JP3279421B2 - camera - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera, and more particularly to a camera having improved focusing characteristics.

[0002]

2. Description of the Related Art In an autofocus system, a subject image obtained through an optical system such as a lens is converted into an electric signal, and contrast information is detected from the electric signal so that the detected contrast information value is maximized. There is one that performs feedback control of the position of an optical system. Autofocus control by such a method is used for a video camera or the like.

FIG. 15 shows a configuration of a conventional still video camera.
Is shown in The position of the lens 2 is driven and controlled by a lens driving circuit 1, and a subject image is captured by a CCD as an image sensor.
3 is imaged. The image signal converted into an electric signal by the CCD 3 is subjected to processing such as γ correction and color separation in an imaging process circuit 4, and is converted into digital data by an A / D converter (ADC) 5. Will be recorded. As is well known, block data of N pixels (for example, 8 pixels) is read from the buffer memory 6 and subjected to orthogonal transform processing in a DCT (discrete cosine transform) circuit 8. The DC transform coefficient obtained by the orthogonal transform by the DCT circuit 8 is supplied to the quantization circuit 9 and the lens AF control circuit 40, while the AC transform coefficient is converted to the quantization circuit 10 and the lens AF control circuit 40. And supplied to. The lens AF control circuit 40 receives the AC conversion coefficient, controls the lens driving circuit 1 to move the lens 2, and controls the position of the lens 2 so that the AC conversion coefficient obtained by the DCT circuit 8 is maximized. Focus control. The pulse generation (SSG) circuit 7 includes a CCD 3,
Various pulses such as horizontal and vertical synchronization signals for controlling the imaging process circuit 4, the A / D converter 5, the buffer memory 6, and the lens AF control circuit 40 are generated.

The DC conversion coefficient quantized by the quantization circuit 9 is supplied to a delay circuit 11 and a subtraction circuit 12, and a prediction error is obtained as an output of the subtraction circuit 12. This prediction error is encoded by the encoding circuit 14 and output to the synthesizing circuit 16. On the other hand, the AC transform coefficients quantized by the quantization circuit 10 are subjected to coefficient rearrangement processing by a so-called zigzag scanning circuit 13, then encoded by an encoding circuit 15, and output to a synthesis circuit 16. D synthesized by the synthesis circuit 16
The prediction error of the C conversion coefficient and the AC conversion coefficient are stored in the recording device 1
7 recorded. The entire sequence of these components is controlled by the system control circuit 18. The DCT circuit 8 is often used at the time of image encoding, and if contrast information is detected by the DCT circuit,
Since the DCT circuit can be shared, the configuration is simplified and the cost is reduced.

[0005]

As described above, in the conventional automatic focusing (AF) control of a camera, a DCT operation is performed on the entire area of a screen, and the obtained DCT coefficients (AC conversion coefficients and DC conversion coefficients) are converted. The AF control is performed based on this (for example, a method described in Japanese Patent Application Laid-Open No. Hei 3-216078). However, when the DCT calculation is performed on the entire screen, there is a problem that the calculation takes time and the focusing speed becomes slow. Further, even when the DCT calculation is performed only on the image of the AF area which is the minimum necessary for the AF control in order to reduce the calculation amount, the delay of the focusing speed still remains, and the information amount required for the AF control is not limited. As a result,
There is a risk of false focusing.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a camera which not only enables high-speed and high-precision focusing while maintaining a simple structure, but also recognizes camera shake.

[0007]

In order to solve the above-mentioned problems, a camera according to the present invention uses an output signal of an image pickup device for A / A.
When a screen related to the video signal is divided into a plurality of blocks, a DCT
DCT calculating means for performing a calculation to obtain a DCT coefficient; AC coefficient sum calculating means for calculating, for each block, a sum of all coefficients corresponding to an AC region with respect to the DCT coefficient obtained by the DCT calculating means; A block for recognizing, as a specific pattern, a pattern of the blocks or a set of the blocks whose total sum of the AC coefficients exceeds a predetermined threshold value by the AC coefficient sum calculating means and which is in a plurality of specific areas set in the screen. It is provided with a specific pattern recognizing means, and a camera shake recognizing means for recognizing whether or not a camera shake is to occur by performing time-series pattern matching on the specific pattern recognized by the specific pattern recognizing means.

