JPH0664226B2 - Autofocus device - Google Patents

Description

The present invention relates to an autofocus device, and more particularly to an autofocus device suitable for reading an image on a microfilm or the like.

[Conventional technology]

A conventional device for electronically reading an image on a microfilm will be described with reference to FIG.

The film 1 is exposed by the light from the light source a that has passed through the lens b, further passes through the lens c, and is then moved by the movable optical path switching mirror d to the screen h side or the reading side
Be selectively led to. The focus is adjusted by tilting the mirror d to the screen side x and adjusting the focus of the lens group c. After this adjustment, tilting the mirror d to the reading side Y switches the optical path, and the motor g. By scanning CCDf in the direction of the arrow, the intensity of incident light is converted into an electric signal by CCDf. e is a mirror.

Here, the image formed on the CCDf and the screen h
The correspondence of the focus with the image formed on the top is adjusted by the adapter lens AL, and this adjustment is usually performed at the time of factory assembly.

Therefore, the focus of the image on the screen h and the focus on the reading CCDf side can be determined due to vibration during transport, transportation, the mechanism (optical path length) due to temperature difference, or the contraction, and the secular change. In addition to the point that it disappears, this adjustment takes a lot of time because the service person moves the adapter lens AL and adjusts it accordingly after the device is installed by the user. High technology is required.

In addition, if the adapter lens AL itself does not clearly form a film image on the screen h, the state of whether or not the operator is in the best focus state (hereinafter referred to as the in-focus position (just focus)) Therefore, the adapter lens AL is required to have very small aberration and distortion and very high resolving power. Therefore, the processing of this adapter lens is required to have high precision and cost. Will also be very expensive.

On the other hand, as described above, even if the image on the screen h and the image on the CCD f obtained via the adapter lens AL have good corresponding adjustments, an unskilled operator can bring the image to the just focus point. It is very difficult to set, and the operator moves the focusing lens up and down to change the focal length and adjust the focus of the image on the microfilm 1, either by manual knob or by electric (motor). Since the movement is performed by a device that decelerates the power from gears by gears, in this type of device, the vertical movement of the focusing lens around the in-focus position is in the range of several μm to several tens μm (depth of focus). However, if it goes out of the range beyond that, the image will be out of focus and the image information cannot be read accurately.

As described above, in this type of apparatus, there are many problems in setting the just focus point (focusing position) only by forming an image on the screen.

Further, in recent years, the digitization of documents has progressed, and so-called optical disc electronic file systems for electronically recording documents have come to be widely used. Needless to say, users who have recorded and filed documents for a long time have made use of the records as microfilm today, but the documents on this microfilm must be converted to optical disc electronic files. It is the microfilm electronic scanner that makes these possible.
In recent years, development of this type of product has been advanced.

However, as described above, it is very inconvenient and time-consuming to adjust the focus on the image on the microfilm by using the eyes of a person skilled in the art, one frame at a time, one frame at a time.

Therefore, in order to eliminate the above-mentioned drawbacks, a method to move and set the focus adjustment lens to the position set at the factory adjustment by adjusting the thickness of the film to be used and inputting the thickness of these films is considered. To be

However, microfilm means that the film thickness varies from manufacturer to manufacturer, and even within the same manufacturer, there are various types of film, such as silver film and diazo film. In that case, usable films are limited, which is very inconvenient. In addition, the above settings were misaligned due to aging and the like during use, and it was very inconvenient to call a service man to make a troublesome adjustment.

On the other hand, in fields such as still cameras, many autofocus mechanisms have been announced. FIG. 10 is a typical example thereof, and a part of the light passing through the lens U passes through the half mirror V, is reflected by the mirror W, and is sent to the beam splitter T. Here, it is divided into three types of optical path lengths.
The third sensors R1 to R3 each detect focus information (in this case, the amount of light is detected, and the greater the amount of light, the better the focus). Thus, when the light amount of the second sensor R2 is the largest, it is determined that the focus is just, and when the third sensor R3 is larger than the second sensor R2, it is the rear pin, and the second sensor R2 is the second. When the light intensity of the sensor R1 of No. 1 is large, it means the front focus.

For such a still camera or the like having a long depth of focus (focusing distance), it is possible to easily detect the focus state by changing the optical path length via the beam splitter as described above. However, in a model with a very short depth of focus, such as a microfilm reader, of a few microns to a dozen microns, physically,
It was impossible to interpose a beam splitter.

