JP2004070226A - Image pickup device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To simplify the constitution for AF processing by using the result of DCT processing in small block units (8×8) for AF processing and to improve AF accuracy by appropriately evaluating a high frequency component in accordance with the frequency thereof. <P>SOLUTION: Image data are DCT-processed in a two-dimensional DCT part for every specified block, quantized according to a quantization modulus and outputted as compressed image data. In an image pickup device having such a data compressing part, evaluation processing decided corresponding to the frequency component of the data is performed to the converted image data obtained by the two-dimensional DCT part, the converted image data processed to be evaluated are integrated, and an AF control signal is formed on the basis of an integrated value. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an imaging apparatus such as a digital still camera, and more particularly to an imaging apparatus having an AF control unit for automatically focusing the camera (autofocus; AF).
[0002]
[Prior art]
In digital still cameras, etc., in order to automatically align the focal point, an active AF method using infrared rays or ultrasonic waves, or a passive AF method that does not require a special distance measuring component, is used as an AF method from an image sensor. An AF method is used in which a difference between adjacent pixels of an image input is obtained, and the difference is adjusted so that the difference value becomes maximum.
[0003]
As a passive AF method, as shown in JP-A-2000-165734, DCT (Discrete Cosine Transform) is performed on image data obtained by imaging at a lens position. The AF control is being performed based on the conversion result.
[0004]
In this conventional AF control using DCT, DCT processing is performed on image data obtained by imaging at an initial focus lens position, an average value of DCT coefficients of a converted image is obtained for each wave number, and the wave number is ascending. The wave numbers having the local minimum value are searched from the average value sequence, and the AF position is determined by one shot using the searched wave number as a defocus parameter. Further, the AF position is determined more finely by using a hill-climbing method based on a high-frequency component of the image with respect to the AF position.
[0005]
[Problems to be solved by the invention]
In the conventional AF control using DCT, it is necessary to perform DCT processing on a fairly large AF area (256 × 256 pixels) during AF in order to perform the above-described processing. In addition, since the high-frequency components of the data subjected to the DCT processing are simply integrated, the evaluation of the high-frequency components is not properly performed.
[0006]
Therefore, the present invention uses a signal from a DCT processing unit in a small block unit (8 × 8 pixels) used for data compression processing such as MPEG or JPEG on photographed image data for AF processing. It is an object of the present invention to provide an imaging apparatus having an AF control unit which simplifies the configuration for AF processing and improves the AF accuracy by appropriately evaluating a high-frequency component according to the frequency. I do.
[0007]
[Means for Solving the Problems]
An image pickup apparatus according to claim 1, wherein the lens system includes a lens and a driving unit thereof,
An image sensor that forms a subject image with light from the lens system,
A front end unit that outputs image data corresponding to a subject image from the image sensor,
The DCT unit performs DCT processing on the image data from the front end unit for each predetermined block to obtain converted image data divided for each frequency component, and the converted image data corresponds to the frequency division by the quantization unit. In the imaging device having a data compression unit that performs a quantization process according to the quantization coefficient that is set as the data and outputs the data as compressed image data,
Further, the image processing apparatus further includes an AF control unit that creates an AF control signal based on the converted image data obtained by the DCT unit and supplies the AF control signal to the lens system.
The AF control unit performs an evaluation process determined corresponding to the frequency component on the converted image data, integrates the converted image data subjected to the evaluation process, and generates the AF control signal based on the integrated value. Is formed.
[0008]
According to the imaging apparatus of the present invention, a converted image that is a DCT processing result in small block units (for example, for 8 × 8 pixels) performed for compression processing such as MPEG or JPEG on captured image data By using the data for the AF processing, the configuration for the AF processing can be simplified. Further, the AF accuracy can be improved by appropriately evaluating each converted high-frequency component based on the evaluation processing criterion determined for the converted image data corresponding to the frequency component.
[0009]
Further, as described above, since the result of the DCT processing at the time of the compression processing is used for AF control, no new configuration is added for AF, and focus adjustment is performed in small block units within the image area, that is, It is possible to perform spot focus.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of an imaging device having an AF control unit according to the present invention will be described with reference to the drawings.