[0008]

According to the present invention, the DCT obtained by the DCT operation means is used.
The sum of all coefficients corresponding to the AC area is calculated for each block on the screen, the specific pattern is recognized for blocks where the sum of the AC coefficients exceeds a predetermined threshold, and camera shake is performed by time-series pattern matching for the specific pattern. Camera shake recognition function for recognizing the situation of

[0009]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a configuration block diagram illustrating an example of a camera related to the present invention. In FIG. 1, FIG.
The components denoted by the same reference numerals indicate components having the same function. In the present example, the entire screen or the inside of the AF area, for example, 8
The pixel block data is supplied to the lens AF calculation control circuit 30. The operation control circuit 30 performs a DCT operation on the supplied block data to obtain a transform coefficient. 8 pixels x 8
FIG. 2 shows an example of a transform coefficient Cuv obtained by performing a DCT operation on block data composed of pixels. In FIG. 2, as u increases, the horizontal frequency component becomes higher, and as v increases, the vertical frequency component becomes higher. Therefore, C00 indicates the magnitude of the DC component (DC conversion coefficient), and higher high frequency components in the horizontal direction toward the right direction (u: 0 to 7) are also lower than C00 (from v: 0). Higher high frequency components in the vertical direction towards 7) will be shown. Here, the remaining transform coefficients excluding the DC transform coefficients are referred to as AC transform coefficients.

Referring to FIG. 3, the AF control based on the conversion coefficient value obtained as described above is performed in such a manner that the coefficient value is small when the lens position is at an almost infinite distance and when the lens position is at the closest distance. This is performed using the feature that the coefficient value shows a peak at the position.

As described above, AF control is performed using the increase / decrease change of the conversion coefficient value. The area of the coefficient used for AF control is fixed, and the AC conversion coefficient of only the part to be used is converted to DC.
The focusing speed can be improved if it is obtained by T calculation. For example, as shown in FIG. 4, if only the transform coefficients in the shaded area (indicating the frequency domain) are used for the AF control, only the shaded area needs to be subjected to the DCT operation.
The speed can be increased about four times as compared with the case where the calculation is performed.

In the above example, the area of the coefficient used for the AF control is fixed. However, if it is arbitrarily changed, the following advantages can be obtained. For example, when performing AF control, since it is not known whether the focal point is at infinity or in the closest direction, the lens is slightly driven in an arbitrary direction, the coefficient increase / decrease change state is checked, and the lens is driven in the increasing direction. Then, the direction is determined as a pre-process for performing the AF control.

In the large blur state, even if the lens is slightly driven, the change amount of the conversion coefficient on the high frequency side is small. Therefore, when the direction is determined, the conversion coefficient on the low frequency side as shown by the hatched portion in FIG. In the AF control, the conversion coefficient on the high frequency side as shown by the hatched portion in FIG. This is because, as shown in FIG. 6, when the lens is far from the in-focus position (close to infinity and closest), the change of the high-frequency side conversion coefficient with respect to the change of the lens position is small, whereas the change of the low-frequency side conversion coefficient is small. The focus is on the fact that the change is relatively large, while the coefficient change on the high frequency side is large near the focus position.

In another example related to the present invention, AF
At the start of the control, all the DCT calculations in the AF area are performed, and the distribution of the transform coefficients is checked. For example, if a lot of frequency components on the vertical side are included, as shown by the hatched portion in FIG. May be automatically varied, for example, by mainly using the conversion coefficient for AF control. Here, a certain amount of conversion coefficient pattern is stored in a ROM or the like, and an optimum pattern is extracted for each AF control, or only a portion where the value of the conversion coefficient is high is extracted using fuzzy logic to perform AF control. It can also be used.

For a long time, such as a movie,
At the time of control, when there is a change in the subject image due to a change in the DC and AC conversion coefficients, the area pattern of the coefficient used for AF control can be changed to an optimum one each time.

