JP4094832B2 - Image data compression apparatus and control method thereof - Google Patents

Description

[0001]
【Technical field】
The present invention relates to an image data compression apparatus for obtaining encoded data by calculating orthogonal transform coefficients, quantizing orthogonal transform coefficients using a quantization table, and encoding quantized orthogonal transform coefficients, and a control method therefor About.
[0002]
BACKGROUND OF THE INVENTION
There is a motion JPEG (joint photographic experts group) that realizes moving image display by continuously reproducing color still images. In motion JPEG, orthogonal transform coefficients are calculated, orthogonal transform coefficients are quantized using a quantization table, and quantized orthogonal transform coefficients are encoded. The encoded image data is recorded on a recording medium such as a memory card.
[0003]
Since the amount of code after compression needs to be a certain value, the compression processing is performed once for the image data for one frame in the next period after the subject is imaged to obtain the image data (first Encoding processing, pre-compression), and the code amount is calculated. A quantization table is calculated based on the calculated code amount. In the next cycle of the first encoding process, the image data is compressed again (second encoding process, main compression) using the calculated quantization table.
[0004]
However, the first encoding process may not end within one cycle. Then, since the timing of the second encoding process is shifted, it becomes impossible to record the image data encoded at a constant interval on the recording medium.
[0005]
DISCLOSURE OF THE INVENTION
An object of the present invention is to make the timing of the second encoding process constant.
[0006]
An image data compression apparatus according to the present invention calculates orthogonal transform coefficients for image data representing an image for one frame, quantizes the orthogonal transform coefficients using a quantization table, and encodes the quantized orthogonal transform coefficients. Encoding means for obtaining encoded data, and performing a first encoding process for encoding the given image data within a given period of time to obtain the first encoded data. A first encoding control unit that controls the encoding unit; and the first encoding process of the encoding unit based on the control of the first encoding control unit does not end within the predetermined period. The stop control means for controlling the first encoding control means so as to stop the first encoding process, obtained by the time when the first encoding process is stopped by the stop control means Based on code amount , Using the quantization table calculating means for calculating the quantization table, and the quantization table calculated by the quantization table calculating means, a second encoding process is performed on the given image data, The second encoding control means for controlling the encoding means so as to obtain the encoded data is provided.
[0007]
The present invention also provides a control method suitable for the image data compression apparatus. In other words, this method calculates the orthogonal transform coefficient, quantizes the orthogonal transform coefficient using the quantization table, and encodes the quantized orthogonal transform coefficient to the image data representing the image for one frame. In an image data compression apparatus having an encoding means for obtaining encoded data, a first encoding process for encoding the given image data is performed within a certain period after the image data is given, The encoding means is controlled to obtain encoded data, and the first encoding process is stopped when the first encoding process of the encoding means does not end within the predetermined period. The quantization table is calculated based on the code amount obtained by controlling the first encoding control means and the first encoding process being stopped. Using the above, Performing a second coding process to the image data, and controls the encoding means so as to obtain a second encoded data.
[0008]
According to the present invention, the first encoding process is executed within a certain period after image data is given. If the first encoding process is completed within the predetermined period, first encoded data is obtained. A quantization table is calculated based on the obtained first encoded data. Using the calculated quantization table, the second encoding process is performed on the given image data. If the first encoding process is not completed within this fixed period, the first encoding process is stopped when the fixed period ends. A quantization table is calculated based on the code amount (code data) obtained when the operation is stopped. The second encoding process is executed using the calculated quantization table.
[0009]
If the first encoding process is not completed within a certain period, the first encoding process is stopped, so that the timing of the second encoding process does not shift. Even when the first encoding process is stopped, the quantization table can be calculated based on the encoded data up to the time point when the first encoding process is stopped. The value is almost constant.
[0010]
When image data representing an image for one frame is periodically given to the encoding means (for example, moving image data given with a period of 1/30 seconds or 1/60 seconds), the first encoding control means is an image. The encoding means is controlled so that the first encoding process is performed in the next period when data is given. The stop control means controls the first encoding control means to stop the first encoding process when the first encoding process does not end during the next period.
[0011]
In this case, control means for controlling the image data given after the second encoding process in the encoding means to repeat the first encoding process and the second encoding process by the encoding means. Is further provided.
[0012]
Usually, it will further comprise recording control means for recording the second encoded data obtained by the second encoding process based on the second encoding control means on the recording medium.
[0013]
[Explanation of Examples]
FIG. 1 shows an embodiment of the present invention and is a block diagram showing an electrical configuration of a digital camera.
[0014]
A digital camera is capable of movie recording and still recording. However, a digital camera capable of only movie recording may be used.
[0015]
The entire operation of the digital camera is controlled by the CPU 2.
[0016]
The digital camera includes a switch 1 including a mode setting switch for setting various modes such as a movie shooting mode and a still shooting mode, and a shutter release button. Various signals output from the switch 1 are input to the CPU 2. The CPU 2 is connected to a cache memory 3 that temporarily stores data.
[0017]
Further, the digital camera includes a direct memory access controller (DMAC) 5, and the DMAC 5 writes image data obtained by imaging to the memory 6 and reads image data written to the memory 6. Is controlled. Furthermore, the digital camera also includes a static random access memory (SRAM) 4.
[0018]
In the memory 6 described above, four areas (image data storage areas 6A to 6D for display) for storing image data representing an image for one frame, an area 6E for storing a quantization table used in quantization described later, A moving image data storage area 6F for recording that temporarily stores moving image data (compressed and encoded as will be described later) to be recorded on the memory card 16 is included.
[0019]
When the movie shooting mode is set by the switch 1, the subject is picked up by the CCD 8 at a period of 1/30 second, and video signals representing the subject image are sequentially output. The video signal is converted into digital image data by the analog / digital conversion circuit 9 and input to the reading control (signal processing) circuit 10. In the reading control circuit 10, predetermined signal processing such as gamma correction and white balance adjustment is also performed.
[0020]
The digital image data is given to the memory 6 through the bus control circuit 7 under the control of the DMAC 5. The display image data recording area 6A, 6B, 6C or 6D in the memory 6 is sequentially overwritten one frame at a time.
[0021]
