JPH0353666A - Picture signal compression coder - Google Patents

Abstract

PURPOSE:To attain proper fixed length processing of a coded data by setting a normalizing coefficient alpha at the outside of a coder and applying fixed length processing with the coefficient. CONSTITUTION:Whether or not a normalizing coefficient alpha is inputted to a coder 28 from an alpha setting section 18 is discriminated, and when the normalizing coefficient alpha is inputted, the normalizing coefficient alpha is used to limit a coded data quantity and after the obtained code quantity is discriminated to be close to an object, the result is outputted from a gate. If the obtained code quantity exceeds the object or is not close to the object, the more proper normalizing coefficient alpha is set by the alpha setting section 18 and sent to an alpha discrimination section 16. Thus, the optimum normalizing coefficient alpha is obtained at the outside of the coder 28 for the fixed length processing. Thus, the proper fixed length processing of a coded data is implemented.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image signal compression encoding device, and more particularly to an image signal compression encoding device that makes the amount of data of a compression encoded image constant.

When storing digital image data in memory, such as image data taken by an electronic still camera, it is necessary to reduce the amount of data and reduce the storage capacity of the memory.
Various compression encoding methods are used. In particular, two-dimensional orthogonal transform encoding is widely used because it can perform encoding at a high compression rate and can also suppress image distortion caused by encoding.

In such two-dimensional orthogonal transform encoding, image data is divided into a predetermined number of blocks, and the image data within each block is subjected to two-dimensional orthogonal transform. The orthogonally transformed image data, that is, the transformation coefficients, are compared with a predetermined threshold, and portions below the threshold are truncated (coefficient truncation). Next, quantization using a predetermined step width, that is, normalization, is performed. This suppresses the value of the conversion coefficient, that is, the dynamic range of the amplitude.

During this time, the comparison with the value and the normalization process are often performed at the same time. That is, when the transform coefficients are normalized by a predetermined normalization coefficient and the result is converted into an integer, the transform coefficient having a value lower than the normalization coefficient becomes O.

The normalized transform coefficients are then Huffman encoded and
It is stored on a recording medium such as a memory card.

In such two-dimensional orthogonal transform encoding, the encoded data of the AC component subjected to Huffman encoding is limited to output at a predetermined data length, that is, so-called fixed length encoding is performed. It is desirable that this fixed length is performed using an optimal amount of data depending on the required image quality, capacity of the recording medium, etc.

However, in conventional devices, the amount of data for fixed length conversion is uniformly set using a specific method, and fixed length conversion is performed thereby, making it impossible to perform appropriate fixed length conversion. This could cause the output buffer to overflow.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the problems of the prior art and to provide an image signal compression encoding device that can appropriately convert encoded data into a fixed length.

According to the present invention, an image signal compression encoding device that divides digital image data constituting one screen into a plurality of blocks and performs two-dimensional orthogonal transform encoding on the image data of each block includes a plurality of image signal compression encoding devices. orthogonal transformation means for two-dimensional orthogonal transformation of digital image data divided into blocks; encoding means for encoding data orthogonally transformed by the orthogonal transformation means; encoded data for each block output from the encoding means The encoding output control means has an encoding output control means for limiting the amount of normalization coefficients, and a normalization coefficient setting means for setting the normalization coefficients used in the encoding means. When input, the amount of encoded data is limited by a predetermined value,
If no normalization coefficient is input from the normalization coefficient setting means, the amount of encoded data is limited based on the activity of image data for each divided block.

DESCRIPTION OF EMBODIMENTS Next, embodiments of an image signal compression encoding apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an embodiment of an image signal compression encoding device according to the present invention.

The device has a memory controller lO. A memory 60 is connected to one memory controller, and one frame of still image data captured by, for example, an electronic still camera is read from the memory 60.

The output from the memory controller 10 is sent to the blocking unit l2.
is input. The blocking unit 12 is composed of a frame buffer, and receives input from the memory controller 10.
The image data stored in the blocking section 12 is divided into a plurality of blocks, read out block by block, and sent to the two-dimensional orthogonal transformation section l4. The two-dimensional orthogonal transform unit l4 performs two-dimensional orthogonal transform on the image data for each block. As the two-dimensional orthogonal transformation, well-known orthogonal transformations such as discrete cosine transformation and Hadamard transformation are used. The image data for each block that has been subjected to two-dimensional orthogonal transformation in the two-dimensional orthogonal transformation unit l4 is arranged vertically and horizontally, similar to the pixel data for each block shown in FIG. 3, with low-order data arranged in the upper left part. , the data becomes higher-order as it moves toward the lower right.