FIG. 8 shows a procedure of the above-described AF control operation and a recording operation processing of a DCT coefficient as image information.
First, a trigger input of the first stage for the AF operation is waited (step S1), and A is calculated by calculating only the DCT coefficient necessary for the AF control.
After performing the F control (step S2), it waits for the AF control to end (step S3). Next, a second-step trigger input for recording operation instruction is waited (step S4), DCT coefficients of all areas are calculated (step S5), and the coefficients are recorded on a storage medium such as a memory card (step S6). finish. As a result, since the minimum necessary calculation is performed during AF, the transition to the recording operation is performed promptly, and the risk of missing a photo opportunity or the like can be reduced.

Although the above-described DCT processing is performed by software processing in the lens AF calculation control circuit 30, it is needless to say that the DCT processing can be implemented by hardware.

Next, an example of reducing the adverse effect of flicker in connection with the present invention will be described. This effect is due to the fact that the coefficient obtained by the DCT operation is affected by flicker of the light source. The false focusing operation by the above-mentioned flicker is shown in FIG.
This will be described with reference to FIG. The curve shown by the broken line in the same figure is a characteristic showing the relationship between the lens position and the coefficient value change when there is no flicker. The characteristic curve indicated by the solid line shows the characteristic when flicker is present. Due to the flicker, the characteristic curve has a sharp peak at a lens position that is greatly irregular and deviates from an accurate focusing position, resulting in a false peak. Causes a false focusing operation.

In this embodiment, flicker is suppressed by using a DC conversion coefficient for the amount of change of the light source. That is, taking advantage of the fact that the DC conversion coefficient is large when it is bright and small when it is dark, the AF control is performed using the coefficient x obtained by the following equation that reflects the change in the DC coefficient. x = (AC conversion coefficient used for AF control) × (DC conversion coefficient at start of AF control) / (DC conversion coefficient of current field) According to the AF control using such a coefficient x, the dotted line in FIG. As a result, a smooth characteristic curve free from flicker is obtained, and stable and reliable AF control becomes possible. If the coefficient y represented by y = x × β is used instead of x, the influence of flicker can be further suppressed. Here, β is obtained from a conversion table as an example shown in FIG. 10 based on a coefficient α defined by the following equation. α = {(DC coefficient at start of AF control) − (DC coefficient of current field)} / (DC coefficient at start of AF control) × 100 [%] β is a DC conversion coefficient that changes according to brightness. Since the AC coefficient does not always change in accordance with the change in accordance with the change, the AC coefficient is compensated for and can be obtained experimentally.

Next, in connection with the present invention, an example of obtaining data for moving object tracking and camera shake correction using a transform coefficient obtained by DCT will be described. As a moving object tracking method, Japanese Patent Publication No. 59-32743 discloses a method of projecting a two-dimensional video signal of a tracking area one-dimensionally, checking a correlation of each image, and tracking a portion having a high correlation. I have. In this method, when the tracking area is enlarged, the amount of data used for correlation increases, and therefore, the time required for the correlation calculation increases, and tracking performance may be degraded. In some cases, tracking becomes impossible due to loss of tracking.

In this example, in FIG. 11, a video of one screen is subjected to DCT calculation, and a tracking area is locked to a subject to be tracked. Next, the sum of the absolute values of all the AC transform coefficients in the DCT coefficients is obtained for each block, and a portion where the sum is equal to or larger than a predetermined threshold (Th) is regarded as a subject to be tracked. Also, for noise suppression, only continuous regions are extracted.
As a result, a main subject can be extracted.

Blocks satisfying the relationship with the threshold value Th shown in the equation in FIG. 12 are indicated by shaded areas in FIG. Among these oblique lines, those located in the tracking area shown in FIG. 11 are the part g, so that the subject in the part g may be tracked from the next screen. Further, when the size of the subject g changes, the tracking area can be sequentially changed in accordance with the change. In addition, full-screen DCT for high speed
DCT only for tracking area and around tracking area without conversion
The calculation may be performed. Further, the threshold value Th may be an arbitrary fixed value, or may be an AC value inside or outside the tracking area.
The difference amount of the sum of the absolute values of the conversion coefficients may be determined automatically, for example, by an operation based on fuzzy or the like, and may be sequentially varied.

