JPH11201717A - Object recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve range finding accuracy by utilizing the defocusing characteristic of an optical system. SOLUTION: The difference A-B between the detection signal amplitude A of a zero-cross detector when the focal position is shifted from a short range to a long range as shown in Fig. (a) and the detection signal amplitude B of the zero-cross detector when the focal position is shifted from a long range to short range as shown in Fig. (b) is found from the relation between 'focal point drift' and 'defocusing of image' to detect a zero-cross point t0 as shown in Fig. (c), and distance can be accurately measured.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object recognizing device for precisely measuring and outputting the positional relationship of an object existing in the front-back direction.

[0002]

2. Description of the Related Art Conventionally, there have been various methods for three-dimensional image measurement as shown in FIG. 9. Iguchi: Recent trends in three-dimensional shape measurement, "Measurement and Control," Vol. No.19
June 1995, pp. 429-434.

[0003]

However, in the above-mentioned focus adjustment method, the blur must be increased in order to accurately determine the focus position. For this purpose, a lens having a very small depth of focus and a large aperture is required. I need. In a lens system like a microscope, the depth of focus is extremely shallow,
A micro height gauge utilizing this principle has been put into practical use,
It is difficult to achieve a sufficiently shallow depth of focus for measurements of several meters or more, and it is difficult to obtain satisfactory measurement accuracy. Iguchi and Sato: 3D image measurement, Shokodo 199
It is pointed out in the first edition, November 19, 2000, pages 18-19.

An object of the present invention is to solve the above-mentioned problems,
It is an object of the present invention to provide an object recognizing apparatus in which distance measurement accuracy is improved by using a blur characteristic of an optical system by combining an optical system for changing a focal position and an image processing system for output of photoelectric conversion.

[0005]

The object recognizing device according to the present invention for achieving the above object is characterized in that the positional relationship of an object is measured by utilizing the focus adjustment of an optical system combined with a lens. I do.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail with reference to the embodiments shown in FIGS. FIG. 1 shows a block diagram of a block circuit according to the first embodiment. In an optical system 1, an objective lens 2, an eyepiece 3, and a photoelectric converter 4 are arranged on an optical path O. The output of the photoelectric converter 4 is connected to a band-pass filter 6 and an output terminal 7.
It is connected to the. The output of the band-pass filter 6 is connected to a display 8 having a display memory characteristic, a memory / image processing unit 9, and a memory / image processing unit 9.
Output is a zero-cross detector 10, a time / position converter 1
1, the output computing unit 12 and the output terminal 13 are sequentially connected.

The output of the lens moving mechanism 5 is sequentially connected to a lens position encoder 14, a horizontal addressing unit 15, and a display 8, and the output of the lens position encoder 14 is connected to a time / position converter 11. ing. The output of the reference clock generator 16 is connected to the synchronizing signal generator 17 and the zero-cross detector 10, and the output of the synchronizing signal generator 17 is sequentially connected to the moving wave generator 18, the lens moving driver 19, and the lens moving mechanism 5. And is connected to the display 8 via a vertical addressing unit 20.

In the above-described configuration, when the optical system 1 is directed to a subject existing in the direction of the optical path O, the subject forms an image on the surface of the photoelectric converter 4 at the position of the eyepiece 3 that satisfies the focusing condition, and is sharp. Images can be obtained. On the other hand, at the position of the eyepiece 3 that does not satisfy the focusing condition,
I get only blurry images. At this time, if the focusing distance is repeated in the order of near → far → near → far across the position of the subject, the focusing condition is satisfied twice during the repetition, and a clear image of the subject is obtained. The relationship between degree and distance L is near →
It is different in the case of far and the case of far → near.

When the signal intensity of a specific frequency component is extracted from the image signal which is the output signal of the photoelectric converter 4 by the band pass filter 6, the signal intensity for the same subject becomes maximum at the time of focusing and decreases as the defocus increases. (Sakano: 12th autofocus technology "Optics")
Vol. 5, No. 5, August 1983, pp. 351-358).

At this time, the relationship between the signal intensity and the distance differs between the case where focusing is performed in the near → far direction and the case where focusing is performed in the direction of far → near. 2A and 2B are explanatory diagrams of the detection of the distance to the subject by moving the lens position. FIG. 2A shows the arrangement of the optical system, and FIG. 2B shows the distance between the output P of the photoelectric element of the photoelectric converter 4 and the subject. FIG. 2C shows a drive waveform when the lens position moves.

Reference clock generator 16 of the entire system
2 (c) is created in the moving waveform generator 18 from the output of the synchronizing signal generator 17 based on the clock signal of FIG. 2 and input to the lens moving driver 19 to drive the lens moving mechanism 5. Then, the eyepiece 3 is moved from near to far to near to far.

