JP3975811B2 - Digital still camera - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a digital still camera, and more particularly to a technique for shortening a shooting interval during continuous shooting while controlling the capacity of an image file to be generated.
[0002]
[Prior art]
FIG. 4 is a block diagram showing a configuration of a conventional digital still camera. Hereinafter, the configuration of a conventional digital still camera will be briefly described with reference to FIG. In FIG. 4, reference numeral 11 denotes an imaging circuit, which converts an input video into corresponding RGB three primary color data. Reference numeral 12 denotes a memory for storing the RGB data output from the imaging circuit, the later-described YC data, and JPEG data. Reference numeral 13 denotes a YC processing circuit which converts RGB data into YC data. A JPEG processing circuit 14 compresses YC data and converts it into an image file. A recording medium 15 stores the image file transferred from the memory 12. The image pickup circuit 11 generally outputs RGB data divided into A field and B field twice. Since the RGB primary colors are not aligned in the output of one field, the RGB data is temporarily stored in the memory 12, and the YC processing circuit 13 reads the RGB data from the memory 12 and processes it. Since the JPEG processing circuit 14 performs processing in units of 16 pixels × 8 pixels or 16 pixels × 16 pixels called macroblocks, YC data is also temporarily placed in the memory 12 and read and processed in the order of macroblocks. An image file generated by JPEG processing is also temporarily placed in the memory 12 before being stored in the recording medium.
[0003]
In order to operate the digital still camera with a limited memory capacity, a method of overwriting YC data on RGB data and recording is used. FIG. 5 is a graph for explaining an example of use of a memory in a conventional digital still camera. In FIG. 5, 31 is a horizontal axis indicating time, 32 is a vertical axis indicating an address, 33 and 34 are lower and upper limits of storage areas allocated to image processing, 35 is a locus of RGB data write addresses, and 36 is RGB The locus of the read address of data, 37 is the locus of the write address of YC data, 38 is the locus of the read address of YC data, and 39 is the point in time when writing of the second RGB data starts. The reason why the locus 35 of the write address of the RGB data passes through the same region twice is that the even-numbered line is written in the first field and the odd-numbered line is written in the second field. The reason why the slope of the RGB data read address locus 36 is smaller than the RGB data write address locus 35 is that the writing of RGB data is performed by skipping one line, while the YC generation is processed line by line. This is because the address speed is halved. The YC data write address locus 37 has the same inclination as the RGB data read address locus 36 because the YC data is overwritten and recorded on the RGB data of the corresponding line. Since JPEG processing reads and converts YC data, there is no problem even if RGB data is overwritten and erased. Compared with the method of storing the RGB data and YC data in different areas, the conventional method can increase the number of continuous shots if the storage capacity is the same because the capacity for storing image files is large. If the number of continuous shots is the same, the same function can be realized with a smaller storage capacity, which is advantageous in terms of cost.
Since JPEG processing reads YC data in units of macroblocks, the locus of addresses is not a simple straight line, but it should be understood that the locus 38 of YC data read addresses represents the center line of the address locus. As shown in FIG. 4, since the YC processing circuit 13 and the JPEG processing circuit 14 are independent, as shown by the address locus in FIG. 5, the processing is performed in parallel within the range where the read and write addresses do not intersect. I can do it.
[0004]
Since JPEG processing includes variable length coding, the capacity of an image file generated varies depending on the content of the image. However, since the capacity of the recording medium for storing the image file is fixed and it is expected that a fixed number of sheets can be recorded on the recording medium, a mechanism for controlling the capacity of the image file is required. Therefore, an appropriate value is given to a parameter called scale factor that affects the compression ratio, the first JPEG processing is performed, the scale factor is corrected according to the capacity of the generated image file, and then the second JPEG processing is performed. In general, a two-pass processing method of obtaining a final image file is used. In FIG. 5, the reason why the locus 38 of the read address of the YC data passes through the same region twice is that the JPEG processing is performed twice.