DC component data is placed at the upper left. The output of the two-dimensional orthogonal transform unit 14 is normalized by a predetermined quantization step value, that is, a normalization coefficient α, in a normalization unit (not shown), and is sent to the AC encoding unit 32 of the encoding unit 28.

The AC encoding unit 32 also receives the normalization coefficient α from the α determination unit l6. The α determination unit l6 determines whether the normalization coefficient α input from the α setting unit l8 is O. If the normalization coefficient α is not O,
and sends it to the bit fixed length conversion unit 36. α setting section l8
sets the normalization coefficient α. The normalization coefficient α is data used in the fixed length conversion unit 35 to convert the output data from the AC encoding unit 32 into a fixed length. The normalization coefficient α is set by predicting the optimal value using a predetermined method such as Newton's method based on the amount of output encoded data, and setting the normalization coefficient α so that the amount of output encoded data converges on the target value. be exposed.

Therefore, the normalization coefficient α is modified by determining whether the amount of encoded data is close to the target value for fixed length conversion, as described later. The initial value of the normalization coefficient α may be a constant value, or may be set to an approximate value each time.

The AC encoding unit 32 further receives a normalization coefficient a from the a calculation unit 44 . The a calculation unit 44 sets the normalization coefficient a based on the total activity calculated by the total activity calculation unit 26. Blocking department! The image data for each block output from the block activity calculation section 20 is also sent to the block activity calculation section 20. The block activity calculation unit 20 calculates the activity of each block, that is, the extent to which image data of the Takagi frequency component is included in the block. The activity of a block is determined by adding the absolute value of the difference between each pixel data constituting the block and the average value of these pixel data.

The activity for each block output from the block activity calculation unit 20 is output to the total activity calculation unit 26 and the bit allocation calculation unit 40.

The total activity calculation section 26 adds up the activities for each block sent from the block activity calculation section 20, and calculates the total value of the activities. The total activity calculation unit 26 outputs the total activity value to the bit allocation calculation unit 40.

As shown in FIG. 4, the AC encoding unit 32 normalizes the alternating current component of the transform coefficient inputted by scanning in a zigzag pattern using the normalization coefficient t'l a sent from the α calculation unit 44, and encodes it. ．． Since the alternating current component of the conversion coefficient often has continuous zeros, the continuous amount of zero value data, that is, the zero run length, is detected, the zero run length and non-zero amplitude are obtained, and this is converted into a two-dimensional Huffman encode. AC encoding section 32
The output from is sent to the fixed length converter 36. On the other hand, the DC component of the image data for each block output from the blocking unit l2 is sent to the DC encoding unit 30 of the encoding unit 28, and is Huffman encoded in the DC encoding unit 30. The DC encoded data output from the DC encoder 30 is sent to the output buffer 38.

The bit allocation calculation unit 40 calculates the activity for each block sent from the block activity calculation unit 20,
Using the total activity value sent from the total activity calculation unit 26, the encoded bits to be allocated to each block are calculated. The encoded bits allocated to each block are two-dimensional Huffman encoded for each block, and determine at what point the data output from the low-frequency component should be cut off, in other words, how many bits of encoded data should be output. This is the specified number of bits. The encoded bits allocated by the bit allocation calculation unit 40 are calculated by the normalization coefficient α from the α size foot unit 16.
is used for fixed length conversion in the fixed length conversion section 36 when the input is not input.

The coded bits allocated to each block are given by the following formula, for example. bit B1= ((act b/act tlx
bit[flagl }+ { Carry Over
} =-fil In the above formula, bit[flagl represents the total number of bits of the entire image, and is the number of bits allocated to the entire image composed of a plurality of blocks in encoding. In other words, when two-dimensional Huffman encoded data for each block is output in order from the low frequency component, the bits indicate how many bits will be allocated to the entire image if the data output is stopped at a predetermined number of bits. It is a number. This number of bits determines the amount of data that is compressed and encoded and output.

act b is the activity for each block.
act t represents the total activity and is the sum of the activities of the block. Therefore, the above equation (
fact b/act tl x bit[flag
] } distributes the number of bits allocated to the entire image according to the ratio of the activity of each block to the total activity of the block. {Carry Over} represents a carryover bit from the previous block, and as described later, the bit carried over from the block immediately before the block for which the allocated encoded bit bit Bl is to be obtained, that is, the bit carried over from the previous block. This is the number of bits that were allocated to the block but were not used. Note that the carryover bit {Ca
Bit allocation may be performed without considering }.