[0011]
FIG. 1 is a diagram illustrating a configuration of an imaging device according to a first embodiment of the present invention. FIG. 2 is a diagram showing the configuration of a small block (8 × 8 pixels) to be subjected to two-dimensional DCT processing. FIG. 3 is a diagram showing the converted values in the small block subjected to two-dimensional DCT processing for each frequency component. It is a figure showing the example of reading which reads. FIG. 4 is a diagram illustrating an example of a weight value table.
[0012]
In FIG. 1, a lens 10 includes an imaging lens and a focus lens, converges light from a subject, and causes an image sensor 20 to form a subject image. The lens drive unit 60 moves and adjusts the position of the lens 10 so that the subject image is focused on the image sensor.
[0013]
The image sensor 20 has an image sensor such as a charge coupled device (CCD) or the like, converts the formed subject image into analog image data of an electric signal, and supplies the analog image data to the front end 30.
[0014]
The front end 30 first performs signal processing such as noise removal and gain adjustment by the signal processor 32 on the analog image data from the image sensor 20, and then converts the analog image data by the A / D converter 34. It is converted into digital image data (hereinafter, original image data). The original image data (for example, 1280 × 1024 pixels) is supplied to the data compression unit 40.
[0015]
The data compression unit 40 includes a two-dimensional DCT unit 42 and a quantization unit 44, and converts the two-dimensional image data into a small block of a predetermined number of vertical and horizontal pixels (for example, 8 × 8, 64 pixels) d (i , J), perform data compression processing for each block, and output compressed image data F (i, j). The size of the small block is usually set to about 8 × 8 or 16 × 16 in consideration of the saturation tendency of the conversion efficiency and the increase in the device scale.
[0016]
In the DCT unit 42, addition, subtraction, and multiplication using COS coefficients are performed by butterfly operation. Further, the quantization unit 44 quantizes the image data by dividing each frequency component of the DCT-processed transformed image data by an appropriate quantization coefficient. By changing the coefficient used for the quantization, the degree of data compression is changed.
[0017]
First, one-dimensional DCT processing is sequentially performed for each pixel in the vertical direction (row direction) for a pixel d (i, j) in a certain small block configured as shown in FIG. The results are sequentially stored in a buffer. Next, the data processed in the vertical direction stored in the buffer is sequentially subjected to one-dimensional DCT processing for each row in the horizontal direction (column direction), and this direction is further divided for each frequency component. In this way, the transformed image data f (i, j) that has been subjected to DCT processing in two dimensions in the vertical and horizontal directions and divided for each frequency component is obtained as shown in FIG.
[0018]
The converted image data f (i, j) is sequentially read from the low frequency side to the high frequency side as shown by the arrow in FIG. The quantization unit 44 quantizes the transformed image data f (i, j) according to a quantization coefficient set corresponding to the frequency division.
[0019]
In this way, the original image data is compressed for each small block, and the compressed image data F (i, j) is stored in the data storage device or transmitted to the communication destination.
[0020]
In the present invention, the AF control unit 50 is provided, and based on the converted image data f (i, j) in the middle of data compression, an AF control signal is created, and the lens 10 is moved so as to focus.
[0021]
The AF control unit 50 inputs the converted image data f (i, j) to the weight value multiplying unit 52 in the order shown by the arrow in FIG. 3, and synchronizes with the converted image data f (i, j). , J) are supplied.
[0022]
FIG. 4 is a diagram showing an example of this weighting value. In accordance with the level of the horizontal frequency i and the vertical frequency j, the weighting value k is set to increase as the frequency increases. In this example, the weight value k is 0 on the low frequency side, and the weight values k of 1 to 6 are set as the frequency increases. This weighting value k is an example, and may be set to another value as described in parentheses (1 to 3).
[0023]
From the weight value multiplying unit 52, a result k · f (i, j) obtained by multiplying the converted image data f (i, j) by the weight value k is sequentially output. The result of the multiplication is integrated by the integration unit 56, and the integrated value Σk · f (i, j) is supplied to the focus adjustment unit 58.