[0005]
[Problems to be solved by the invention]
In the case where continuous shooting is performed continuously in a short time, the next RGB data can be started in FIG. This is because reading the YC data for the second JPEG process must precede the writing of the second RGB data in order to place the YC data in the same area as the RGB data. If YC data is placed in a different area from the RGB data, writing of the second RGB data can be started without waiting for the JPEG processing twice, but it is disadvantageous in terms of cost because the memory capacity needs to be increased as described above. Become.
[0006]
Japanese Patent Laid-Open No. 10-150633 proposes a method of shortening the processing time by thinning out macroblocks in the first JPEG processing. However, this method is effective for digital still cameras that execute YC processing and JPEG processing in parallel, even if it is effective for old-generation digital still cameras that start the first JPEG processing after the completion of one-screen YC processing. There is no. This is because the JPEG processing is performed after the YC processing waits for YC data to be generated as indicated by the YC data write address trajectory 37 and the YC data read address trajectory 38 in FIG. This is because the end of the first JPEG process is immediately after the end of the YC process for one screen regardless of whether or not.
[0007]
Thus, there is a problem in terms of cost if YC data and RGB data are placed in different areas, and if they are placed in the same area, there is a problem that the continuous shooting interval becomes longer by the amount of JPEG processing performed twice.
[0008]
An object of the present invention is to solve the above problems, and to propose a method capable of shortening the continuous shooting interval and appropriately controlling the image file capacity without increasing the memory capacity.
[0009]
[Means for Solving the Problems]
In order to solve the above-described problems, the present invention provides an imaging unit that converts an input video into input image data, a YC processing unit that processes the input image data to generate YC data including luminance and color difference information, An image compression means for converting the YC data to obtain an image file whose capacity is compressed, wherein the YC processing means divides the input image data into a plurality of small images for processing. The YC processing means processes a group of small images ahead of other small images, and the image compression means converts the YC data generated in advance to perform a first compression process, A capacity of an image file obtained by processing the entire input image data is estimated from a capacity of the image file generated by the first compression processing, and the image is determined according to the capacity of the image file. Adjust the condensation means, characterized in that performing a second compression process of converting the input image overall YC data thereon.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
The invention according to claim 1 of the present invention is an imaging means for converting an input video into input image data, and a YC processing means for processing the input image data to generate YC data comprising luminance and color difference information. An image compression means for converting the YC data to obtain an image file whose capacity is compressed, wherein the YC processing means divides the input image data into a plurality of small images for processing. The YC processing means processes a group of small images in advance of other small images, and the image compression means converts the YC data generated in advance to perform a first compression process, A digital still camera that estimates the capacity of an image file obtained by processing the entire input image data from the capacity of the image file generated by the first compression processing. The time required for estimating the capacity of the JPEG file by performing the first JPEG process for estimating the capacity of the JPEG file and the YC process necessary as a premise thereof on the small image obtained by dividing the input image and thinning it out Can be shortened.
[0011]
According to a second aspect of the present invention, in the digital still camera according to the first aspect, a first storage unit that stores the input image data and a second storage unit that stores the YC data. And the first storage means reads out the input image data to be stored by thinning out the rectangular small image as a unit, and the second storage means is delayed from the YC data of the rectangular area generated in advance. YC data of the entire input image consisting of YC data of the generated rectangular area is stored, and the image compression means is adjusted according to the capacity of the image file selected by the first compression processing, and then the A digital still camera which performs a second compression process for converting YC data of the entire input image stored in the second storage means, and is generated for the first JPEG process JPEG file capacity for final recording with high accuracy by adjusting the image compression means according to the file capacity generated in the first JPEG process and performing the second JPEG process without wasting the YC data. Can be controlled.
[0012]
According to a third aspect of the present invention, in the digital still camera according to the second aspect, the second storage means also serves as the first storage means, and is adjusted in advance to a capacity of YC data. By disposing the input image data in a distributed manner, the YC data is stored in the same storage area as the input image data, and the digital still camera is characterized in that the input image data storage area Since it is not necessary to provide a YC data storage area independently, the capacity of the memory to be used can be reduced and manufacturing can be performed at low cost.
[0013]
According to a fourth aspect of the present invention, there is provided the digital still camera according to the first, second or third aspect, wherein the small image is obtained by dividing an input image horizontally. In addition, by processing a horizontally long small image as a unit, recording of the input image in the memory and reading for YC processing are facilitated in parallel.