The old coded bits to be allocated to each block, which are output from the bit allocation calculating section 40, are sent to the fixed length converting section 36. If the normalization coefficient a is input, the fixed length section 3
6 is not fixed length based on activity. However, to prevent the output buffer 38 from overflowing,
The bit allocation (bitB) of each block is all fixed to the maximum value. Therefore, the AC encoded data is output to the output buffer 38 within a range that does not exceed the maximum value.

The output buffer 38 receives AC encoded data for each block whose number of bits is limited by the fixed length converter 36 . The output buffer 38 stores DC encoded data with a fixed code length and AC encoded data with a limited number of bits, and these data are sent to the gate 42.

The gate 42 temporarily stores the data sent from the output buffer 38, and sends this data to the memory card 50 through the connector 44 in response to a gate enable signal from the control section 22. The data is thereby recorded on the memory card 50. In addition to the memory card 50, a magnetic disk, an optical disk, or other recording medium may be used as the recording medium.

Data output from output buffer 38 is also sent to symbol counter 24. The code counter 24 counts the number of codes sent from the output buffer 38. The output of the code counter 24 is sent to the control section 22. The control section 22 is a control section that controls the operation of each section of the apparatus through signal lines (not shown). The α setting unit 18 calculates the normalization coefficient α using a predetermined method as described above. The control unit 22 requests the α setting unit l8 to create a normalization coefficient α. The control unit 22 also controls the fixed length converting unit 36 according to the determination from the α determining unit l6.

The operation of this device will be explained using the flow shown in Figure 2. First, the DC code amount, that is, the number of bits allocated to DC encoded data is calculated (10) α setting unit l8
The normalization coefficient α is manually input to the control unit 22 from
transfers the normalization coefficient α to the α determination unit l6 of the encoder 28+102) The α determination unit l6 determines whether the normalization coefficient t'l a input from the α setting unit l8 is O. +1041. When the normalization coefficient α is O, that is, when the normalization coefficient a is not sent from the α setting unit l8, the DC component is encoded by the DC encoding unit 30 (1301. Next, the encoding A block pit allocation is created in the bit allocation calculation unit 40 of the device 28 (1321
In the α calculation unit 44, a normalization coefficient a is created according to the data from the total activity calculation unit, and based on this, the AC component is encoded in the AC encoding unit 32 (134
3. The encoded AC component data is converted into a fixed length by the fixed length conversion unit 36 by block bit allocation from the bit allocation calculation unit 40 (1361. It is determined whether or not encoding has been completed for all blocks, f1381, If the encoding has not been completed, the process returns to step 130. This process ends when all blocks have been encoded. If it is determined in step 104 that the normalization coefficient α input from the a setting unit 18 is not O, That is, when the normalization coefficient α is sent from the α setting unit l8, α
Normalization is performed using the normalization coefficient α input from the setting unit l8. In this case, the DC component of the block sent from the blocking unit l2 to the DC encoding unit 30 is encoded (step 1101) and sent to the output buffer 38. Next, the DC component of the block sent from the blocking unit l2 to the DC encoding unit 30 is encoded (step 1101). The AC component of the coefficient is A
It is sent to the C encoding unit 32 for encoding, and is sent to the α setting unit l.
Encoding is performed using the normalization coefficient α sent from 8 to the AC encoding unit 32 through the α determination unit l6. At this time, the bit allocation value in the fixed length converting unit 36 is fixed to a predetermined constant value regardless of the activity value (C step 112), and the fixed length data is accumulated in the output buffer 38, The amount of data is counted by the code counter 24. Determine whether DC component encoding and AC component encoding have been completed for all blocks (step 114);
If it has not finished, return to step 110 and
Repeat component encoding and AC component encoding. D.C.
When the encoding of the components and the encoding of the AC components are completed for all blocks, the amount of data counted by the code counter 24 is sent to the control unit 22 (step 116), and the control unit 22 determines that the amount of code is the target of fixed length. (Step 11U. This target value is set in advance as a predetermined number of bits as the code amount to be allocated to one encoded image data. If the code amount exceeds the target value If the value is not close to the target value, the α setting section 18 newly sets the normalization coefficient α.
(Step 1201. Then, Step 10
Returning to step 2, the normalization coefficient α is input again from the α setting unit 18 and transferred to the α determination unit 16 (102) If the code amount from the code counter 24 does not exceed the target value and is sufficiently close to the target value. In the case of 11221 . 1124 sending the encoded data stored in the output puffer 38 to the memory card 50 through the connector 44;
1 The encoded data is thereby recorded on the memory card 50. According to the present embodiment, it is determined whether the normalization coefficient α is input to the encoder 28 from the α setting unit l8. If it is input, the amount of coded data is limited by this, and it is determined that the amount of code obtained is close to the target value and output from the gate. If the obtained code amount exceeds the target value or is not close to the target value, a more suitable α is set in the α setting unit l8 and sent to the α determining unit l6. Therefore, the optimal normalization factor α
Since it is possible to obtain a fixed length, it is possible to perform appropriate compression according to the required image quality, capacity of the memory card 50, etc. Since the normalization coefficient α set as the optimum value in this way is set outside the encoder 28, it can also be used for compression encoding of other image data. Furthermore, when the normalization coefficient α is input from the α setting unit l8, fixed length conversion is performed using the encoded bits calculated in the bit allocation calculation unit 40 inside the encoder, so the output buffer 38 will not overflow. In the above embodiment, the encoded data output from the fixed length converting section 36 is multiplied by M into the output buffer 38, and the gate 4 is multiplied by the gate enable signal from the control section 22.
Although the output buffer 38 is not provided in the encoder 28, an output buffer may be provided in the gate 42. In this case, the control unit 22 determines the code amount of the coded data counted by the code counter 24, and if it is close to the target value, each block is coded using the normalization coefficient α at that time. , just output the encoded data to the output buffer. According to the present invention, the compression encoding device sets a normalization coefficient α outside the encoder, and uses this to perform fixed length encoding. Since the normalization coefficient α is set outside the encoder, the normalization coefficient α can be used to compress multiple image data. Moreover, since the upper limit of the data to be encoded and output is determined, the output buffer will not overflow.

