parallel and the number of candidates for a boundary bit position. <P>SOLUTION: A user mode setting section 107 sets the size of an image to be encoded. An entropy encoding control section 108 determines a greater N1 as the set image size becomes larger, and determines a smaller N2 as the set image size becomes greater. Then, a coefficient data acquiring section 106 supplies to a boundary bit position determining section 109 coefficient values of N1 pieces of pixel blocks, thereby calculating N2 pieces of candidates for a boundary bit position for each pixel block. Furthermore, the coefficient data acquiring section 106 supplies the coefficient values of the N1 pieces of pixel blocks to N2 pieces of entropy encoding sections, respectively. A code amount comparing section 111 selects and outputs encoded data of which the encoded data amount is minimum, from among N2 pieces of encoded data generated from the coefficient value of one pixel block repeatedly N1 times. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for encoding image data.

  A JPEG encoding method is known as a still image encoding technique. In JPEG encoding, first, input image data is divided into pixel blocks composed of 8 pixels in the vertical direction and 8 pixels in the horizontal direction. Then, DCT conversion processing is performed for each pixel block, and 8 × 8 frequency component coefficients (hereinafter, DCT coefficients) are obtained. One coefficient in the upper left corner of the 8 × 8 DCT coefficients represents a DC (direct current) component, and the remaining 63 coefficients represent an AC (alternating current) component. Next, each DCT coefficient is quantized with reference to a preset quantization table. Each quantized DCT coefficient is entropy encoded and output as encoded data. Entropy coding methods include Huffman coding and arithmetic coding. The DC coefficient and the AC coefficient take different procedures and are coded.

  JPEG Baseline employs Huffman coding as a coefficient entropy coding method. The DC coefficient and the AC coefficient are encoded by different procedures. The DC component of the pixel block of interest is encoded after the difference value with the DC component (predicted value) of the immediately preceding pixel block is calculated. On the other hand, each AC component in the pixel block of interest is Huffman-encoded with a combination of a zero consecutive number (zero run) and a non-zero coefficient value.

  As can be seen in JPEG Baseline, in an encoding method that encodes a continuous number of zeros, a coefficient that becomes 0 must exist to some extent in order to perform encoding efficiently. In other words, when all 63 AC coefficients are non-zero, the coding efficiency is very poor.

  When the AC coefficient is quantized, in order to increase the number of occurrences of zero, it is necessary to increase the quantization step value. However, the larger the quantization step value, the lower the image quality of the image obtained by decoding.

  Therefore, in order to make the coding function efficiently, each coefficient is separated into the upper bit part and the lower bit part at the boundary bit position, and run length coding is applied to the upper bit part. A method can be considered. By interposing such separation processing, even if the DCT coefficient after quantization is non-zero, if the boundary bit position is set to an appropriate position, the higher-order bit portion is more likely to be “0”, which is efficient. Encoded data can be generated. Patent document 1 is mentioned as literature which discloses this technique.

In addition, a method of searching for and applying an optimum parameter for efficient compression at the time of encoding is also known (for example, Patent Document 2). This method searches for the optimum k parameter for Golomb-Rice encoding, performs Golomb-Rice encoding for each k parameter, and generates encoded data. Find good k-parameters.
JP 2000-013797 A JP 2000-115782 A

  However, in the technique disclosed in Patent Document 1, for example, the bit division position when the DCT coefficient is divided into the upper bit part and the lower bit part is fixed, and depending on the contents of the DCT coefficient, the coding efficiency is not necessarily good. You can't get results. In order to set the boundary bit position to an optimum position, it is necessary to perform encoding at all the bit positions and find the most code amount among them. In other words, it takes a lot of time for the encoding process to be realized by software, and when it is realized by hardware, the circuit scale becomes large, and problems such as an increase in the size of the apparatus, high costs, and an increase in power consumption remain. .

  The present invention has been made in view of the above-described conventional example. In an apparatus having a plurality of encoding processing units, the number of pixel blocks to be encoded in parallel and the number of boundary bit position candidates are determined according to the size of the image. By deciding, it is intended to provide a technology that can appropriately set the encoding speed and the encoding efficiency.

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 performs frequency conversion and quantization processing in units of pixel blocks from multivalued image data, and encodes each multivalued coefficient obtained by the quantization processing,
Each coefficient value of the pixel block is separated into an upper bit part higher than a set boundary bit position and a lower bit part below the set boundary bit position, and run-length encoding is performed on the upper bit part A plurality of encoding means for encoding the lower bit part in bit plane units;
For each coefficient of the pixel block of interest, the number of bits from the bit position that first becomes “1” from the most significant bit to the least significant bit to the least significant bit is detected as the effective bit number, and the detected effective bit number appears. Boundary bit position determining means for determining a plurality of candidates for the boundary bit position for each coefficient of the pixel block of interest based on the number of times;
Setting means for setting the size of the image to be encoded;
When the number of the plurality of encoding means is N0, the number of pixel blocks to be encoded in parallel is N1, and the number of boundary bit position candidates to be referred to when encoding each pixel block is N2, the setting means The larger the size set in (1), the larger the N1 is determined, and the larger the size set by the setting means is, the smaller the N2 is determined, the encoding parameter determining means;
By supplying the coefficient values of the N1 pixel blocks determined by the encoding parameter determining means to the boundary bit position determining means, N2 boundary bit position candidates are calculated for each pixel block, and each pixel The coefficient value of the block is supplied to each of the N2 encoding units, and the N2 boundary bit position candidates are set in each of the N2 encoding units, so that N2 units are provided for each pixel block. Encoding control means for generating encoded data;
An output means for selecting and outputting encoded data having a minimum encoded data amount among N2 encoded data generated from coefficient values of one pixel block for each of the N1 pixel blocks; Is provided.

  According to the present invention, it is possible to appropriately set the encoding speed and the encoding efficiency by determining the number of pixel blocks to be encoded in parallel and the number of boundary bit position candidates according to the image size. .

  Hereinafter, the present invention will be described in detail according to preferred embodiments with reference to the accompanying drawings.

  The image encoding apparatus described in the present embodiment performs frequency conversion and quantization processing in units of pixel blocks. Then, the multi-value coefficient value obtained by the quantization process is separated into an upper bit part higher than the boundary bit position (the setting will be described later) and a lower bit part below the boundary bit position and encoded. To do.

[First Embodiment]
FIG. 1 is a block configuration diagram of an image encoding device according to the present embodiment. Note that this image encoding apparatus will be described as being incorporated as an image encoding processing unit in a digital camera. A digital camera has an optical unit, an image sensor, a display unit, an operation unit, and a storage unit that stores a rewritable nonvolatile storage medium. These are directly connected to the encoding processing apparatus that is the subject of the present invention. Is not related and the description is omitted because it is the same as a known configuration. However, in the digital camera according to the embodiment, it is assumed that the user can set the resolution (size) of the photographed image by operating the operation unit before photographing. There are three selectable sizes (resolutions): L, M, and S. The L size is 4000 pixels in the horizontal direction and 3000 pixels in the vertical direction (hereinafter referred to as 4000 × 3000 pixels), the M size is 2000 × 1500 pixels, and the S size is 1600 pixels × 1200 pixels.

  As illustrated in FIG. 1, the image coding apparatus according to the embodiment includes an image data input unit 101, a block division unit 102, a series conversion unit 103, a coefficient quantization unit 103, a coefficient buffer 103, and a coefficient data acquisition unit 106. The apparatus also includes a user mode setting unit 107, an entropy encoding control unit 108, a boundary bit position determination unit 109, an entropy encoding processing unit 110, a code amount comparison unit 111, and an image data output unit 112. The entropy encoding processing unit 110 includes a plurality of entropy encoding units. In the case of the embodiment, the entropy encoding processing unit 110 includes four entropy encoding units 110a, 110b, 110c, and 110d.