[0014]
The invention according to claim 5 of the present invention is the digital still camera according to claim 1, 2 or 3, wherein the small image is obtained by dividing an input image vertically and horizontally. Thus, by limiting the width of the small image, the storage capacity of the line memory required by the YC processing circuit can be reduced.
[0015]
The invention according to claim 6 of the present invention is the digital still camera according to claim 5, wherein the small images to be processed in advance are thinned out so as to form a checkered pattern, This has the effect that the JPEG file capacity can be estimated with higher accuracy by distributing the input image area to be estimated for the JPEG file capacity more uniformly.
[0016]
A seventh aspect of the present invention is the digital still camera according to the second, third, fourth, fifth, or sixth aspect, wherein the second storage means has two modes, and the first mode Then, when performing the first compression processing, the second storage means reads out the input image data by thinning out a rectangular small image as a unit, and in the second mode, when performing the first compression processing. The second storage means reads input image data without thinning out, and uses the first mode for continuous shooting and uses the second mode for normal shooting. If it is a camera and the processing time is enough, the JPEG file capacity can be estimated with higher accuracy by not thinning out the area of the input image for which the JPEG file capacity is to be estimated. DOO can have the effect that.
[0017]
The invention according to claim 8 of the present invention is the digital still camera according to claim 7, wherein the second mode is also used when processing the last one of the continuously shot images. This is a digital still camera that uses the fact that there is no subsequent input image for the last one of the continuous shots, so that there is room for processing time, and the area of the input image that is the target of JPEG file capacity estimation By not subtracting, the JPEG file capacity can be estimated with higher accuracy.
The invention according to claim 9 of the present invention is the digital still camera according to claim 7 or 8, further comprising third storage means, wherein the third storage means is an image generated by the immediately preceding compression process. The file capacity information is stored, and in the processing of the second and subsequent images at the time of continuous shooting, the first compression process is not performed, and the image file immediately before being output by the third storage means is stored. A digital still camera that adjusts the image compression means using information on the capacity and then performs the second compression processing on the image compression means, and the capacity of the immediately preceding JPEG file in the second and subsequent shots This has the effect that the amount of signal processing can be reduced to reduce the power consumption, instead of the estimation result of the JPEG file capacity.
[0018]
(Embodiment 1)
Hereinafter, embodiments of the present invention will be described in detail with reference to FIG. 1, FIG. 2, and FIG. FIG. 1 is a block diagram illustrating the configuration of a digital still camera according to the present invention. In FIG. 1, reference numeral 11 denotes an image pickup circuit as image pickup means, which converts an input video into corresponding RGB primary color data. Reference numeral 12 denotes a memory which stores the RGB data, YC data and JPEG data described later output from the imaging circuit 11. The memory 12 includes a first storage unit that records RGB data output from the imaging circuit 11, and a second storage unit that stores YC data. Reference numeral 13 denotes a YC processing circuit as YC processing means, which converts RGB data into YC data. Reference numeral 14 denotes a JPEG processing circuit as image compression means, which compresses YC data and converts it into an image file. A recording medium 15 stores the image file transferred from the memory 12. Reference numeral 16 denotes a shutter, and the user presses the shutter 16 to instruct photographing. Reference numeral 17 denotes a control circuit that detects pressing of the shutter 16 and controls the photographing operation.
[0019]
(A), (B), (C), and (D) of FIG. 2 are memory maps showing the arrangement of image data storage areas in the embodiment of the present invention. In FIG. 2, 21 is an area allocated for storing RGB data and YC data, 22 is a storage area for RGB data, 23 is a storage area for RGB data of a small image group to be processed in advance, and 24 is a small area for processing with a delay. A storage area for RGB data of the image group, 25 is a storage area for YC data generated in advance, 26 is a storage area for YC data generated late, and (A) is assigned to storage of RGB data and YC data. A memory map showing the arrangement of the storage area 22 for RGB data in the area 21. (B) is a memory map showing the arrangement of the storage area 23 for RGB data of a small image group to be processed in advance. (C) is a memory map showing an arrangement of the RGB data storage area 24 and the YC data storage area 25 generated in advance, and (D) is generated in advance. 10 is a memory map showing an arrangement of a YC data storage area 25 and a YC data storage area 26 generated with a delay.