Further, even if the normalization coefficient α is not supplied from outside the encoder, it is possible to perform fixed length conversion based on the activity of the block.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the image signal compression encoding device according to the present invention, FIG. 2 is a flowchart showing the operation of the device shown in FIG. 1, and FIG. 3 is pixel data for each block. FIG. 4 is a diagram showing the coding order of AC components of transform coefficients. Main part. Explanation 14. ．． ．． Two-dimensional orthogonal transformation unit 1G. ．． ．． α determination unit 18. ．． ．． α setting section 20. ．． Block activity calculation unit 24. ．． Sign counter 26. ．． ．． Total activity calculation unit 28. ．． ．．
Encoding unit 30. ．． ．． DC encoding unit 32. ．． ．． AC encoding unit 36. ．． ．． Fixed length section output buffer pit allocation calculation section, gate α calculation section memory card

Claims (1)

[Claims] 1. An image signal compression encoding device that divides digital image data constituting one screen into a plurality of blocks and performs two-dimensional orthogonal transform encoding on the image data of each block, the device comprising: , orthogonal transformation means for two-dimensional orthogonal transformation of the digital image data divided into the plurality of blocks; encoding means for encoding the data orthogonally transformed by the orthogonal transformation means; and output from the encoding means. encoding output control means for limiting the amount of encoded data for each block; and normalization coefficient setting means for setting a normalization coefficient used in the encoding means; When the normalization coefficient is input from the normalization coefficient setting means, the amount of encoded data is limited by a predetermined value, and when the normalization coefficient is not input from the normalization coefficient setting means. An image signal compression encoding device characterized in that the amount of encoded data is limited based on the activity of image data for each divided block. 2. The apparatus according to claim 1, further comprising: block activity calculation means for calculating the activity of the image data for each divided block; and an activity for each block calculated by the block activity calculation means. encoded data amount allocation means for calculating the amount of encoded data to be allocated to each block based on the encoded data amount allocation means; . An image signal compression encoding device, characterized in that the amount of encoded data is limited in accordance with the output from the encoded data amount distribution means.