JPH11119090A - Automatic focus detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic focus detector capable of more accurately detecting out-of-focus amount by judging the distance from the optical axis of an image reforming lens in sensor data and correcting the out-of-focus amount in accordance with the distance. SOLUTION: An area sensor 6 is scanned by a sensor driving means 8 so as to read in the charge of a pair of subject images. An out-of-focus amount arithmetic operation means 7 calculates the out-of-focus amount from transferred stored data concerning all the lines of the area sensor 6, and further an optical axis deviation calculation means 9 calculates the distance from the optical axis of the image reforming lens. Thus, an out-of-focus amount correction means 10 selects a correction block, calculates correction data by a corresponding correction expression, and subtracts the correction data from the out-of-focus amount calculated previously so as to perform correction. The out-of-focus amount is accurately detected concerning all the areas.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a phase difference detection type automatic focus detection device used for a camera or the like.

[0002]

2. Description of the Related Art Conventionally, a phase difference detection type automatic focus detection device uses a one-dimensional line sensor as a photoelectric conversion element. The same two images are formed on this line sensor by a pair of re-imaging lenses. Since the distance between these two images changes according to the amount of defocus on the film surface,
The amount of defocus can be detected by measuring the image interval.

While the prior art uses a one-dimensional line sensor, the present applicant has proposed an automatic focus detection device that uses a two-dimensional area sensor as a photoelectric conversion element in order to improve detection accuracy. Nippon 8-330420,
Japanese Patent Application No. 8-330421). This automatic focus detection device
By using the area sensor, a subject image formed by the re-imaging lens can be captured in a wide area. As a result, a great deal of information that could not be obtained by the line sensor can be obtained, and the detection accuracy can be improved.

[0004]

Incidentally, the re-imaging lens is a minute optical system, and it is difficult to perform optical correction for improving image performance like a photographing lens of a camera. Therefore, the image formed on the area sensor slightly deteriorates as the distance from the optical axis of the re-imaging lens increases. This results in a detection error of the defocus amount. The present invention solves the above-described problem. The problem is to determine the distance of the sensor data from the optical axis of the re-imaging lens, and correct the defocus amount according to the distance, thereby accurately adjusting the defocus amount. It is an object of the present invention to provide an automatic focus detection device capable of detecting the focus.

[0005]

In order to solve the above-mentioned problems, an automatic focus detection device according to the present invention comprises a field mask for determining a field of view, and a pair of object images formed by reflected light of the object passing through the field mask. A pair of re-imaging lenses, a photoelectric conversion element in which a pair of light receiving elements for converting a pair of subject images formed by the re-imaging lens into electric charges are two-dimensionally arranged, and Calculating means for calculating the focus shift amount; optical axis shift amount calculating means for calculating how much the output of the photoelectric conversion element is data shifted from the optical axis of the re-imaging lens; Correction means for correcting the defocus amount. The correction means can correct the defocus amount using an approximate expression of a defocus amount and an image interval error in each line of the photoelectric conversion element.

[0006]

According to the above arrangement, the amount of defocus can be detected without error in a wide area of the image formed by the re-imaging lens, and the detection accuracy can be improved and a wide detection area can be realized.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in more detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of an automatic focus detection device according to the present invention. In this embodiment, the film equivalent surface 1, the field mask 2, the condenser lens 3, the aperture mask 4, the re-imaging lens 5, the photoelectric conversion element 6, the sensor driving means 8, the defocus amount calculating means 7, the re-imaging lens And an optical axis shift amount calculating means 9 and a focus shift amount correcting means 10. On the photoelectric conversion element 6, an image of a region defined by the field mask 2 is formed as a pair of identical images by the pair of re-imaging lenses 5. In the focus detection device of the phase difference method, the distance between the images changes according to the amount of defocus on the film surface. Therefore, by calculating this image distance, the amount of defocus on the film surface can be known. .

In a conventional one-dimensionally arranged photoelectric conversion element (line sensor), the sensor is a re-imaging lens 5 as shown in FIG.
In a two-dimensional array of photoelectric conversion elements (area sensors), as shown in FIG.
There is also a sensor outside the optical axis. Therefore, with the two-dimensional array of area sensors 6, a wide detection area that cannot be obtained with the line sensor can be obtained. Since the area sensor sends each line output sequentially, output data of a plurality of lines can be obtained. When the amount of defocus is calculated based on this output data, the result can be obtained for the number of lines.

FIG. 4 shows a plot of the value of the defocus amount (for example, 128 lines in the area) obtained from all the line data under the condition that the defocus amount is constant in the field mask region. Since the focus shift is constant in the area, the calculated focus shift should be constant, but the focus shift on the line deviated from the optical axis of the re-imaging lens greatly deviates from the focus shift on the optical axis. I have. That is, it can be understood that the image interval changes as the distance from the optical axis increases. This is due to the optical characteristics of the re-imaging lens and the condenser lens. In particular, changes in the center of gravity of the light beam due to coma aberration and distortion are the main causes. Although it is possible to correct these aberrations on the optical design, the size and complexity of the optical system become large, and it becomes impossible to satisfy the demands such as cost reduction. In the present invention, the deviation of the distance measurement result of each line in the area sensor is corrected, and an accurate defocus amount is detected over the entire area.

Next, a correction method will be described with reference to FIGS. The error of the defocus amount shown in FIG. 4 changes depending on the distance from the optical axis of the re-imaging lens and the defocus. Therefore, the correction value is obtained from these two values. FIG.
FIG. 5 is a graph in which the distance xob from the optical axis of the re-imaging lens as a parameter is used to plot the calculated amount of deviation of the image interval due to a change in defocus. A shift in the image interval causes an error in the focus shift amount. This error is determined by the optical characteristics of the re-imaging lens. This error amount is stored as a correction value under each condition, and the correction value under the corresponding condition is read and corrected by the correction means during the actual focus detection operation.