[0020]
FIG. 3 is a graph for explaining how to use the memory in the digital still camera of the present invention. In FIG. 3, 31 is a horizontal axis indicating time, 32 is a vertical axis indicating an address, 33 and 34 are lower and upper limits of a storage area allocated to image processing, 35 is a locus of RGB data write addresses, and 36 is The locus of RGB data read addresses, 37 is the locus of YC data write addresses, 38 is the locus of YC data read addresses, and 39 is the time when writing of the second RGB data starts.
[0021]
Hereinafter, embodiments of the present invention will be described step by step with reference to FIGS. 1, 2, and 3.
[0022]
In FIG. 1, when the user presses the shutter 16 to instruct shooting, the control circuit 17 detects the pressing and starts the shooting operation. First, the imaging circuit 11 outputs the RGB data first from the even lines, and the memory 12 stores this. At this time, the memory 12 stores each line with a distance of 1 line or more, and then stores the RGB data of odd lines with an interval of 1 line or more.
[0023]
FIG. 2A shows a memory map in a state where all the RGB data is stored in the memory 12, and stores an area not occupied by the RGB data 22 for each line with a large space. Thus, the 1st reason which memorize | stores with a clearance gap is for using the same area | region for memory | storage of YC data later. The RGB data output by the imaging circuit 11 is around 10 bits per pixel, whereas YC data has a total of 16 bits per pixel, so that one line of RGB data occupies the capacity occupied by one line of RGB data. If it is not set larger than the net capacity, YC data cannot be placed in the same area.
[0024]
Further, the second reason for storing the data with a gap is that the YC processing generates one line of YC data from three lines of RGB data. 2B is a storage area for RGB data of a small image group to be processed in advance, and 24 of FIG. 2C is a storage area for RGB data of a small image group to be processed later. Have overlapping parts.
[0025]
If the small image of RGB data is divided and processed so as not to overlap, the output small image generated by the YC processing is two lines less than the input small image, so the YC data of the boundary portion is equivalent to two lines. Will not be generated.
[0026]
In order to prevent this, it is necessary to divide the input small image so that it overlaps by two lines. As a result, as shown in FIG. 2C, the gap in the storage area of the RGB data of the small image group to be processed with delay. Is reduced by two lines. In YC processing, since the number of pixels in one line of the output image is smaller than the number of pixels in one line of the input image, the conversion is not simple, but the RGB is slightly larger than the storage capacity of YC data per line of subtraction. If the capacity of one line of data is secured, as shown in FIG. 2C, the storage area of YC data generated prior to the gap of the RGB data storage area 24 of the small image group to be processed with delay. 25 can be stored.
[0027]
FIG. 2D shows a state in which all RGB data is finally converted to YC data. There is a gap of 2 lines between small images of YC data. However, since JPEG processing is performed in units of macroblocks, the number of lines of small images is a multiple of 16, and the boundary of small images is the boundary of macroblocks. If it matches, JPEG processing will not be an obstacle. When the number of lines of the small image is 64, the ratio of the capacity of the area occupied by the YC data and the net capacity of the YC data is 66 ÷ 64≈1.03, and the overhead is slight.
[0028]
The operation so far will be described mainly from the movement of the address with reference to FIG.
[0029]
The imaging circuit 11 in FIG. 1 outputs RGB data, and the control circuit 17 controls the memory 21 to store the RGB data in the range from the lower limit 33 to the upper limit 34 of the storage area in FIG. Since the imaging circuit 11 divides the RGB data into two fields and outputs them in an interlaced order, the RGB data write address 35 passes through the range from the lower limit 33 to the upper limit 34 of the storage area once for each field.