On the other hand, the signal output of the photoelectric converter 4 is supplied to a display 8 having a display memory characteristic via a band pass filter 6, and the signal intensity is displayed as a change in luminance or a change in color. At this time, in order to synchronize the horizontal display position on the display 8 with the position of the eyepiece 3, the horizontal addressing unit 15 is driven by the output of the lens position encoder 14. The right end of the screen of the display 8 is made to correspond to the position at infinity.

The vertical address of the display 8 is selected by the vertical address specifying unit 20. That is, when the subject exists at the distances L1, L2, L3 as shown in FIG. 2 (b), a pattern as shown in FIG. In FIG. 3, the width d indicates the sharpness of the image, the arrow f indicates the address designation direction, the left end R indicates the position of the objective lens 2, and the right end ∞ indicates the infinity position of the subject.

Further, the output of the band-pass filter 6 is also guided to a memory / image processing unit 9, where signal processing is performed as shown in FIG. Fig. 4 (a) shows the eyepiece lens 3 close.
FIG. 4B is a graph of the output amplitude signal A of the zero-crossing detector 10 when moving far, FIG. 4B is a graph of the output amplitude signal B when moving far to near, and FIG. FIG. 4 shows a graph of a difference AB between amplitude signals A and B.

The zero-cross point to obtained by the zero-cross detector 10 shown in FIG. 4C is precisely detected in units of a clock cycle which is the output of the reference clock generator 16, and the time / position converter 11 obtains distance information. Can be At this time, if the subject is stationary, the information of the distances L1, L2, L3 is vertically arranged on the display screen of the display 8 as shown in FIG. When the subject is moving in the direction of the optical path O, the positions of the distances L1, L2, and L3 move left and right,
By calculating the temporal change of the moving amount by the output calculator 12, the moving direction and the moving speed of the movement of the subject can be detected.

The state of the above-described exercise is displayed on the display 8 from moment to moment. If the subject is too close to the present system and there is a danger of collision, a warning is issued or the entire system is isolated from the subject by another drive mechanism. And can be maintained at a fixed distance. Also, distances L1, L2, L3 in FIG.
Information is divided into zones, and a specific color is added to the subject entering each zone. You can also make it.

Thus, "defocus" versus "image blur"
By detecting the "difference" between the case where the focal position shifts from a short distance to a long distance and the case where the focus position shifts from a long distance to a short distance, more accurate distance measurement can be performed. Further, by continuously measuring the time lapse of the “difference”, data on the moving direction and the moving speed of the subject can be collected.

FIG. 5 shows a side view of the second embodiment, and FIG.
The eyepiece 3 and the lens moving mechanism 5 are replaced with a liquid lens 25. The liquid lens 25 is formed of a transparent liquid 26 having a refractive index larger than that of air and flexible circular transparent insulating films 27a and 27b, and the transparent liquid 26 is covered with insulating films 27a and 27b adhered at the peripheral edges. And the insulating films 27a and 27b are covered with transparent electrodes 28a and 28b each having a predetermined area. The liquid lens 25 is attached to the inner walls 30a, 30b of the optical system 1 via circular bellows 29a, 29b.

FIG. 6 shows a front view of the liquid lens 25 and its driving means. The transparent electrodes 28a and 28b have a central electrode 31 at the center, a ring-shaped electrode 32 outside thereof, and semicircular electrodes 33a and 33b outside. The outermost quarter-arc electrodes 34a, 34b, 34c, and 35d are provided. The electrodes 31, 32, 33a, 33b, 34a to 34d have
Lead wires 35, 36, 37a, 37b, 38a respectively
To 38d are connected. And each lead wire 35 ~
The output of the lens control voltage generation unit 39 is connected to 38d, and the output of the moving waveform generator 18 is connected to the lens control voltage generation unit 39. Other configurations are the same as those in FIG.

In the above-described configuration, a charge q1 is applied to the transparent electrode 28a on the insulating film 27a side via the lead wires 35 to 38d to the liquid lens 25, and the transparent electrode 2 on the insulating film 27b side is provided.
When a charge q2 is given to 8b, a force F expressed by the following equation acts between the insulating films 27a and 27b according to Coulomb's law. F = (1 / 4πε ₀ ) · (q1 · q2 / r ² ) (1)

Here, ε ₀ is the dielectric constant of the transparent insulator portion sandwiched between the electrodes, and r is the distance between the electrodes.

Therefore, the lens control voltage generating unit 39
By controlling the amounts of the electric charges q1 and q2, the curvature of the curved surface formed by the insulating films 27a and 27b can be changed, and the focal length of the liquid lens 25 can be changed. Thereby, from the case where the focus position of the optical system 1 is at a close distance, for example, as shown by the ray trajectory R1 of FIG. 5, to the case where the focus position is at infinity, as shown by the ray trajectory R2. Focusing can be controlled.