[0024]
The integration in the integrating unit 56 may be performed by collecting a plurality of small blocks, or may be performed by distributing the small blocks to a plurality of locations. Further, in order to perform spot focus, it may be performed only for a single small block.
[0025]
Now, the AF control of the image pickup apparatus of FIG. 1 configured as described above will be described with reference to the flowchart of FIG.
[0026]
First, when the AF control is started (Step S101), the lens 10 is moved to an initial position (Step S102). As the initial position, the closest end or the farthest end is assumed. Here, the closest end will be described below.
[0027]
The image data at this initial position is captured (step S103). Next, a small block used for AF control is cut out (step S104), and two-dimensional DCT processing is performed on the image data d (i, j) in the cut out small block (step S105).
[0028]
Next, a high-frequency component of a DCT coefficient is extracted from the DCT-processed transformed image data f (i, j) (step S106). The extraction of the high frequency component in step S106 corresponds to the processing in the weight value multiplication unit 52, the weight value table 54, and the integration unit 56 in FIG. 1, that is, the acquisition of the integration value Σk · f (i, j).
[0029]
Next, it is determined whether or not the extracted high frequency component is larger than the previous one (step S107). If it is larger (Yes), the position of the lens 10 is moved to the far end by one step, and the process returns to step S103. To capture the image data at that position. Hereinafter, these processes are repeated until the determination result in step S107 becomes “not larger than the previous time (No)”.
[0030]
If the result of the determination in step S107 is “not greater than previous time (No)”, it means that the extracted high frequency component has passed the peak, so the position of the lens 10 is returned to the near end by one step (step S109). ). Thereby, the position of the lens 10 is set to the in-focus position where the high-frequency component is the largest, so that the AF control of the imaging device ends (step S110). The processing in step S109 is not necessarily performed, and can be omitted.
[0031]
Note that it is also possible to perform AF control from the current lens position without performing step S102. In this case, the directivity is determined based on the mutual magnitude comparison of the high-frequency components extracted in the first two high-frequency component processings of the DCT coefficients, and then the peak point of the high-frequency components is searched. .
[0032]
According to the first embodiment, a converted image that is a DCT processing result in units of small blocks of 8 × 8 pixels, which is performed for compression processing such as MPEG or JPEG on captured image data. Since the data f (i, j) is used for the AF processing, the configuration for the AF processing can be simplified.
[0033]
In addition, the evaluation of the converted image data f (i, j) is performed using the weight value k set so as to become larger toward the higher frequency side according to the level of the horizontal frequency i and the vertical frequency j. By using the weighting value k corresponding to the frequency as the evaluation processing reference, it is possible to appropriately evaluate each high-frequency component and improve the AF accuracy.
[0034]
In addition, since the DCT processing result in small block units used for compression processing such as MPEG or JPEG is used for AF control, one area or a plurality of areas in the image area can be used without adding a new configuration for AF. Focus adjustment, that is, spot focus can be performed using the data of the area. Examples of the region include a DCT processing region in a central portion of the image area, a DCT processing region dispersed in a plurality of portions of the image area, and a plurality of DCT processing regions in the central portion of the image area. Can be set variously according to the characteristics of
[0035]
FIG. 6 is a diagram illustrating a configuration of an imaging apparatus according to the second embodiment of the present invention, in particular, an AF control unit 50.
[0036]
In FIG. 6, a weight value k to be supplied to the weight value multiplying unit 52 can be selected. A first weight value table 54 and a second weight value table 55 are provided, and the weight value of either table is determined. You can select whether to use a value. The selection is performed by switching the switch 57 to one of the weight value tables 54 and 55 by the selection signal CS. Other configurations are the same as those of the first embodiment in FIG.
[0037]
As shown in FIG. 4, the weight values k of the weight value tables 54 and 55 are set in the same area and have different sizes (0 to 6, 0 to 3). Further, a weight value table in which regions for setting weight values are different may be provided. The number of weight value tables may be an arbitrary number of 2 or more.