The user mode setting unit 107 is provided in an operation unit (not shown) and can be set for the following three items.
-Image quality level (in the embodiment, high image quality and standard image quality)
・ Encoding mode (encoding speed priority / encoding amount suppression priority / auto)
・ Image size (L, M, S)
The user mode setting unit 107 can be composed of, for example, a display unit and operation keys of a digital camera. FIG. 15 shows an example of the setting screen of the display unit. The user operates the keys on the operation unit to make settings for three items. The example of FIG. 15 shows an example in which the image quality level is set to “high image quality”, the encoding mode is set to “Auto”, and the image size is set to “L”.

  The user mode setting unit 107 outputs a signal indicating the set image quality level among the above three items to the coefficient quantization unit 104 described later. Also, the user mode setting unit 107 outputs the other two items, that is, {encoding speed priority mode or code amount suppression priority mode or auto mode}, {L or M or S} to the entropy encoding control unit 108. To do.

  Normally, a digital camera generates R, G, and B three-component data for each pixel. However, the image sensor of the digital camera to which the present embodiment applies is monochrome multi-value image data (only luminance components) ( Assume that 8 bits per pixel = 256 gradations). This is for convenience and to simplify the following description. When one pixel is RGB three components, it is clear that the processing described below may be performed for each component. In addition, the color space is not limited to the one-component luminance space and the RGB color space, but may be a Luv space, a YCbCr space, a YMCK space, or the like. When the image encoding apparatus according to the present embodiment is applied to a device other than a digital camera, for example, when applied to a digital copying machine, the image input source is an image scanner. If applied to a PC, the image input source may be an image scanner or a video camera, or may be a computer-readable storage medium storing an uncompressed image data file in some cases. That is, the present invention is not limited by the type of image data input source.

  Hereinafter, the processing content of each processing unit will be described with reference to FIG.

  The image data input unit 101 inputs monochrome multi-valued image data captured by an image sensor (not shown) in the order of raster scanning, and outputs it to the block dividing unit 102. The block division unit 102 has an internal buffer memory for temporarily storing the image data from the image data input unit 101, and a coefficient for an image composed of 8 × 8 pixels (hereinafter referred to as a pixel block). The data is output to the conversion unit 103.

  The series conversion unit 1003 frequency-converts the pixel block output from the block division unit 1002 to generate 8 × 8 coefficient values and outputs the generated coefficient values to the coefficient quantization unit 104.

The frequency conversion described here applies DCT (discrete cosine transform) used in JPEG (ISO / IEC 154444-1). The JPEG discrete cosine transform processing is described in the following international standard recommendation, and therefore detailed description thereof is omitted here.
ITU-T Recommendation T.81 | ISO / IEC 10918-1: 1994, Information technology-Digital compression and coding of continuous-tone still images: Requirements and guidelines.

  As in the case of JPEG, level shift processing for subtracting an intermediate value from a pixel value is performed prior to application of discrete cosine transform. Since one pixel in this embodiment has a luminance component value represented by 8 bits, the cosine transform is performed after subtracting 128 from the pixel value.

  The coefficient quantization unit 104 outputs, for each pixel block output from the coefficient conversion unit 103, each DCT coefficient value included in the pixel block corresponding to the image quality level set by the user mode setting unit 107. Quantization is performed using the quantization matrix. If the image quality level is high, each quantization step value in the selected quantization matrix will have a smaller value than that for standard image quality. The coefficient quantization unit 104 stores each quantized coefficient value in the coefficient buffer 105. In the following description, when it is not necessary to distinguish the presence or absence of quantization, the “quantized coefficient value” stored in the coefficient buffer 105 is simply referred to as “coefficient value”. In addition, one coefficient value generated by the coefficient quantization unit 104 is assumed to be 9 bits in total, that is, 1 bit indicating a positive / negative sign and 8 bits indicating an absolute value portion. However, the number of bits expressing the coefficient value is not limited to this. In particular, 10 bits or more may be used if the accuracy is increased.

  As described above, from the user mode setting unit 107 to the entropy encoding control unit 108, signals indicating {encoding speed priority mode or code amount suppression priority mode or auto mode}, {L or M or S} are received. Supplied.

  The entropy encoding control unit 108 functions as an encoding parameter determination unit that determines the encoding parameters to be given to the coefficient data acquisition unit 106 and the boundary bit position determination unit 109. Specifically, the processing of the entropy coding control unit 108 when the coding speed priority mode is selected is equivalent to {auto mode & image L size} regardless of which image size is specified. Each processing unit is set as follows. Further, the processing of the entropy coding control unit 108 when the code amount suppression priority mode is selected is {auto mode & image S size} regardless of which of the image sizes L, M, and S is designated. Each processing unit is set so that the processing is equivalent to. Therefore, the encoding speed priority mode and the code amount suppression priority mode should be understood in accordance with the description of the auto mode described below.

  Hereinafter, a case where the auto mode is selected will be described.

  When the entropy encoding control unit 108 receives the information indicating that the auto mode is set, the entropy encoding control unit 108 generates a control signal (encoding parameter) corresponding to the image size selected by the user, the coefficient data acquisition unit 106, and the boundary bit position determination. Output to the unit 109 and the code amount comparison unit 111.

The processing of the coefficient data acquisition unit 106, the boundary bit position determination unit 109, and the code amount comparison unit 111 is as follows. The coefficient data acquisition unit 106 acquires the coefficient value of the pixel block from the coefficient buffer 105 in the raster scan order. The pixel block at the upper left corner of the image is defined as _0th, and the pixel block is defined as MB0. The i-th pixel block is represented as a pixel of interest MB _i . Here, i has a relationship of 0 ≦ i ≦ total number of pixel blocks−1.

-For image size L:
The coefficient data acquisition unit 106 acquires, from the coefficient buffer 105, coefficient values of a total of four pixel blocks, that is, the pixel block MB _{i of interest} and the subsequent three pixel blocks BM _{i + 1} , BM _{i + 2} , and BM _{i + 3.} To do. Then, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block of interest MB _i to the entropy encoding unit 110a in the entropy encoding processing unit 110. Further, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block MB _{i + 1} next to the pixel block of interest to the entropy encoding unit 110b. Further, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block of interest MB _{i + 2} to the entropy encoding unit 110c. Then, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block of interest MB _{i + 3} to the entropy encoding unit 110d.

Further, the coefficient data acquisition unit 106 outputs the coefficient values of the four pixel blocks of the pixel blocks MB _i , MB _{i + 1} , MB _{i + 2} , MB _{i + 3} to the boundary bit position determination unit 109 in this order. To do.

The boundary bit position determination unit 109 calculates a plurality of boundary bit position candidates from the coefficient values of the pixel block MB _i . Then, the boundary bit position determination unit 109 outputs the first boundary bit position candidate to the entropy encoding unit 110a. Similarly, the boundary bit position determination unit 109 outputs the first boundary bit position candidates of the pixel blocks MB _{i + 1} , MB _{i + 2} and MB _{i + 3} to the entropy encoding units 110b, 110c, and 110d. To do.

As a result, the entropy encoding units 110a to 100d encode the pixel blocks MB _i , MB _{i + 1} , MB _{i + 2} , and pixel block MB _{i + 3} using the given boundary bit positions, respectively. The encoded data is output to the code amount selection unit 111.