However, in this method, as described above, the matrix must have correction values that change under the two conditions of the distance from the optical axis and the defocus, and if the correction accuracy is to be improved, the division of the conditions will be fine. And the amount of correction values to be stored becomes enormous, which is not practical. Of course, this method may be used if a large amount of correction data can be stored in the system. In the present embodiment, the error characteristic curve obtained by the optical calculation is replaced with an approximate expression, and the error amount is obtained by a simple calculation. FIG. 6 shows in FIG.
5 is a graph obtained by dividing an error amount into four blocks according to a defocus amount and linearly approximating each block.

FIG. 7A is a plot of the change in the slope of each block of the approximate straight line in FIG. FIG.
FIG. 7B is a graph obtained by approximating FIG. 7A by the least square method. FIG. 8A is a plot obtained by extracting the value of the y-intersection of each block of the approximate straight line in FIG. FIG. 8B is a graph obtained by approximating FIG. 8A by the least square method.

The correction equation for each block obtained from the approximation of FIGS. 7B and 8B can be expressed as follows. A and b of each block in the form of correction value = a × D + b
The value of the first block a = −0.2 × X ² b = 0.26 × X The second block a = −0.56 × X ² b = 0.9 × X The third block a = 0.04 × Xb = 0.28 × X Fourth block a = 0.3 × X ² b = −3.0 × X ^{2 where} D: Defocus amount X: Deviation amount from optical axis of re-imaging lens By having the formula, the correction amount (error amount) can be obtained by a simple calculation. The deviation amount of the re-imaging lens from the optical axis can be easily calculated from the line interval of the sensor because the line of the optical axis when the sensor is incorporated is determined by design.

FIG. 9 is a flow chart for explaining the operation of the automatic focus detection device according to the present invention. First, the area sensor 6 is controlled by the sensor driving means 8 to start accumulating electric charges (step (hereinafter referred to as "S") 901). The control means (not shown) determines that the accumulation is completed (S90).
2) When the accumulation is completed, the accumulated data of the area sensor 6 is read and transferred to the outside (S903). The defocus amount calculating means 7 calculates the defocus amount for each line from the obtained data of all lines of the area sensor 6 (S9).
04). Further, the optical axis shift amount calculating means 9 calculates the distance of the line data from the optical axis of the re-imaging lens (S90).
5).

The defocus amount correcting means 10 selects the above-described correction block (S906), and calculates correction data using a corresponding correction formula (S907). Since this correction data becomes the error amount itself, correction is performed by subtracting the correction data from the previously calculated defocus amount (S908). It is determined whether or not the above-described correction operation has been completed for all lines (109), and if not all have been completed, the process returns to S904. If all have been completed, terminate. The above operation shows the case where all the lines are calculated, but only the necessary lines may be calculated.

[0016]

As described above, the present invention provides a field mask for determining a field of view, a pair of re-imaging lenses for forming a pair of subject images from reflected light of the subject passing through the field mask, A photoelectric conversion element in which a pair of light receiving elements for converting a pair of subject images formed by the re-imaging lens into electric charges are two-dimensionally arranged; calculating means for calculating a defocus amount from an output of the photoelectric conversion element; Is calculated by calculating how much data is deviated from the optical axis of the re-imaging lens, and correcting means for correcting the defocus amount according to the optical axis deviation amount. The amount of defocus can be detected without error in a wide area of the image formed by the imaging lens. Therefore, it is possible to improve the accuracy of detecting the amount of defocus and to realize a wide detection area, and to use the data of the area sensor widely. Also, if an approximate expression is used as the correction means, it is not necessary to have a huge storage area for correction values, and it is possible to easily improve the focus detection accuracy without increasing the size of the optical system and without increasing the cost. There is an effect that can be.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of an automatic focus detection device according to the present invention.

FIG. 2 is a perspective view showing a configuration example of an optical system using an area sensor.

FIG. 3 is a perspective view showing a configuration example of an optical system using a line sensor.

FIG. 4 shows all lines (each distance from the optical axis of the re-imaging lens)
5 is a graph showing the measured amount of defocus at the point.

FIG. 5 is a graph showing an image interval error with respect to a defocus amount using a distance from the optical axis of the re-imaging lens as a parameter.

FIG. 6 is a graph obtained by linearly interpolating FIG. 5;

FIG. 7 is a diagram illustrating a slope characteristic of an interpolation straight line.

FIG. 8 is a diagram showing y-intersection characteristics of an interpolation straight line.

FIG. 9 is a flowchart illustrating a focus detection operation.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Film equivalent surface 2 ... Field mask 3 ... Condenser lens 4 ... Aperture mask 5 ... Re-imaging lens 6 ... Photoelectric conversion element (area sensor) 7 ... Defocus amount calculation means 8 ... Sensor control means 9 ... Optical axis shift amount Calculation means 10: Defocus amount correction means 20: Photoelectric conversion element (line sensor)

Claims (2)

[Claims] 1. A field mask for determining a field of view, a pair of re-imaging lenses for forming a pair of subject images from reflected light of the subject passing through the field mask, and a re-imaging lens A photoelectric conversion element in which light receiving elements for converting the pair of subject images into electric charges are two-dimensionally arranged; an arithmetic unit for calculating an amount of defocus from the output of the photoelectric conversion element; An optical axis shift amount calculating means for calculating how much data is shifted from the optical axis, and a correcting means for correcting a focus shift amount in accordance with the optical axis shift amount. Detection device. 2. The automatic focus detection device according to claim 1, wherein the correction unit corrects the defocus amount using an approximate expression of a defocus amount and an image interval error in each line of the photoelectric conversion element. apparatus.