In the formula of FIG. 12, the sum of the absolute values of all the AC conversion coefficients is used.
Considering the subject brightness, the degree of focusing, etc.), the sum of the absolute values of only the AC conversion coefficients related to the area optimal for each condition may be used. Alternatively, the sum of the absolute values of the AC conversion coefficients in the area used during the AF control may be used. AC for this optimal region
The conversion coefficient can be obtained by calculation according to the distribution state of the AC conversion coefficient, or an arbitrary fixed coefficient can be used. Further, in the above description, the subject is regarded as being equal to or greater than the threshold (Th), but may be regarded as being subject according to conditions. Also, D
Although the C coefficient is not used, the subject may be extracted by focusing only on the change of the DC coefficient, or the subject may be extracted by comprehensively judging both changes of the DC coefficient and the AC coefficient. As described above, tracking of a moving object and extraction of an arbitrary subject can be performed.

Next, utilizing such subject extraction,
An embodiment of the present invention for detecting a screen shift amount (for example, a camera shake amount) will be described. For example, as shown in FIG. 13A, appropriate areas i, j, k, and l are defined in four corners of the screen, and subjects at each of the four corners are extracted by the equation in FIG. 12 (FIG. 13B). The same process is performed on the next screen (FIG. 14A) to obtain a subject pattern as shown in FIG. 14B. Next, a pattern matching process is performed on each of the four corners in FIGS. 13B and 14B to detect the amount of screen shift.

In the above-described embodiment, four corners are set as the DCT target areas.
It is clear that there is only one area, and the number and the size of the block for performing DCT (in this example, 8 ×
8 pixels) is also arbitrary and can be set in consideration of cost and accuracy.

According to the present embodiment, high-speed and accurate focusing control can be performed without adding special hardware. Obviously, the present invention can be applied not only to an electronic still camera but also to other cameras such as a movie camera.

[0027]

As described above, according to the camera of the present invention, not only the focusing operation can be performed at high speed and high precision, but also the camera shake can be recognized.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram illustrating an example of a camera related to the present invention.

FIG. 2 is a diagram showing transform coefficients obtained by performing a DCT operation on block data of 8 pixels in length and width.

FIG. 3 is a diagram for explaining AF control based on a transform coefficient obtained by DCT.

FIG. 4 is a diagram illustrating an example in which a coefficient area used for AF control is fixed.

FIG. 5 is a diagram illustrating a conversion coefficient area used when determining a lens driving direction during AF control and during AF control.

FIG. 6 is a diagram illustrating a relationship between a lens position and a high-frequency conversion coefficient and a low-frequency conversion coefficient.

FIG. 7 is a diagram illustrating an example of a distribution of conversion coefficients.

FIG. 8 is a diagram showing an AF control operation and an image information recording operation processing procedure.

FIG. 9 is a diagram illustrating a relationship between a lens position and a conversion coefficient in AF control based on the presence or absence of flicker according to the present invention.

FIG. 10 is a diagram showing an example of a conversion table between coefficients α and β in an example related to the present invention.

FIG. 11 is a diagram for explaining an operation tracking method according to the present invention.

FIG. 12 is a diagram for explaining the operation of the example shown in FIG. 11;

FIG. 13 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 14 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 15 is a block diagram of one configuration of a camera based on conventional AF control.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 lens drive circuit 2 lens 3 CCD 4 imaging process circuit 5 A / D converter 6, 21 buffer memory 7 pulse generation circuit 8 DCT circuit 9, 10 quantization circuit 11 delay circuit 12 subtraction circuit 13 zigzag scanning circuit 14, 15 encoding Circuit 16 Synthesis circuit 17 Recording device 18 System control circuit 30, 40 Lens AF control circuit

Claims (1)

(57) [Claims] 1. A digital video signal obtained by A / D conversion of an output signal of an image sensor is subjected to a DCT operation for each block when a screen related to the video signal is divided into a plurality of blocks.
DCT calculating means for obtaining CT coefficients; AC coefficient sum calculating means for calculating, for each block, a sum of all coefficients corresponding to an AC region for the DCT coefficients obtained by the DCT calculating means; A specific pattern recognizing means for recognizing, as a specific pattern, a pattern of a block or a set thereof, in which the sum of the AC coefficients exceeds a predetermined threshold value, and which is present in a plurality of specific areas set in the screen; And a camera shake recognizing means for recognizing whether or not a camera shake is caused by time-series pattern matching of the specific pattern recognized by the specific pattern recognizing means.