  The code amount comparison unit 111 outputs the encoded data from the entropy encoding units 110a, 110b, 110c, and 100d to the encoded data output unit 112 in this order.

As described above, four pixel blocks are encoded in parallel. Therefore, when the encoding process of the four pixel blocks is completed, the coefficient data acquisition unit 106 regards the pixel block MB _{i + 4} as the target pixel block and ends the encoding of the final pixel block without performing the above process. Until this process is repeated. Therefore, although the amount of encoded data to be generated is not necessarily minimized, the operation rate of the entropy encoding processing unit 110 is 100%, and high-speed encoding processing can be realized.

-For image size M:
The coefficient data acquisition unit 106 acquires, from the coefficient buffer 105, coefficient values of a total of two pixel blocks, the pixel block MB _{i of interest} and the subsequent pixel block BM _{i + 1} . Then, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block MB _i of interest to the entropy encoding units 110a and 110b in the entropy encoding processing unit 110. Also, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block MB _{i + 1} next to the pixel block of interest to the entropy encoding units 110c and 110d.

The boundary bit position determination unit 109 outputs the upper two boundary bit position candidates from the coefficient values of the pixel block MB _i to the entropy encoding units 110a and 110b. Similarly, the upper two candidates of the boundary bit position are output from the pixel block MB _{i + 1} to the entropy encoding units 110c and 110d.

  As a result, the entropy encoding units 110 a to 100 d encode the given coefficient values and output the encoded data to the code amount selection unit 111.

  The code amount comparison unit 111 first compares the data amounts of the encoded data from the entropy encoding units 110a and 110b. Then, the encoded data with a small amount of data is selected as the encoded data of the pixel block of interest and is output to the encoded data output unit 112. Next, the code amount comparison unit 111 compares the data amounts of the encoded data from the entropy encoding units 110c and 110d. Then, the encoded data with a small data amount is selected as the encoded data of the pixel block next to the pixel block of interest, and is output to the encoded data output unit 112.

  As described above, four entropy encoding units perform encoding processing for two pixel blocks. That is, the operating rate of the entropy encoding processing unit 110 can be set to 100%. Also, when the image size is M, two encoding processes are performed for one pixel block, and the one with the smaller encoded data amount is selected, so that the encoding efficiency can be further increased.

-For image size S:
The coefficient data acquisition unit 106 acquires the coefficient value of the pixel block of interest MB _i from the coefficient buffer 105. Then, the coefficient data acquisition unit 106 outputs the 64 coefficient values of the pixel block MB _i of interest to the entropy encoding units 110a, 110b, 110c, and 110d in the entropy encoding processing unit 110.

The boundary bit position determining unit 109 outputs the upper four candidates for the boundary bit position to the entropy encoding units 110a to 110d from the coefficient values of the pixel block MB _i .

  The code amount comparison unit 111 compares the data amounts of the four encoded data from the entropy encoding units 110a to 110d. Then, the encoded data with the smallest data amount is selected as the encoded data of the pixel block of interest, and is output to the encoded data output unit 112.

  As a result, since one pixel block is encoded at four boundary positions, the operation rate of the entropy encoding processing unit 110 can be 100%. Also, four encoding processes are performed for one pixel block, and the one with the smaller encoded data amount is selected, so that the encoding efficiency can be further increased. Further, since the total number of pixel blocks when the image size is S is smaller than that of the image sizes M and L, the time required for encoding does not matter.

  The process according to each image size (resolution) when the auto mode is selected has been described above.

  When the encoding speed priority mode is selected, the four entropy encoding units 110a to 110d perform encoding processing for four consecutive pixel blocks regardless of the image size. Also, when the code amount suppression priority mode is selected, each of the four entropy encoding units 110a to 110d performs encoding processing according to four different boundary bit positions for one pixel block regardless of the image size. The minimum encoded data is selected and output.

  Next, the boundary bit position determination unit 109 and the entropy encoding processing unit 110 will be described in more detail.

  First, the boundary bit position determining unit 109 will be described. FIG. 3 is a block configuration diagram of the boundary bit position determination unit 109. The boundary bit position determination unit 109 includes a valid bit determination unit 301, a frequency table generation unit 302, a frequency table buffer 303, and a boundary bit selection unit 304.

  In order to simplify the description, a process for determining the boundary bit position of one pixel block (target pixel block) will be described. Each of the 64 coefficient values in the pixel block of interest is represented as Cn.

The effective bit determination unit 301 obtains the effective bit number B (Cn) of each coefficient value Cn to be input. The number of effective bits indicates a bit position where “1” first appears when searching from the upper bit to the lower bit when the absolute value of the coefficient value Cn is expressed in a natural binary number. For example, when the coefficient value Cn of interest is “−5”, the natural value of the absolute value “5” is “0... 0101”. Since the most significant “1” is the third digit from the least significant, B (−5) = 3. In other words, it can be said that the effective number of bits of the coefficient value having the absolute value “5” is 3 bits from the least significant bit. When Cn = 0, there is no “1” bit, so the number of effective bits B (Cn) is zero. In short, the effective bit number B (Cn) can be said to be the smallest positive integer value P that satisfies the condition of | Cn | <2 ^P with respect to the absolute value | Cn | of the coefficient Cn.

  The frequency table generation unit 302 updates the occurrence frequency table held in the frequency table buffer 303 for B (Cn) output from the valid bit number determination unit 301. The coefficient Cn in the embodiment is represented by 8 bits excluding the positive and negative sign bits. Therefore, the frequency table buffer 303 has a capacity for storing the appearance frequencies N (0) to N (8) of the valid bit numbers 0 to 8, respectively. The frequency table buffer 308 is initialized with “0” when determining the frequency of one pixel block. When the effective bit determination unit 301 determines the number of effective bits B (Cn) of the coefficient of interest value Cn, the frequency table generation unit 302 increases N (B (Cn)) in the frequency table buffer 303 by “1”. Let That is, the frequency table generating unit 302 updates the frequency table buffer 303 every time the valid bit determining unit 301 determines a valid bit of one coefficient value.

  FIG. 4 shows an example in which the appearance frequency of the effective bits of all coefficient values in the pixel block of interest obtained as described above is shown as a graph. FIG. 4 shows an example in which there are 0 to 7 effective bits and the number of occurrences of 8 effective bits is 0.

  When the effective bit number determination and the frequency table update processing are performed for all the coefficient values of the target rectangular block, the boundary bit selection unit 304 calculates boundary bit candidates in the target rectangular block. Specifically, it is as follows.

In the boundary bit selection 304, first, the number N (b) of occurrences of each valid bit number b stored in the frequency table buffer 303 (b is an arbitrary integer value) is changed from the maximum value to the minimum value of the valid bit number. By performing cumulative addition, conversion into a table S (b) of cumulative frequency as shown in FIG. 5 is performed. In the pixel block of interest, when the maximum value of the number of effective bits having an appearance count of 1 or more is bmax (bmax = 7 in the case of FIG. 4), the boundary bit selection unit 304 uses variables b = bmax, bmax−1, Update bmax-2... to obtain a cumulative frequency table S (b) represented by the following equation.
S (b) = N (b) When b = bmax S (b) = N (b) + S (b + 1) When b ≠ bmax

Conversely, when the variable b = 0, 1, 2,..., Bmax is updated, it can be expressed as follows using the summation function Σ.
S (bmax−b) = ΣN (bmax−i) (i = 0, 1,..., B)

  In short, the cumulative frequency table S (b) indicates that when b ≠ 0, the cumulative frequency table S (b) represents the number of coefficient values whose absolute value is 2 ^ b or more. S (0) is equal to the number of all coefficient values in the target rectangular block.