As described above, there are many problems in the microfilm reading device, and it is the current situation that the microfilm reading device equipped with the autofocus function cannot be commercialized.

[Problems to be solved by the invention]

The present invention focuses on the above-mentioned problems, and in order to solve the problems, prior to the autofocus operation, a binarization threshold value of an image signal that facilitates the autofocus operation is found in advance, It is an object of the present invention to provide an autofocus device which has a high success rate and can improve accuracy by binarizing an image signal according to the threshold value.

[Means for solving problems]

In order to achieve such an object, the present invention provides a sensor including a plurality of light receiving elements arranged in rows or on a surface, an optical means for projecting an image on the sensor, and a binary image signal from the optical means. Means, a threshold value changing means for changing the threshold value input to the binarizing means, a means for counting binary change points from the binarized image signal, and a count value from the counting means. The focus of the optical means is adjusted on the basis of the optical means, and prior to the adjustment of the focus, the threshold value changing means changes the threshold value at a position near the focus and a plurality of non-focus points, and the counting means makes a plurality of values at each position. The binary change points with respect to the threshold value are counted, and the threshold value that maximizes the difference between the count value in the vicinity of the focus position and the count value in the out-of-focus position is found and set to this threshold value. After adjusting the focus Characterized in that and means.

[Work]

In the autofocus device configured in this way,
When an image is projected on a sensor in which a plurality of light receiving elements are arranged via an optical means and an image signal obtained from the sensor is output as a binary image signal, an autofocus operation is first set. The threshold values for binarization are changed in the vicinity of the focal point and the lens positions of non-focus points before and after the focal point, and the count values of the number of binary change points are obtained respectively. The threshold value that gives the largest value among the count values obtained from each state is obtained, and after setting such a threshold value, the autofocus operation is performed to ensure accurate and accurate operation. It becomes possible to guarantee.

〔Example〕

FIG. 1 shows an embodiment of the present invention. Here, 10 is a microfilm on which an image is recorded. The light projected from the light source a is condensed and transmitted by the condenser lens 11 to the microfilm 10 placed at a predetermined position, and the image converged through the focusing lens 12 for focusing is displayed. The image can be projected on the image sensor 13 composed of a CCD element array.

Thus, the image sensor 13 reads an image by, for example, main scanning and sub-scanning, and sends it to the upper amplifier 14 which exchanges an electric sensor signal. Reference numeral 15 is a comparator, which compares the threshold value signal from a threshold value setting circuit 16, which will be described later in detail, with the image signal supplied from the amplifier 14, and binarizes the image signal. An image signal (binary image signal) can be supplied to a printer (not shown) or an image storage memory. An apparatus that can accurately perform an autofocus operation based on the binary image signal will be described below. I will describe the configuration of.

In this embodiment, the number of changing points of the binary image signal, that is, the number of rising or falling edges (hereinafter referred to as focus information)
In the state where the lens is moved to the vicinity of the in-focus position and the out-of-focus position based on the principle that the optimum focus adjustment position is obtained when the count value shows the maximum value in the scanning of the same line. By changing the threshold value, information that changes the count value of the focus information is obtained, and from this information, the threshold value is set so that the count value becomes the maximum value, and the autofocus operation is performed.

Still to explain the principles described above by a 11A view and a 11B view, wherein the first 11A graphs show signal binarization stage when the lens 12 is in the focused position, l ₁ in now (I) - When the sensor scans between l ₂ 's, an image signal as shown in (II) is obtained, and when this is binarized with a threshold value TL, an output signal as shown in (III) is obtained.

On the other hand, FIG. 11B shows the case where the lens 12 is at the non-focus position and the image signal as shown in (II) is obtained when the sensor scans between l _{1 and} l ₂ in (I). By binarizing this with the same threshold value TL as above, an output signal as shown in (III) is obtained.

Thus, comparing the waveform of (III) in FIG. 11A and the waveform of (III) in FIG. 11B, (II) in FIG.
Although the rising edge is obtained as l ₁ -l _{7 in} I), there are three edges e ′ ₁ , e ′ ₄ and e ′ ₆ in (III) of FIG. 11B. Therefore, if the number of edges in the binary image signal is counted, it can be determined that the lens is at the in-focus position when the number of edges is maximum in one scan.