[0030]
The control circuit 17 in FIG. 1 starts the YC processing circuit 13 in FIG. 1 with a slight delay from the start of writing the second field RGB data, and controls the memory 21 to read the first round RGB data and write the YC data. start. This operation is represented by the locus of the RGB data read address 36 and the YC data write address 37 in FIG. The control circuit 17 in FIG. 1 interrupts the processing every time 64 lines of YC data are written, and restarts by incrementing the read and write start addresses by 64 lines. In FIG. 3, this operation is represented as a jump between the read address 36 for RGB data and the write address 37 for YC data.
[0031]
Since the RGB data write address 35 advances by two lines every time one line is written, the gradient between the RGB data read address 36 and the YC data write address 37 corresponding to the YC processing for each line is equal to the RGB data write address 35. Half of the slope. However, the gradient of the read address 36 for RGB data and the write address 37 for YC data is the same as the gradient of the write address 35 for RGB data as a whole by performing an operation of jumping each time an input image is divided into small images. Become. The control circuit 17 in FIG. 1 starts the JPEG processing circuit 14 in accordance with the generation of YC data, and starts the first JPEG processing. Since the JPEG process also proceeds while jumping to the YC data generated at the jump address, as shown by the locus of the read address 38 of the YC data in FIG. At the time of delay, the result of JPEG processing is obtained by thinning the input screen in half.
[0032]
Since the capacity of the JPEG data output by the JPEG processing so far is the result of thinning the input image evenly, it can be considered that it is almost half the capacity of the JPEG data when the entire input image is processed. Therefore, the control circuit 17 in FIG. 1 evaluates the capacity of the JPEG data, changes the setting of the scale factor of the JPEG processing circuit 14, and starts the second JPEG processing.
[0033]
In parallel with this, the control circuit 17 in FIG. 1 controls the YC processing circuit 13 and the memory 21 to return the start position to the second small image and perform the YC processing and jump for the second small screen unit. Resume. Since the thinning is not performed in the second JPEG process, the gradient of the read address 38 of the YC data becomes gentler than the gradient of the first address, and does not catch up with the write address 37 of the YC data.
[0034]
In this way, the input image is divided into small images, thinned out, YC processed, and YC data of the small image is subjected to JPEG processing to estimate the JPEG data amount of the entire input image, thereby speeding up the start of the second JPEG processing. It is possible.
[0035]
The control circuit 17 in FIG. 1 monitors the shutter 16, and if the shutter 16 continues to be pressed, the process shifts to the second shooting at the shortest time interval. At this time, as shown in FIG. 3, it is necessary to select a timing at which the read address 38 of the YC data and the write 35 of the RGB data do not intersect, so the writing of the second RGB data is started from the point 39 in FIG. It will be. If the time required for YC processing without thinning out the entire input image is equivalent to 2 fields, the time shortened by performing thinning is equivalent to 1 field, so the effect of shortening the processing time is great.
[0036]
In the first JPEG processing, since the input image is thinned out, it is undeniable that the accuracy of estimating the JPEG file capacity is slightly lower than when the entire input image is JPEG processed. Therefore, as proposed in claim 7, the control circuit 17 of FIG. 1 has two processing modes, and if the shutter 16 is kept depressed at the start of writing the second field of RGB data, the continuous shooting mode is set. Then, the input image is thinned out in small image units to perform YC processing and JPEG processing. If the shutter 16 is not pressed at the time point, the camera enters single shooting mode and performs YC processing and JPEG processing without thinning out the input image. By doing so, it is possible to control the JPEG file capacity with high accuracy at the time of single shooting with a sufficient processing time.
[0037]
As for the continuous shooting mode performed on the YC data generated by thinning out the first JPEG process, as proposed in claim 6 of the present invention, the input image is divided vertically and horizontally, and the YC process and the JPEG process are performed in the first round. The accuracy of estimating the JPEG file capacity can be increased by performing thinning so that the small images to be checked form a checkerboard pattern, thereby sampling the input images more evenly.
[0038]
Further, as proposed in claim 9 of the present invention, by utilizing the fact that the input image does not change significantly from the immediately preceding image after the second continuous shooting, the first JPEG processing is omitted, and The scale factor can be adjusted based on the JPEG file size.