  After obtaining the cumulative frequency table S (b), the boundary bit selection unit 304 examines the value of S (b) from the maximum value to the minimum value of the number of effective bits, that is, in descending order, and sets a preset threshold value. The maximum integer b exceeding Th is searched. Then, the boundary bit selection unit 304 determines the obtained b as the first candidate B1 of the boundary bit position.

  Here, the threshold Th is set to be a predetermined ratio with respect to the total number of coefficient values (in the case of the embodiment, since the pixel block has a size of 8 × 8 pixels, the total number of coefficient values is 64). This threshold value Th can be set or selectable by the user as appropriate, but from a statistical viewpoint, this threshold value Th is suitably ¼ to ３ of the total number of coefficients in the rectangular block. is there. For example, if the ratio 1/3 is adopted, the threshold Th is “21” (≈64 / 3).

  Further, the boundary bit selection unit 304 determines the second and subsequent candidates in the order of the bit distance from the first candidate B1. For example, “B1 + 1” is determined as the second candidate B2, “B1-1” is determined as the third candidate B3, and “B1 + 2” is determined as the fourth candidate B4. In other words, when the first candidate B1 is determined, the boundary bit selection unit 304 determines the second and subsequent candidates in the order of increasing distance from the center.

  Although details will be described later, each of the entropy encoding units 110a to 110d in the entropy encoding processing unit 110 separates the coefficient value Cn into upper and lower order by the boundary bit B, and has a continuous number of zeros for the upper bit part. Encoding, that is, run-length encoding is applied. Therefore, the role of the threshold Th here is for adjusting the ratio of the coefficient value in which the higher-order bit part of the coefficient value is non-zero and the coefficient value of zero, and aims to make run-length encoding function effectively. It can be said that it is set as. In the example shown in FIG. 5, when “5” is the maximum b exceeding the threshold Th, the boundary bit position B1 of the first candidate is “5”. The second candidate B2 is “6”, the third candidate B3 is “4”, and the fourth candidate B4 is “7”. When selecting a candidate, if the value of the candidate exceeds the possible range of the coefficient Cn, the value is ignored and determined within an effective range. For example, if the boundary bit position B is “8”, 7, 6, and 5 are determined as the values of the second to fourth candidates.

Here, the processing of the boundary bit position determination unit 109 when the auto mode described above is selected is summarized as follows. In the following description, the first candidate of the boundary bit position for the pixel block of interest MB _i is represented as B1 _i , the second candidate is represented as B2 _i , the third candidate is represented as B3 _i , and the fourth candidate is represented as B4 _i .
-When the image size is L:
The boundary bit position determination unit 109 includes first candidate boundary bit positions B1 _i and B1 _{i +} from a total of four pixel blocks MB _i , MB _{i + 1} , MB _{i + 2} , and MB _{i + 3} including the pixel block of interest. ₁ , B1 _{i + 2} and B1 _{i + 3} are calculated. In other words, the boundary bit position determination unit 109 does not output the boundary bit positions after the second candidate of each pixel block. Then, the boundary bit position determining unit 109 outputs the boundary bit position B1 _i to the entropy encoding unit 110a. Similarly, the boundary bit position determination unit 109 outputs the boundary bit positions B1 _{i + 1} , B1 _{i + 2} , and B1 _{i + 3} to the entropy encoding units 110b, 110c, and 110d, respectively.
・ When the image size is M:
The boundary bit position determining unit 109 selects the boundary bit positions B1 _i , B2 _i , B1 of the first candidate and the second candidate from MB _i and MB _{i + 1} of the total two pixel blocks of the pixel block of interest and the next pixel block, respectively. _{i + 1} and B2 _{i + 1} are calculated. Then, the boundary bit position determination unit 109 outputs the boundary bit position B1 _i to the entropy encoding unit 110a, and outputs the boundary bit position B2 _i to the entropy encoding unit 110b. Similarly, the boundary bit position determination unit 109 outputs the boundary bit positions B1 _{i + 1} and B2 _{i + 1} to the entropy encoding units 110c and 110d.
・ When the image size is S:
The boundary bit position determination unit 109 calculates first to fourth candidate boundary bit positions B1 _i , B2 _i , B3 _i , and B4 _i from the pixel block MB _{i of} interest. The boundary bit position determination unit 109 then outputs the boundary bit positions B1 _i , B2 _i , B3 _i , and B4 _i to the entropy encoding units 110a to 110d.

  Next, the entropy encoding processing unit 110 will be described. As shown in FIG. 1, the entropy encoding processing unit 110 according to the embodiment includes four entropy encoding units 110a to 110d. Each entropy encoding unit encodes each coefficient Cn (64 pieces) in one pixel block input from the coefficient data acquisition unit 106 according to the boundary bit position input from the boundary bit position determination unit 109. The configuration is the same. Therefore, only the entropy encoding unit 110a will be described below.

  FIG. 2 is a block configuration diagram of the entropy encoding unit 110a. As illustrated, the entropy encoding unit 110a includes an upper / lower bit separation unit 200, an upper bit encoding unit 201, a lower bit encoding unit 202, a code buffer 203, and a code string forming unit 204.

  Hereinafter, the boundary bit position set by the boundary bit position determination unit 109 to the entropy encoding unit 110a will be described as “B”.

  The upper / lower bit separation unit 200, for the coefficient value Cn of the pixel block of interest output from the coefficient data acquisition unit 106, the upper bit part obtained by removing the lower B bits in the code absolute value expression, and the boundary bit position and below The lower B bits are separated. However, positive and negative sign bits are not processed. More specifically, the upper / lower bit separation unit 200 includes the upper bit part Un (= Cn >> B) and the lower bit part Ln (= | Cn | & ((1 << B) of the target coefficient value Cn. ) -1)) is obtained. Here, Y >> X indicates that the value Y is shifted in the lower direction by X bits, and Y << X indicates that the value Y is shifted in the upper direction by X bits. | Y | represents the absolute value of the value Y, and X & Y represents the logical product (AND) of the value X and the value Y in bit units.

  The upper / lower bit separation unit 200 outputs the bit part Un of the target coefficient value Cn to the upper bit encoding 201 and outputs the lower bit part Ln to the lower bit encoding unit 202.

  It should be noted that for the coefficient Cn whose value excluding the sign bit of the upper bit part Un is “0”, the sign indicating the sign of the coefficient Cn is also passed to the lower bit encoding part 202.

  FIG. 6 shows an example in which “−41, −7, 28, −1,...” Is input as an example of the coefficient value Cn, and the upper bit part Un and the lower bit part Ln are separated at the boundary bit position B = 4. .

  In the case illustrated, the upper / lower bit separation unit 200 outputs 64 upper bit units Un including one bit of the positive / negative code to the upper bit encoding unit 201.

  When the value represented by 4 bits excluding the sign bit of the input upper bit part Un is “0”, the upper bit coding unit 201 does not code the sign bit here. In the upper bit part Un, the value represented by 4 bits excluding the sign bit is “0” in the case of FIG. 6, in the case of FIG. 6, coefficient values Cn = “− 7”, “−1”. When the value represented by 4 bits excluding the sign bit of the upper bit part Un is non- “0”, the upper bit part Un including the sign bit is directly subjected to run length encoding.

  Accordingly, in the case of FIG. 6, the upper bit encoding unit 201 performs run-length encoding in the order of multi-value data “−2”, “0”, “1”, “0”. Become. As a result, the probability that the upper bit part is “0” is increased, and the encoding efficiency can be increased.