Returning to FIG. 1 again, the explanation will be continued. In the system controller 17 including the central processing unit CPU, the driving signal is supplied to the lens driving device 18 so that the imaging lens 12
Can be moved in the direction of the arrow to change its lens position, and at the same time, an image reading instruction signal is supplied to the image sensor drive circuit 19 and an image at that lens position is given to the image sensor 13 by the scanning signal from the drive circuit 19. Can be read.

Therefore, in this example, the position of the imaging lens 12 is gradually changed by the drive signal from the system controller 17, and the binary change point on the scanning line of the same line read by the image sensor 13 is binary at each position. Change point counter 20
The maximum value of the count value is obtained by the following procedure by supplying the count value from the counter 20 to the system controller 17 so as to be taken into.

When the threshold value setting signal is supplied from the system controller 17 to the threshold value setting circuit 16, the threshold value can be changed according to the instruction signal, and such a threshold value can be continuously supplied to the comparator.

Hereinafter, the operation procedure performed by the system controller 17 will be described in principle. FIG. 3A and FIG. 3B are diagrams for explaining the relationship between the image signal and the threshold value when the focus is just in focus and when the defocusing is out of focus. If Tb, Tc, and three levels are set, the count value of the focus information obtained from the binarized image signal is set when the most appropriate threshold value Tb is set at the time of focusing in FIG. 3A. Largest and first
At the time of non-focusing in FIG. 3B, the above-mentioned count value becomes small even at this threshold value Tb.

Further, when the threshold value Ta is higher than this, the count value is small even at the time of focusing shown in FIG. 3A, and when the threshold value Tc lower than Tb is set, the count value is shown in FIG. 3A. Similarly, the count value does not change so much both at the time of focusing and at the time of non-focusing of FIG. 3B, and both are small values.

FIGS. 4A to 4C show the relationship between the lens position and the count value for each of the threshold values Ta, Tb and Tc. That is, when the threshold value is set to an appropriate threshold value Tb, the peak of the count value is large and sharp as shown in FIG. 4B, and the maximum value point of the peak is easily discriminated. Can be accurately determined. Further, in the case of the threshold value Ta, the peak is small as shown in FIG. 4A, and in the case of the threshold value Tc, the peak is gentle as shown in FIG. 4C, and in any case, it is difficult to find JP accurately. I understand.

Therefore, before executing a series of autofocus procedures, the system controller 17 first sets a threshold value via the threshold setting circuit 16 at a preset lens position (empirically obtained near focus position). The count value is read while changing the value. This lens position is indicated by X in FIG. However, as shown in Fig. 6A, the data string of the count value for the threshold value obtained by scanning at this time is shown in Fig. 6A. And

Next, the lens position is moved from X by a constant value a1 to defocus the image. Then, while changing the threshold value again, the count value data is taken in the same way, and the data string as shown in Fig. 6B is obtained. Is obtained.

Further, next, in FIG. 5, a constant value in the opposite direction from X
Move by a2 to defocus, and in this state, take the count value data while changing the threshold value in the same way as the previous time. Is obtained. Then, the data string is calculated From data column Counted data and data string From data column The data string as shown in FIG. 7 by adding the count value data obtained by subtracting Ask for. Now this data string Looking at the graph of, the threshold value with the largest count is
SL, that is, the peak of the count value is significantly high when the focus is in focus, so even if you try to find the difference from the count value when it is out of focus, the peak of the count value on the graph becomes sharp. I understand. Note that FIG. 8 shows a graph of the count value with respect to the lens position.

That is, when the threshold value having the maximum count value in FIG. 7 is SL and the values before and after it are SL1 and SL2, respectively, when the threshold values are set to SL, SL1, and SL, respectively,
The count values for the lens position are shown in FIG. As is clear from this figure, the peak of the graph when the threshold is set to SL is the sharpest, and therefore when the threshold is set, the highest focus position JP is obtained, and the accuracy is the easiest. You can ask well.

System controller for autofocus
The operation procedure of 17 will be further described with reference to FIG. First, in step S1, the imaging lens 12 is set at a predetermined position,
In this state, the threshold values are changed via the threshold value setting circuit 16 while taking in the count values via the counter 20 in steps S2 and S3, and the count values are taken in with respect to all the threshold values in step S4. If the count values for all the threshold values are obtained, the process proceeds to step S5 and the imaging lens
12 is defocused by a predetermined value a1 in the positive direction.

Thus, in this state, steps S6 to S8 repeat the steps S2 to S4, and if count values for all threshold values are obtained, the process proceeds to step S9, where defocus is performed in the negative direction. .