[0039]
Further, as proposed in claim 8 of the present invention, when the control circuit 17 in FIG. 1 detects the release of the shutter 16 during the continuous shooting process, or the predetermined number of continuous shots has been reached. In this case, it is determined that it is the last one in continuous shooting, the first JPEG processing is performed without thinning out, and the JPEG file capacity is estimated with high accuracy, and the total capacity of the JPEG file generated during continuous shooting is the target. It is also possible to control with high accuracy so as to approach the value.
[0040]
As described above, according to this embodiment, the low-cost configuration in which the YC data is arranged in the same area as the RGB data, the JPEG file capacity control is not deviated, and the continuous shooting interval is greatly shortened. It is possible to configure a digital still camera.
[0041]
【The invention's effect】
As described above, according to the present invention, the low-cost configuration in which the YC data is arranged in the same area as the RGB data, the JPEG file capacity control is not deviated, and the continuous shooting interval is greatly shortened. An excellent effect that a still camera can be configured is obtained.
[Brief description of the drawings]
FIG. 1 is a block diagram of an embodiment of the present invention.
FIG. 2 is a diagram showing a memory map showing an arrangement of storage areas for image data in the embodiment of the present invention.
FIG. 3 is a graph illustrating how to use a memory according to an embodiment of the present invention
FIG. 4 is a block diagram showing a configuration of a conventional digital still camera.
FIG. 5 is a graph illustrating an example of using a memory in a conventional digital still camera
[Explanation of symbols]
11 Imaging circuit
12 memory
13 YC processing circuit
14 JPEG processing circuit
15 Recording media
16 Shutter
17 Control circuit

Claims (9)

Imaging means for converting input video into input image data, YC processing means for processing the input image data to generate YC data consisting of luminance and color difference information, and converting the YC data to compress the capacity An image compression means for obtaining an image file, wherein the YC processing means divides the input image data into a plurality of small images for processing, and the YC processing means is a group of small images. The image is processed in advance of other small images, and the image compression means converts the YC data generated in advance to perform the first compression processing, and the image file generated by the first compression processing. A digital still camera, wherein the capacity of an image file obtained by processing the entire input image data is estimated from the capacity of the image. 2. The digital still camera according to claim 1, further comprising: a first storage unit that stores the input image data; and a second storage unit that stores the YC data. The first storage unit stores the first storage unit. Input image data is read out by thinning out a rectangular small image as a unit, and the second storage means includes input of YC data of a rectangular area generated in advance and YC data of a rectangular area generated later. YC data for the entire image is stored, the image compression means is adjusted according to the capacity of the image file selected by the first compression processing, and then the entire input image stored by the second storage means A digital still camera that performs a second compression process for converting the YC data. 3. The digital still camera according to claim 2, wherein the second storage means also serves as the first storage means, and the input image data is distributed and arranged in advance according to the capacity of YC data. Thus, the YC data is stored redundantly in the same storage area as the input image data. 4. The digital still camera according to claim 1, wherein the small image is obtained by dividing an input image into horizontal sections. 4. The digital still camera according to claim 1, wherein the small image is obtained by dividing an input image vertically and horizontally. 6. The digital still camera according to claim 5, wherein a small image group to be processed in advance is thinned out so as to form a checkered pattern. 7. The digital still camera according to claim 2, 3 or 4 or 5 or 6, wherein the second storage means has two modes, and the first compression process is performed when the first compression processing is performed in the first mode. The second storage means reads out the input image data by thinning out a rectangular small image as a unit. In the second mode, the second storage means reads the input image data when performing the first compression processing. A digital still camera, which reads out without thinning out, uses the first mode for continuous shooting and uses the second mode for normal shooting. 8. The digital still camera according to claim 7, wherein the second mode is also used when processing the last one of continuously shot images. The digital still camera according to claim 7 or 8, further comprising a third storage unit, wherein the third storage unit stores information on a capacity of an image file generated by the immediately preceding compression process, In the processing of the second and subsequent images at the time of continuous shooting, the first compression processing is not performed, and the image compression means is adjusted using the information on the capacity of the image file immediately before output from the third storage means. Then, the second compression processing is performed on the digital still camera.

Images