  Note that one DC coefficient and 63 AC coefficients are obtained from one pixel block. The upper bit encoding unit 201 in the embodiment uses the same method as the encoding of DC and AC coefficients in JPEG Baseline. That is, when the upper bit part Un of the DC coefficient of the pixel block of interest is encoded, the upper bit part Un of the DC coefficient of the immediately preceding pixel block (which is also an already encoded block) is used as a predicted value. Then, the difference between the upper bit portion Un of the pixel block of interest and the predicted value is calculated, and Huffman coding is performed based on the difference. Then, the Huffman coding is performed on the upper bit part Un of the AC coefficient by combining the number of consecutive zeros and the non-zero coefficient. Similar to JPEG Baseline, a special code representing EOB (End Of Block) is used when the end of the block ends with a sequence of zeros.

  On the other hand, the upper / lower bit separation unit 200 outputs the 64 lower bit parts Ln of the target rectangular block to the lower bit encoding unit 202. However, as described above, the upper / lower bit separation unit 200, for the coefficient Cn whose value of 4 bits excluding the sign bit of the upper bit part Un is “0”, converts the sign bit to the lower bit. It is added to the part Ln and output to the lower bit encoding part 202.

  In the example of FIG. 6, the coefficient values “−7” and “−1” indicate that the value of the upper bit part Un excluding the sign bit is “0”. Therefore, the addition position of the plus / minus sign bit is set to a position where “1” first appears from the high order to the low order of the low order bit part Ln. In the case of FIG. 6, the position indicated by the diagonal lines is the position where the positive and negative bits are inserted.

  The lower bit encoding unit 202 encodes the lower bit part Ln output from the upper / lower bit separation unit 200. Since the lower bit part Ln is generally data that is difficult to compress with respect to the upper bit part Un, the lower bit encoding part 202 of this embodiment outputs the lower bit part Ln without compression (however, in pursuit of compression performance) If you do, you may apply arithmetic codes). Since the data is not compressed, the input data is basically encoded data as it is, but the bit data is arranged according to the degree of influence on the image quality, in other words, the decoding apparatus can perform progressive reproduction. Therefore, the data are rearranged in the order of bit planes and output.

  In the case of FIG. 6, the lower bit encoding unit 202 stores the input lower bit part Ln in an internal buffer (not shown), and is configured with the most significant bit digit (Bth digit) of each lower bit part Ln. Output a bit plane. This process is repeated up to the least significant bit digit (first digit).

  An example of data output by the lower bit encoding unit 202 in the case of the coefficient value sequence “−41, −7, 28, −1,...” (Boundary bit position B = 4) illustrated in FIG. Shown in

  The plane (bit information at the same bit position) output at the beginning of the lower bit part Ln is a B-digit bit plane “1010...”. The coefficient values “−41” and “28” correspond to “1”. Since the value represented by the upper bit part Un excluding the sign bits of the coefficient values “−41” and “28” is non-zero, the sign bit is not inserted.

  Focusing on the next bit plane of B-1 digits, the first “1” bit representing the coefficient value “−7” appears (see FIG. 6). Further, the value of the upper bit part Un excluding the sign bit of the coefficient value “−7” is “0”. Therefore, as shown in the figure, positive and negative sign bits are inserted at positions following the first “1” representing the coefficient value “−7”. Hereinafter, the B-2 digit bit plane and the B-3 digit bit plane are output. In the B-3 digit bit plane, since the first “1” representing the coefficient value “−1” appears, the sign bit is inserted subsequently.

  The encoding buffer 203 stores encoded data in units of pixel blocks output from the upper bit encoding unit 201 and the lower bit encoding unit 202. The encoding buffer 203 also stores information on boundary bit positions set by the boundary bit position determination unit 109.

  The code string forming unit 204 forms encoded data of a pixel block having a structure as shown in FIG. 8 from the encoded data of one pixel block stored in the code buffer 203, and outputs the encoded data to the code amount comparing unit 111. In FIG. 8, information on the boundary bit position B of the pixel block of interest is stored in the block header. Subsequently, the encoded data of the upper bit part Un and the encoded data of the lower bit part Ln are stored.

  The processing content of the entropy encoding unit 110a has been described above. The other entropy encoding units 110b, 110c, and 110d are different only in the coefficient values or boundary bit positions given from those of the entropy encoding unit 110a.

  The code amount comparison unit 111 is as described above, but is summarized as follows based on the above description. The following is an explanation when the user operates the user mode setting unit 107 to select the auto mode.

It is assumed that the code amount comparison unit 111 includes an internal buffer having an appropriate capacity.
-When the image size is L:
The entropy encoding units 110a to 110d output encoded data of the pixel blocks MB _i , MB _{i + 1} , MB _{i + 2} , MB _{i + 3} . Therefore, the code amount comparison unit 111 assumes that each encoded data is the encoded data of the pixel blocks MB _i , MB _{i + 1} , MB _{i + 2} , MB _{i + 3} and outputs the encoded data to the encoded data output unit 112 as it is. To do.
・ When the image size is M:
Entropy encoding unit 110a, 110b outputs the encoded data to the first candidate B1 _i boundary bit position of the pixel block MB _i, the second candidate B2 _i as a parameter. Therefore, the code amount comparison unit 111 compares the data amounts of the two encoded data, selects the smaller code amount as the encoded data of the pixel block MB _i , and outputs the selected data to the encoded data output unit 112. Also, the code amount comparison unit 111 discards non-selected encoded data.

Further, the entropy encoding units 110c and 110d output the encoded data of the first candidate B1 _{i + 1} and the second candidate B2 _{i + 1} at the boundary bit positions of the pixel block MB _{i + 1} . Therefore, the code amount comparison unit 111 compares the data amounts of the two encoded data, selects the smaller code amount as the encoded data of the pixel block MB _{i + 1} , and outputs it to the encoded data output unit 112. To do. Also, the code amount comparison unit 111 discards non-selected encoded data.
・ When the image size is S:
The entropy encoding units 110a to 110d output encoded data of the first candidate B1 _i , the second candidate B2 _i , the third candidate B3 _i , and the fourth candidate B4 _i at the boundary bit position of the pixel block MB _i . Therefore, the code amount comparison unit 111 compares the data amounts of these four pieces of encoded data, selects the encoded data having the minimum code amount as the encoded data of the pixel block MB _i , and outputs the encoded data output unit 112. Output to. Also, the code amount comparison unit 111 discards the three non-selected encoded data.

  Next, the encoded data output unit 112 will be described.

  The encoded data output unit 112 generates and outputs a file header prior to inputting the encoded data from the code amount comparison unit 111. In this file header, the number of horizontal pixels of the image, the number of vertical pixels, the number of color components (the number of color components is one because it is an example of a monochrome image in the embodiment), the number of bits of each pixel, etc. Various information necessary for decoding is added as a header.

  Thereafter, the encoded data output unit 112 outputs the encoded data of each pixel block from the code amount comparison unit 111 so as to follow the previously generated file header. FIG. 9 shows a file structure of encoded image data in the embodiment.

  Note that the file structure of the finally generated encoded image data is not limited to FIG. For example, the file structure shown in FIG. 10 may be used.