[0038]
In the imaging apparatus according to the second embodiment shown in FIG. 6, a plurality of weight value tables having different weight values are provided. AF control can be performed.
[0039]
FIG. 7 is a diagram illustrating a configuration of an imaging device, particularly, an AF control unit 50 according to a third embodiment of the present invention.
[0040]
In FIG. 7, as compared with the first embodiment of FIG. 1, a selection unit 51 is provided instead of weight value multiplication unit 52 of FIG. 1, and a selection bit is provided instead of weight value table 54 of FIG. A table 53 is provided. Other configurations are the same as those of the first embodiment in FIG.
[0041]
As shown in FIG. 8, the selection bit table 53 sets the selection bit to 0 so as not to select the low frequency side and selects a predetermined area on the high frequency side according to the level of the horizontal frequency i and the vertical frequency j. The selection bit is set to "1". Further, the selection unit 51 responds by inputting the sequentially input converted image data f (i, j) to a gate circuit controlled to be opened or closed by a corresponding selection bit k (0 or 1). Only the converted image data f (i, j) whose selection bit is 1, that is, the converted image data f (i, j) on the high frequency side passes. The passed converted image data f (i, j) is integrated by the integration unit 58 and supplied to the focus adjustment unit 58.
[0042]
In the imaging apparatus according to the third embodiment of FIG. 7, the selection unit 51 can be configured by a gate that is opened or closed according to the selection bit 0 or 1, so that the configuration of the AF control unit 50 is simplified. .
[0043]
【The invention's effect】
According to the imaging apparatus of the present invention, a converted image that is a DCT processing result in small block units (for example, 8 × 8 pixels) performed for compression processing such as MPEG or JPEG on captured image data By using the data for the AF processing, the configuration for the AF processing can be simplified. Further, the AF accuracy can be improved by appropriately evaluating each converted high-frequency component based on the evaluation processing criterion determined for the converted image data corresponding to the frequency component.
[0044]
In addition, since the processing result of DCT processing in small blocks used for compression processing such as MPEG and JPEG is used for AF control, one small image in the image area can be added without adding a new configuration for AF. It is possible to perform focus adjustment, that is, spot focus, using data of an area or a plurality of small areas.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of an imaging device according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a configuration of a small block (8 × 8 pixels) on which two-dimensional DCT processing is performed.
FIG. 3 is a diagram showing a read example in which frequency components of small blocks are sequentially read.
FIG. 4 is a diagram showing an example of a weight value table.
FIG. 5 is a flowchart of AF control of the imaging apparatus in FIG. 1;
FIG. 6 is a diagram showing a configuration of an AF control unit according to a second embodiment of the present invention.
FIG. 7 is a diagram showing a configuration of an AF control unit according to a third embodiment of the present invention.
FIG. 8 is a diagram showing a configuration of a selection bit table of FIG. 7;
[Explanation of symbols]
Reference Signs List 10 lens 20 image sensor 30 front end 32 signal processor 34 A / D converter 40 data compression unit 42 two-dimensional DCT unit 44 quantization unit 50 AF control unit 51 selection unit 52 weight value multiplication unit 53 selection bit tables 54, 55 Weight value table 56 Integrator 57 Switch 58 Focus adjuster 60 Lens driver

Claims (1)

A lens system having a lens and a driving unit thereof,
An image sensor that forms a subject image with light from the lens system,
A front end unit that outputs image data corresponding to a subject image from the image sensor,
The DCT unit performs DCT processing on the image data from the front end unit for each predetermined block to obtain converted image data divided for each frequency component, and the converted image data corresponds to the frequency division by the quantization unit. In the imaging device having a data compression unit that performs a quantization process according to the quantization coefficient that is set as the data and outputs the data as compressed image data,
Further, the image processing apparatus further includes an AF control unit that creates an AF control signal based on the converted image data obtained by the DCT unit and supplies the AF control signal to the lens system.
The AF control unit performs an evaluation process determined corresponding to the frequency component on the converted image data, integrates the converted image data subjected to the evaluation process, and generates the AF control signal based on the integrated value. An imaging device characterized by forming:

Images