The image data written in the display image data storage areas 6A to 6D are sequentially read one frame at a time and supplied to the digital / analog conversion circuit 13 via the bus control circuit 7 and the display control circuit 12. The digital image data is converted into an analog video signal by the digital / analog conversion circuit 13 and given to the display device 14. The captured subject image is displayed on the display screen of the display device 14.
[0022]
If the shutter release button in the switch 1 is pressed while the movie shooting mode is set, the image data obtained by imaging while the shutter release button is pressed is compressed and stored in the memory. Recorded on card 16. Specifically, the following operation is performed in the digital camera.
[0023]
The image data written in the display image data storage areas 6A to 6D are sequentially read out frame by frame and input to the compression encoding circuit 11 via the bus control circuit 7.
[0024]
FIG. 2 shows details of the compression encoding circuit 11.
[0025]
The image data read from the memory 6 is supplied to a DCT (discret cosine transform) circuit 22 via a buffer memory 21, and a DCT coefficient (orthogonal transform coefficient) is calculated for each block of 8 pixels × 8 pixels. . The calculated DCT coefficient is given to the quantization circuit 23 and quantized. Details of the processing in the quantization circuit 23 will be described later. The quantized DCT coefficient is encoded by the encoding circuit 24. The encoded image data (compressed image data) is output from the compression encoding circuit 11 via the buffer memory 25.
[0026]
Referring again to FIG. 1, the encoded image data (for a plurality of frames) output from the compression encoding circuit 11 is given to the memory 6 via the bus control circuit 7, and stored in the recording moving image data storage area 6F. Temporarily stored. The encoded image data is read from the moving image data storage area 6F for recording and given to the media control circuit 15 via the bus control circuit 7. It is recorded on the memory card 16 by the media control circuit 15.
[0027]
3 and 4 are time charts showing compression processing in the digital camera, and FIG. 5 is a flowchart showing movie recording processing procedures.
[0028]
When the movie shooting mode is set, as described above, the subject is imaged by the CCD 8 at a 1/30 second period (a read signal VI is output from the CPU 2 and given to the CCD 8 at a 1/30 second period), and the subject image is displayed. Representing image data is read. For example, referring to FIG. 3, image data representing an image A11 for one frame is obtained between times t11 and t12. The obtained image data is given to the display memory 6A and stored as described above. At time t12, image data representing an odd field image (A21) is read from the display memory 6A. As described above, the digital / analog conversion circuit 13 converts the signal into an analog video signal via the bus control circuit 7 and the display control circuit 12 and applies the analog video signal to the display device 13. At time t13, image data representing an odd field image (A22) is read from the display memory 6A. As described above, the digital / analog conversion circuit 13 converts the signal into an analog video signal via the bus control circuit 7 and the display control circuit 12 and applies the analog video signal to the display device 13. The subject image is displayed with the period (1/60 second) of the vertical periodic signal VD.
[0029]
Further, in the digital camera according to this embodiment, the above-described compression processing is performed twice for image data representing an image of one frame. A quantization table to be used in the quantization circuit 23 is calculated from the code amount obtained in the first compression process (pre-compression). In the second compression process (main compression), the compression process is performed again using the quantization table calculated in the first compression process. By performing the compression process twice in this way, the code amount of the compressed image data becomes substantially constant.
[0030]
In particular, in the digital camera according to this embodiment, as will be described in detail later, the first compression processing is not completed within the reading period of image data (within 1/30 second), and the reading signal VI is When input, the first compression process is interrupted. The quantization table is calculated based on the code amount obtained up to the time when it was interrupted.
[0031]
As described above, when reading of image data for one frame is completed between times t11 and t12 (YES in step 31), the first compression process is performed in the compression encoding circuit 11 from time t12 (step 32). ). In the first compression, the DCT coefficient is calculated by the DCT circuit 22 and the quantization circuit 23 performs the quantization and encoding in the encoding circuit 24 as described with reference to FIG. The quantization in the first compression process is performed using the quantization table 1 obtained from Equation 1. The initial quantization table is a quantization table stored in advance in the quantization table storage area 6E of the memory 6. The initial coefficient is a predetermined coefficient.
[0032]
Quantization table 1 = initial quantization table × initial coefficient.
[0033]
The quantization table 2 used for quantization in the second compression process is calculated from Equation 2.
[0034]
Quantization table 2 = initial quantization table × normalization coefficient Equation 2
[0035]
However, the normalization coefficient is obtained from Equation 3.
[0036]
Normalization coefficient = initial coefficient × (target code amount / SIZE) ^{−1 /} β (formula 3)
[0037]
However, SIZE is the code amount (referred to as SIZE1) obtained in the first compression process when the first compression process is completed within the reading period, and the second compression process is performed within the reading period. If not completed, the estimated code amount (referred to as SIZE2) obtained from Equation 4 is obtained.
[0038]
SIZE2 = (the amount of image data for one frame / the amount of image data encoded up to the point of interruption) × the amount of encoding obtained up to the point of interruption
[0039]
If the first compression process is completed within the reading period from time t12 to t13 (YES in step 33), the amount of image data encoded for one frame obtained in the first compression process is It is temporarily stored (step 35). A quantization table used for the second compression process is calculated according to Equation 2 using the encoding amount obtained by the first compression process (step 38). A second compression process is executed between times t13 and t14 using the calculated quantization table (step 39). Steps 31 to 39 are repeated until the moving image recording is completed (step 40).
[0040]
In FIG. 4, image data representing an image (A41) for one frame is read from time t21 to t22. If the signal VI is interrupted at time t23 without completing the first compression process within the reading period from time t22 to t23 (NO in step 33, YES in step 34), the first compression process is interrupted. (Step 36). The estimated code amount SIZE2 described above is calculated according to Equation 4 using the encoding amount obtained up to the point of interruption (step 37). Expression 3 is calculated using the calculated estimated code amount SIZE2, and a normalization coefficient is calculated. By substituting the calculated normalization coefficient into Equation 2, a quantization table used for the second compression process is calculated (step 38).
[0041]
A second compression process is performed from time t23 using the calculated quantization table (step 39).
[0042]
When the first compression process does not end within the reading period, it is interrupted, so the second compression process is not delayed. Sufficient time can be given to the second compression processing.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an electrical configuration of a digital camera.
FIG. 2 is a block diagram showing details of a compression encoding circuit.
FIG. 3 is an operation time chart of the digital camera.
FIG. 4 is an operation time chart of the digital camera.
FIG. 5 is a flowchart showing a processing procedure of the digital camera.
[Explanation of symbols]
2 CPU
6 Memory 8 CCD
10 Reading control circuit
11 Compression coding circuit
15 Media control circuit
16 Memory card
22 DCT circuit
23 Quantization circuit
24 Coding circuit