  For example, as shown in FIG. 10, various information necessary for decoding, such as the number of pixels in the horizontal direction and the number of pixels in the vertical direction, the number of color components, and the number of bits of each pixel, are added to the file header. This header also includes boundary bit position information for all blocks. That is, if this header is analyzed on the decoding side, the boundary bit position of each block can be obtained. After the header, the upper bit encoded data is arranged for all the blocks of the image, and then the lower bit encoded data is arranged. In the figure, N represents the total number of blocks, and L represents the maximum value of the boundary bit position B searched for all blocks. The upper bit encoded data is configured by concatenating the upper bit portion encoded data of each block in the order of raster scanning. The low-order bit encoded data is configured by arranging low-order bit part encoded data of each block in the order of bit planes. The bit plane i is configured by collecting the data of the bit plane i of each block in the raster scan order, but does not include the data from the block where the boundary bit B <i, and therefore the data is not necessarily collected from all the blocks. It is not done.

  The configuration and processing of the image encoding device according to the embodiment have been described above. The above embodiment is summarized as follows in an easy-to-understand manner.

  First, it is assumed that the number of entropy encoding units included in the entropy encoding processing unit 110 is N0 (in the embodiment, N0 = 4). Further, the number of pixel blocks to be encoded at consecutive positions is defined as N1, and the number of boundary bit position candidates used when encoding one pixel block is defined as N2. Here, the relationship of N0, N1, and N2 is N0 = N1 × N2.

  Under this definition, the encoding process of the present embodiment will be described. As the image size increases, the encoding parameter N1 is increased and N2 is decreased to perform the encoding process. Conversely, the smaller the image size, the smaller N1 and the larger N2, and the encoding process is performed. Then, the coefficient values of the determined N1 pixel blocks are supplied to the boundary bit position determination unit 109, so that N2 boundary bit position candidates are calculated for each pixel block. Then, the coefficient value of each pixel block is supplied to each of the N2 entropy encoding units, and the N2 boundary bit position candidates are set in each of the N2 entropy encoding units. N2 encoded data are generated every time. As a result, N2 encoded data are generated from the coefficient values of one pixel block. Then, the code amount comparison unit 111 selects and outputs encoded data that is the minimum encoded data amount among the N2 encoded data. The code amount comparison unit 111 performs this process for each of the N1 pixel blocks.

  As a result, the larger the image size, the greater the number of pixel blocks to be encoded per unit time. On the other hand, the smaller the image size, the smaller the number of pixel blocks to be encoded per unit time, but the encoding efficiency can be increased.

  In the embodiment, it has been described that N0 = N1 × N2, but this requirement is not necessarily satisfied.

For example, when N0 = 4, the block MB _i is encoded using two boundary bit position candidates, and the blocks BM _{i + 1} and MB _{i + 2} are encoded using one boundary bit position candidate. It doesn't matter. In this case, for the block MBi, the one with the smaller amount of encoded data is selected, but whichever is selected does not affect the image quality. This is because the entropy coding itself is reversible and the image quality is determined by the quantization step value by the coefficient quantization unit 104. In short, the number of boundary bit position candidates is reduced as high-speed encoding is desired. In other words, the boundary bit positions may be increased as the encoding efficiency is increased. As a result, it is possible to cope with a digital camera capable of photographing at a resolution exceeding three types.

  In the above embodiment, the pixel block size is 8 × 8 pixels, and the boundary bit position is determined in units of the pixel block. However, the present invention is not limited to this. For example, one pixel block has a size of n × m pixels (n and n are integers of 2 or more), and series conversion and coefficient quantization are executed in units of this pixel block. Then, the boundary bit position determining unit 109 determines the boundary position in units of macroblocks composed of a plurality of pixel blocks (p × n) × (q × m), and each entropy encoding unit is also in units of macroblocks. In addition, the macro bit may be encoded in units by separating the upper bit part and the lower bit part. In this case, it is desirable that the prediction value used when encoding the DC component of the first pixel block in one macroblock is “0”. When the DC component of the second and subsequent pixel blocks is encoded, the DC component value of the immediately preceding pixel block is used as the predicted value. In this way, it becomes possible to decode the macro block alone.

  In the embodiment, an example in which there are three types of image sizes and four encoding units (entropy encoding units in the embodiment) has been described. However, the present invention is not limited to this number.

  As described above, according to the present embodiment, when the auto mode is selected, the operation rate of the plurality of encoding units can be set to 100% according to the size (resolution) of the image to be encoded. it can. In addition, when encoding a large image, although there is a possibility that the encoding efficiency may slightly decrease, it is possible to encode at a high speed. On the other hand, in the case of an image with a small size, since the number of blocks as a unit for encoding processing is small, it is possible to perform processing for improving the encoding efficiency by that amount.

  In the embodiment, the example applied to the digital camera has been described. However, the image data input source may be applied to an apparatus (for example, a digital copying machine) on which an image scanner is mounted or connectable. In this case, “image size” can be easily understood by replacing “original size”. Further, when the reading resolution of the image scanner is variable, the above embodiment can be used as it is.

[Modification of First Embodiment]
An example in which processing equivalent to that in the first embodiment is realized by a computer program will be described below as a modification of the first embodiment.

  FIG. 14 is a block configuration diagram of an information processing apparatus (for example, a personal computer) in the present modification.

  In the figure, reference numeral 1401 denotes a CPU which controls the entire apparatus using programs and data stored in a RAM 1402 and a ROM 1403 and executes image encoding processing and decoding processing described later. A RAM 1402 includes an area for storing programs and data downloaded from the external device via the external storage device 1407, the storage medium drive 1408, or the I / F 1409. The RAM 1402 also includes a work area used when the CPU 1401 executes various processes. Reference numeral 1403 denotes a ROM which stores a boot program, a setting program for the apparatus, and data. Reference numerals 1404 and 1405 denote a keyboard and a mouse, respectively. Various instructions can be input to the CPU 1401.

  A display device 1406 includes a CRT, a liquid crystal screen, and the like, and can display information such as images and characters. Reference numeral 1407 denotes a large-capacity external storage device such as a hard disk drive device. The external storage device 1407 stores an OS (operating system), a program for image encoding and decoding processing described later, image data to be encoded, encoded data of an image to be decoded, and the like as files. The CPU 1401 loads these programs and data into a predetermined area on the RAM 1402 and executes them.

  A storage medium drive 1408 reads programs and data recorded on a storage medium such as a CD-ROM or DVD-ROM and outputs them to the RAM 1402 or the external storage device 1407. In addition, a program for image encoding and decoding processing described later, image data to be encoded, encoded data of an image to be decoded, and the like may be recorded on the storage medium. In this case, the storage medium drive 1408 loads these programs and data into a predetermined area on the RAM 1402 under the control of the CPU 1401.

  Reference numeral 1409 denotes an I / F, which connects an external device to the present apparatus through the I / F 1409 and enables data communication between the present apparatus and the external apparatus. For example, image data to be encoded, encoded data of an image to be decoded, and the like can be input to the RAM 1402, the external storage device 1407, or the storage medium drive 1408 of this apparatus. A bus 1410 connects the above-described units.

  In the above configuration, when the power of this apparatus is turned on, the CPU 1401 loads the OS from the external storage device 1407 to the RAM 1402 in accordance with the boot program stored in the ROM 1403. As a result, the keyboard 1404 and the mouse 1405 can be input, and the GUI can be displayed on the display device 1406. When the user operates the keyboard 1404 or the mouse 1405 to give an instruction to start an application program for image encoding processing stored in the external storage device 1407, the CPU 1401 loads the program into the RAM 1402 and executes it. As a result, the present apparatus functions as an image encoding apparatus.