Next, the steps S10 to S12 are repeated by repeating the steps S6 to S8, and the data strings of the count values obtained by the respective steps up to S4, S8 and S12. From in step S13 Is calculated, and in step S14 Finds the threshold value SL that maximizes, and sets the threshold value to SL in step S15.

Next, in step S16, the binary change point is newly counted and captured by the counter 20 in the state where the threshold value is set to SL, and further, in step S17, the imaging lens 12 is moved, and in step S18, everything is changed. Repeat steps S16, S17 until the count value is obtained at the lens position of, and if the count value at all lens positions is obtained, proceed to step S19, and obtain the largest count value from the results obtained by step S18. Find the lens position JP that becomes
At, the lens is moved to this lens position JP, and thus the autofocus operation is completed.

Note that the sensor described as an image sensor in the above description may be a solid-state sensor or an image pickup tube, and may be a one-dimensional sensor or a two-dimensional sensor to obtain the same effect by the same procedure as described above. You can Further, instead of driving the imaging lens, other lens, lamp, microfilm or image sensor may be driven for focusing, and the above operation is performed instead of using the CPU for the system controller. It is also possible to configure using only executable hardware. Further, it can be applied to reading of 35 mm film, X-ray film, etc. in addition to microfilm.

〔The invention's effect〕

As described above, first, the threshold value is changed at the in-focus position and the non-focus position to measure how the count value changes, and the difference between the count values at each position is the maximum value. By obtaining the threshold value, the optimum threshold value for performing the autofocus operation can be obtained by this threshold value, so that the lens can be reliably set to the in-focus position by the autofocus operation. The accuracy and success rate of autofocus can be remarkably improved as compared with the conventional case.

Furthermore, according to the present invention, not only is there a feature that auto-focusing can be performed without setting a threshold value in advance, but since the threshold value is set to the most appropriate value, focusing accuracy and success rate are improved. Since this is high, it is also possible to apply this device as it is to an automatic density adjusting device.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an example of the configuration of an autofocus device of the present invention, FIG. 2 is a flow chart showing the operating procedure, and FIGS. 3A and 3B are binarization of image signals. FIG. 4A is a waveform diagram showing the form of an image signal at the time of in-focus and at the time of out-of-focus, respectively, in order to explain the state in which the number of rises or falls of the binarization changes depending on the threshold value. FIGS. 4B and 4C are characteristic curve diagrams showing the relationship between the lens position and the count value when the threshold values are Ta, Tb and Tc in FIG. 3A, and FIG. 5 shows the present invention. FIGS. 6A to 6C are schematic views of an optical part for explaining a state in which the imaging lens is moved, and FIGS. 6A to 6C are data diagrams of three modes of threshold value-count value obtained during the operation procedure in the present invention. , FIG. 7 shows 6A in the operating procedure of the present invention. 〜Characteristic curve diagram of threshold value obtained from the result of FIG. 6C, FIG. 8 is a characteristic curve diagram showing as an example the relationship between the lens position and the count value obtained based on FIG. 7, and FIG. 9 is FIG. 10 is a perspective view showing an example of the configuration of a conventional autofocus device, FIG. 10 is an explanatory diagram showing an example of the configuration of a conventional camera autofocus device, and FIGS. 11A and 11B are when the lens is at the in-focus position. 3A and 3B are explanatory diagrams respectively showing a process of binarizing an image signal at a non-focused position. a: light source, 10: micro film, 11: condenser lens, 12: imaging lens, 13: image sensor, 14: amplifier, 15 ... comparator, 16 ... threshold setting circuit, 17 …… System controller, 18 …… Lens drive device, 19 …… Image sensor drive circuit, 20 …… Binary change point counter.

Claims (1)

[Claims] 1. A sensor comprising a plurality of light receiving elements arranged in rows or on a plane, an optical means for projecting an image on the sensor, and a means for binarizing an image signal from the optical means.
Threshold value changing means for changing the threshold value input to the binarizing means, means for counting binary change points from the binarized image signal, and the optical means based on the count value from the counting means. Means for adjusting the focus of the means, and prior to the adjustment of the focus, the threshold value is changed by the threshold value changing means at a position in the vicinity of the focus and a plurality of non-focus positions, and the counting means is provided. The binary change points with respect to a plurality of threshold values at each position are counted by the above, and a threshold value is calculated so that the difference between the count value at the near-focus position and the count value at the out-of-focus position becomes the maximum value. And a means for adjusting the focus after setting the threshold value.