As described above, the operation speed can be improved by removing the eyepiece position moving mechanism 5 from FIG. 1 and electrically performing the function of the mechanism 5 by the liquid lens 25. At the time of moving speed measurement, speed measurement on a higher speed side than in the first embodiment becomes possible.

FIG. 7 is a block diagram of the third embodiment, and FIG. 8 is an explanatory diagram of stereoscopic image collection by the depth sampling method.
Each output of the image forming apparatus 40 is connected to an input terminal of an image display device 42 via a transmission path or a recording / storage device 41. A subject S1, a subject S2 at a short distance, a subject S2 at a medium distance, and a subject S3 at a long distance sequentially exist in front of the observer's head H or the optical system 1 in FIG. From the eyeball position E of the observer H or the object S1 from the objective lens 2 of the optical system 1,
Distances L1, L2, and L3 to the front surfaces of S2 and S3, respectively.

In the above arrangement, when collecting stereoscopic image information by the depth sampling method, conventionally, distances L1 to L3
Although sampling is performed at equal intervals in the direction, it is not necessary to sample the shadows where the subjects S1 to S3 do not exist or are not seen by the observer H, for example, the spaces of D1 and D2.
Also, as for the visible part, it is not necessary to sample a part having no unevenness and a low spatial frequency, and it is possible to interpolate from the preceding and following sample values to make up for it.

Therefore, sampling is performed at non-uniform intervals as indicated by dotted lines in FIG. 8, and the sampling plane is determined based on the distance information according to the first embodiment. That is, image information of each sampling plane is created based on the distance information, and is reproduced on the image display device 42 via the transmission path or the recording / storage device 41 together with the distance information.

The sampling planes 1, 2, 3,..., J,.
The image information of m and n and their respective position information are mixed in the image forming apparatus 40 with the image information from the output terminal 7 and the position information from the output terminal 13 in FIG. , J,..., M, n terminals, and reproduces a stereoscopic image on the image display device 42 via the transmission path or the recording / storage device 41 or directly.

As described above, unlike the conventional depth sampling method in which the space is cut at a uniform distance difference, while reducing the total number of sampling surfaces, only the surface on which a subject having a large amount of stereoscopic information exists, or By performing dense sampling only on the irregular surface of the irregularities, the total amount of information to be transmitted or recorded / stored can be reduced.

[0029]

As described above, the object recognition apparatus according to the present invention utilizes the focus adjustment of an optical system in which a lens is combined with an object to accurately measure a positional relationship in units of a clock frequency cycle. be able to.

Further, the arrangement of the object and whether the object is stationary or moving can be detected and displayed, and the present invention can be applied to the setting of a sampling plane when forming a three-dimensional image by the depth sampling method. it can.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a first embodiment.

FIG. 2 is an explanatory diagram of distance detection to a subject by moving a lens position.

FIG. 3 is an explanatory diagram of a display pattern on a display.

FIG. 4 is an explanatory diagram of the operation of the zero-cross detector.

FIG. 5 is a side view of a liquid lens according to a second embodiment.

FIG. 6 is an explanatory diagram of a liquid lens driving unit.

FIG. 7 is an explanatory diagram of the connection of the image forming / display device of the third embodiment.

FIG. 8 is an explanatory diagram of stereoscopic image collection by a depth sampling method.

FIG. 9 is an explanatory diagram of a conventional three-dimensional image measurement method list.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Optical system 2 Objective lens 3 Eyepiece 4 Photoelectric converter 5 Lens moving mechanism 6 Bandpass filter 8 Display 9 Memory / image processing unit 10 Zero cross detector 11 Time / position converter 14 Lens position encoder 15 Horizontal addressing unit 16 Basic clock generator 17 Synchronous signal generator 18 Moving waveform generator 20 Vertical addressing unit 25 Liquid lens 39 Lens control voltage generating unit 40 Image forming device 41 Transmission path or recording / storage device 42 Image display device

Claims (8)

[Claims] 1. An object recognition apparatus characterized in that a positional relationship of an object is measured by using focus adjustment of an optical system combined with a lens. 2. The object recognition apparatus according to claim 1, wherein the focus adjustment is a focusing degree of the optical system capable of focusing from a close distance to an infinite distance. 3. The object recognition apparatus according to claim 2, wherein the fact that the degree of focusing changes with the distance to the object is used. 4. The object recognition device according to claim 3, wherein the degree of focusing is different between the near side and the far side of the object position. 5. The object recognition apparatus according to claim 2, wherein the focusing is performed by a variable focus liquid lens. 6. The object recognition device according to claim 1, wherein the position of the object is displayed on a display to output a stationary state or a moving state of the object. 7. The object recognition device according to claim 6, wherein the moving state of the object is a moving direction and a moving speed. 8. The object recognition apparatus according to claim 1, wherein the positional relationship is position information obtained by three-dimensional image measurement using a depth sampling method.