  Hereinafter, the processing procedure of the application program for image encoding processing executed by the CPU 1401 will be described with reference to the flowchart of FIG. Basically, this program includes functions corresponding to the components shown in FIG. However, the areas of the coefficient buffer 105 in FIG. 1, the code buffer 203 in FIG. 2, and the frequency table buffer 303 in FIG. 3 are reserved in the RAM 1402 in advance. In this modification, an example in which image data to be encoded is stored as a file in the external storage device 1407 and the result of encoding processing is also stored as a file in the external storage device 1407 will be described. However, the image data input source may be an image scanner connected to the interface 1409, and the output destination of the encoded image data may be a network.

  When the user designates an image file to be encoded, the CPU 1401 can determine the size (number of pixels) of the image by analyzing the file header. In step S <b> 1, the CPU 1401 compares the determined image size with a preset threshold value and determines whether the image corresponds to L, M, or S.

  If it is determined in step S1 that the image to be encoded is size S, the process proceeds to step S2, and a variable Pn that defines the number of pixel blocks to be encoded in parallel is set to “1”. Also, “4” is set to the variable Qn that defines the number of boundary bit position candidates to be referred to.

  If it is determined in step S1 that the image to be encoded is size M, the process proceeds to step S3, and the variable Pn that defines the number of pixel blocks to be encoded in parallel is set to “2”. . Further, “2” is set to the variable Qn that defines the number of candidates for the boundary bit position to be referred to.

  If it is determined in step S1 that the image to be encoded is size L, the process proceeds to step S4, and the variable Pn that defines the number of pixel blocks to be encoded in parallel is set to “4”. . Further, “1” is set to the variable Qn that defines the number of candidates for the boundary bit position to be referred to.

  The processes in steps S2 to S4 correspond to the entropy encoding control unit 108 in FIG.

  Next, the process proceeds to step S5, and Pn pixel blocks are input. This is a process corresponding to the coefficient data acquisition unit 106.

  In step S6, Qn boundary bit position candidates are calculated for each of the Pn pixel blocks. This is a process corresponding to the boundary bit position determination unit 109.

  In step S7, Pn × Qn encoded data is generated in each of the Pn pixel blocks using Qn boundary bit threshold values (corresponding to the entropy encoding processing unit 110).

  In step S8, code amount comparison & output processing is performed. The contents of this process are the same as those of the code amount comparison unit 111 described above.

  In step S9, it is determined whether or not the encoding process for all pixel blocks has been completed. If not, the process from step S5 is performed to encode the next Pn pixel blocks.

  If it is determined in step S9 that the encoding process has been completed for all pixel blocks, this process ends.

  As described above, according to the present modification, processing equivalent to that of the first embodiment described above can be realized by a computer as a computer program.

[Second Embodiment]
A second embodiment will be described. The apparatus configuration of the second embodiment is the same as that of FIG. 1 except that the user mode setting unit 107 can set either a moving image shooting mode or a still image shooting mode. Therefore, the apparatus configuration will be described with reference to FIG. In order to simplify the description, it is assumed that the auto mode of the first embodiment is set in the still image shooting mode. Further, in the moving image shooting mode, it is assumed that one input image corresponds to one frame in the moving image, and that moving image shooting continues until a moving image shooting stop instruction is input. .

  FIG. 12 is a flowchart showing the processing contents of the entropy encoding control unit 108. Hereinafter, the processing procedure of the entropy encoding control unit 108 will be described with reference to FIG.

  First, in step S <b> 21, the entropy encoding control unit 108 acquires a shooting mode from the user mode setting unit 107. In step S22, it is determined whether the set shooting mode is a moving image shooting mode or a still image shooting mode. If it is determined that the mode is the moving image shooting mode, the process proceeds to step S23, and a variable Pn that defines the number of pixel blocks to be encoded in parallel is set to “4”. Further, “1” is set to the variable Qn that defines the number of candidates for the boundary bit position to be referred to. In the case of a moving image, if the frame rate is 30 frames / second, encoding per frame must be performed within 1/30 seconds. Therefore, in the embodiment, “4” is determined for the variable Pn and “1” is determined for the variable Qn.

  On the other hand, if it is determined in step S22 that the still image shooting mode is set, the process proceeds to step S24. In the still image shooting mode, the time required for encoding one image is not required to be as fast as a moving image. Therefore, in step S24, the variable Pn and the variable Qn are determined according to the auto mode of the first embodiment. That is, when the image size is L, Pn is determined to be “4” and the variable Qn is determined to be “1”. When the image size is M, Pn is determined to be “2” and Qn is determined to be “2”. When the image size is S, Pn is determined to be “1” and Qn is determined to be “4”.

  Thereafter, the process proceeds to step S25, and the entropy encoding control unit 108 outputs the determined Pn and Qn to the coefficient data acquisition unit 106 and the boundary bit position determination unit 109, respectively.

  The processing of the coefficient data acquisition unit 106 is the same as that in the first embodiment. The processing of the boundary bit position determination unit 109 may be the same as that of the first embodiment, and when it is necessary to perform processing at a higher speed, a fixed bit position at the boundary bit position for all pixel blocks is used. The bit position may be set. For example, when the coefficient bit string is formed of 8 bits, the boundary bit position may be determined to be the third bit from the lower order.

  The second embodiment has been described above. According to the second embodiment, in the moving image shooting mode, in consideration of the necessity of encoding each frame in real time, the number of boundary bit position candidates is set to “1”, and the four entropy encoding units 110a. From 4 to 110d, four consecutive pixel blocks are encoded. As a result, it is possible to increase the encoding speed and to perform encoding with no missing frames.

  In the second embodiment, since the configuration of FIG. 1 is adopted, the number of entropy encoding units is four. However, the present invention is not limited to this number. The reason is as described in the first embodiment.

[Third Embodiment]
In the third embodiment, an example will be described in which the variable Pn and the variable Qn for encoding a still image are set according to the free space of a storage medium (such as a CF memory card) that stores compressed and encoded image data. In order to simplify the explanation,
The apparatus configuration is the same as in the first embodiment, and a description thereof is omitted. However, although the configuration for detecting the free capacity of the output destination storage medium of the encoded data output unit 112 is increased, a normal digital camera can access the file system of the storage medium and detect the free capacity. Since this is done, the configuration is omitted.

  FIG. 13 is a flowchart showing a processing procedure of the entropy coding control unit 108 in the third embodiment. Hereinafter, description will be given with reference to FIG.

  First, in step S31, the entropy encoding control unit 108 acquires a free space Mc of the storage medium. In step S32, it is determined whether or not the acquired free capacity Mc is equal to or less than a preset threshold value Thm. When it is determined that Mc ≦ Thm, that is, when it is determined that the free capacity is equal to or less than the capacity indicated by the threshold Thm, the process proceeds to step S33, and a variable that defines the number of pixel blocks to be encoded in parallel is determined. Pn is set to “1”. Also, “4” is set to the variable Qn that defines the number of boundary bit position candidates to be referred to. As a result, encoded data with a high compression rate is generated.

  On the other hand, if it is determined in step S32 that Mc> Thm, that is, the storage medium has sufficient free space, the process proceeds to step S34. In step S34, the variable Pn and the variable Qn are determined according to the auto mode of the first embodiment. That is, when the image size is L, Pn is determined to be “4” and the variable Qn is determined to be “1”. When the image size is M, Pn is determined to be “2” and Qn is determined to be “2”. When the image size is S, Pn is determined to be “1” and Qn is determined to be “4”.

  Thereafter, the process proceeds to step S35, and the entropy encoding control unit 108 outputs the determined Pn and Qn to the coefficient data acquisition unit 106 and the boundary bit position determination unit 109, respectively.