Claims (4)

Encoding to obtain encoded data by calculating orthogonal transform coefficients, quantizing the above orthogonal transform coefficients using a quantization table, and encoding the quantized orthogonal transform coefficients to image data representing an image of one frame means,
A first encoding process is performed to control the encoding means so as to obtain first encoded data by performing a first encoding process for encoding the supplied image data within a certain period after the image data is provided. Encoding control means,
The first encoding process is stopped so that the first encoding process of the encoding means based on the control of the first encoding control means is stopped when it does not end within the predetermined period. Stop control means for controlling the encoding control means of
Based on the code amount obtained until the first encoding process is stopped by the stop control means, the quantization table calculating means for calculating the quantization table, and the quantization table calculating means Second encoding control means for performing second encoding processing on the given image data using the calculated quantization table and controlling the encoding means so as to obtain second encoded data ,
An image data compression apparatus. Image data representing an image for one frame is periodically given to the coding means, and the first coding control means performs the first coding processing in the next period when the image data is given. The stop control means is configured to stop the first encoding process when the first encoding process does not end during the next period. Controlling the first encoding control means;
Control means for controlling the image data given after the second encoding process in the encoding means to repeat the first encoding process and the second encoding process by the encoding means;
The image data compression apparatus according to claim 1, further comprising: 2. The image data compression apparatus according to claim 1, further comprising recording control means for recording the second encoded data obtained by the second encoding processing based on the second encoding control means on a recording medium. Encoding to obtain encoded data by calculating orthogonal transform coefficients, quantizing the above orthogonal transform coefficients using a quantization table, and encoding the quantized orthogonal transform coefficients to image data representing an image of one frame In an image data compression apparatus comprising means,
Performing a first encoding process for encoding the given image data within a certain period after the image data is given, and controlling the encoding means so as to obtain the first coded data;
Controlling the first encoding control means to stop the first encoding process when the first encoding process of the encoding means does not end within the predetermined period; The above quantization table is calculated based on the amount of code obtained until the encoding process is stopped.
A second encoding process is performed on the given image data using the calculated quantization table, and the encoding means is controlled to obtain second encoded data;
A method for controlling an image data compression apparatus.

Images