  The processing of the coefficient data acquisition unit 106 and the boundary bit position determination unit 109 is the same as that in the first embodiment.

  As described above, according to the third embodiment, when the free capacity of the storage medium is reduced, the parameter is switched to a parameter for increasing the compression rate, so that the number of photos that can be stored can be increased. Become.

[Other Embodiments]
A fourth embodiment will be described. The apparatus configuration in the fourth embodiment is also the same as that in the first embodiment. However, the battery driving mode is different in that either a power saving mode or a normal mode (non-power saving mode) can be set. This setting is performed from an operation panel (not shown).

  Now, let us say that the operating rate of the entropy encoding processing unit 110 is 100% in both the power saving mode and the normal mode. In this case, in order to reduce the power consumption, it is necessary to shorten the time required for encoding.

  Therefore, when the power saving mode is set, Pn = 4 and Qn = 1 are set, and in the normal mode, the same processing as in the first embodiment may be performed.

  As described above, each embodiment according to the present invention has been described, but it is obvious that the processing according to the second to fourth embodiments can be realized by a computer program, as in the modification of the first embodiment described above. It is. In general, the computer program is stored in a computer-readable storage medium such as a CD-ROM, and is set in a reading device (CD-ROM drive or the like) included in the computer and copied or installed in the system. Become executable. Therefore, it is obvious that such a computer readable storage medium is also within the scope of the present invention.

It is a block block diagram of the image coding apparatus which concerns on 1st Embodiment. It is a block block diagram of the entropy encoding part in 1st Embodiment. It is a block block diagram of the boundary bit position determination part which concerns on 1st Embodiment. It is a figure which shows the frequency stored in the frequency table buffer in 1st Embodiment as a graph. It is a figure which shows the example of the data of the accumulation frequency table in 1st Embodiment. It is a figure which shows the outline of isolation | separation of the coefficient value by an upper / lower bit separation part. It is a figure which shows the example of the bit coding data of a low-order bit part. It is a figure which shows the data structure of the encoded pixel block. It is a figure which shows the data structure of the encoding image data file in 1st Embodiment. It is a figure which shows the other example of the data structure of the encoding image data file in 1st Embodiment. It is a flowchart which shows the process sequence of the modification of 1st Embodiment. It is a flowchart which shows the process sequence of the entropy encoding control part in 2nd Embodiment. It is a flowchart which shows the process sequence of the entropy encoding control part in 3rd Embodiment. It is a block block diagram of a computer. It is a figure which shows an example of the screen of the mode setting in the digital camera employ | adopted by embodiment.

Claims (9)

An image encoding device that performs frequency conversion and quantization processing in units of pixel blocks from multivalued image data, and encodes each multivalued coefficient obtained by the quantization processing,
Each coefficient value of the pixel block is separated into an upper bit part higher than a set boundary bit position and a lower bit part below the set boundary bit position, and run-length encoding is performed on the upper bit part A plurality of encoding means for encoding the lower bit part in bit plane units;
For each coefficient of the pixel block of interest, the number of bits from the bit position that first becomes “1” from the most significant bit to the least significant bit to the least significant bit is detected as the effective bit number, and the detected effective bit number appears. Boundary bit position determining means for determining a plurality of candidates for the boundary bit position for each coefficient of the pixel block of interest based on the number of times;
Setting means for setting the size of the image to be encoded;
When the number of the plurality of encoding means is N0, the number of pixel blocks to be encoded in parallel is N1, and the number of boundary bit position candidates to be referred to when encoding each pixel block is N2, the setting means The larger the size set in (1), the larger the N1 is determined, and the larger the size set by the setting means is, the smaller the N2 is determined, the encoding parameter determining means;
By supplying the coefficient values of the N1 pixel blocks determined by the encoding parameter determining means to the boundary bit position determining means, N2 boundary bit position candidates are calculated for each pixel block, and each pixel The coefficient value of the block is supplied to each of the N2 encoding units, and the N2 boundary bit position candidates are set in each of the N2 encoding units, so that N2 units are provided for each pixel block. Encoding control means for generating encoded data;
An output means for selecting and outputting encoded data having a minimum encoded data amount among N2 encoded data generated from coefficient values of one pixel block for each of the N1 pixel blocks; An image encoding device comprising: The boundary bit position determining means includes
The number of occurrences of the effective number of bits b is defined as N (b), the maximum number of effective bits with the number of occurrences of 1 or more is defined as bmax, and the summation function of the following equation for the variable b = 0, 1, 2,.
S (bmax−b) = ΣN (bmax−i) (i = 0, 1,..., B)
In relation to a preset threshold value Th,
S (bmax−b)> Th
The maximum integer b that satisfies the following condition is obtained, and the second integer and the subsequent candidates are determined in order of the first integer at the boundary bit position and the bit distance from the first candidate. Image encoding device. Furthermore, an imaging means is provided,
The image encoding apparatus according to claim 1, wherein the setting unit sets a size of an image captured by the imaging unit. Furthermore, an image pickup means, a selection means for selecting either a moving image shooting mode or a still image shooting mode using the image pickup means,
The encoding parameter determining means sets the N1 to N0 and the N2 to "1" only when the moving image shooting mode is selected by the selecting means. 2. The image encoding device according to 2. Furthermore, a detection means for detecting the free capacity of the storage medium for storing the encoded image data is provided,
The encoding parameter determining means sets the N1 to N0 and the N2 to “1” only when the free space detected by the detecting means is equal to or less than a preset capacity. The image encoding device according to claim 1 or 2. Furthermore, a selection means for selecting either the power saving mode or the non-power saving mode is provided,
3. The encoding parameter determining means sets the N1 to N0 and sets the N2 to “1” only when the power saving mode is selected by the selecting means. The image encoding device described in 1. A control method for an image encoding device that performs frequency conversion and quantization processing in units of pixel blocks from multi-value image data, and encodes each multi-value coefficient obtained by the quantization processing,
Each coefficient value of the pixel block is separated into an upper bit part higher than a set boundary bit position and a lower bit part below the set boundary bit position, and run-length encoding is performed on the upper bit part A plurality of encoding steps for encoding the lower bit part in bit plane units;
For each coefficient of the pixel block of interest, the number of bits from the bit position that first becomes “1” from the most significant bit to the least significant bit to the least significant bit is detected as the effective bit number, and the detected effective bit number appears. A boundary bit position determining step for determining a plurality of candidates for the boundary bit position for each coefficient of the pixel block of interest based on the number of times;
A setting step for setting the size of the image to be encoded;
When the number of the plurality of encoding steps is N0, the number of pixel blocks to be encoded in parallel is N1, and the number of boundary bit position candidates to be referred to when encoding each pixel block is N2, the setting step An encoding parameter determining step of determining the larger N1 as the size set in (2) is larger and determining the smaller N2 as the size set in the setting step is larger;
By supplying the coefficient values of N1 pixel blocks determined in the encoding parameter determination step to the boundary bit position determination step, N2 boundary bit position candidates are calculated for each pixel block, and each pixel The coefficient value of the block is supplied to each of the N2 encoding processes, and the N2 boundary bit position candidates are set in each of the N2 encoding processes, so that N2 number of pixels for each pixel block are set. An encoding control step for generating encoded data;
An output step of selecting and outputting encoded data having a minimum encoded data amount among N2 encoded data generated from coefficient values of one pixel block for each of the N1 pixel blocks; A method for controlling an image encoding device comprising:   A computer program that causes a computer to function as the image encoding device according to claim 1 by being read and executed by a computer.   A computer-readable storage medium storing the computer program according to claim 8.

Images