JPS58135411A - Distance measuring device - Google Patents

Abstract

PURPOSE:To increase the detecting range of distance and to improve the measuring performance of distance without having a large and complicated electric circuit system, by dividing a distance section to be detected into plural regions and providing plural distance measuring systems to each divided regions. CONSTITUTION:Mirrors MR1 and MR2 are set with a prescribed length of base line secured between them, and the standard and reference optical systems are formed with distance measuring lenses LS1 and LS2, mirror faces MR3 and MR4 of a prism PR, and an image sensor IS. Then a measuring distance section is divided into two regions S and R. The output of the sensor IS is fed to the regions S and R via a quantizing circuit QZ under the control of a control circuit SQ. The quantized signals of the regions S and R are supplied to distance measuring arithmetic circuits EF1 and EF2 respectively. The coincidence or discordance is detected for the picture element of a standard viewfield image obtained through a sensor region S to (M+L) units of continuous picture elements of a reference viewfield image obtained through sensor regions R1 and R2. Thus the distance measuring information is delivered. Therefore the distance measuring performance is improved as the detecting range of distance is increased.

Description

[Detailed description of the invention] The superposition is for the distance measuring device.

For a distance measuring device that detects the distance between one cup and C, use the bladder.
-There are many good ideas, some of which are put to practical use in camera weddings. Among such fibrillation #, for example, there is no -1 constant rLk M of 11 moving parts.
By detecting the relative positional deviation gIkm between 2tIk of an object formed with a relative positional deviation corresponding to the distance 1111c, the distance can be detected by There is.

Here, the new Corriage Sohn method is used to detect the relative positional shift between the two images, and for example, for one image that should be used as the d% reference, continuous Mill I pixels are
information, and other information to be referred to (i.e.,
At any position among the M continuous pixels with respect to the reference visual field image, there is an image portion consisting of N consecutive pixels with respect to the reference visual field image. Object distance is detected by detecting whether there are image parts that can be considered to be substantially the same. When performing a colliage calculation on the information chest of (mK#, long word length), if all combinations of each member are performed in parallel, the circuit size will increase. As the word length of the reference information becomes longer, and as the difference in word length between image information becomes larger, the size becomes extremely large. Therefore, it is possible to easily classify a wealth of information using a small-scale circuit. To make it a bit pie
Depends on the bit! This is a station where you have to settle for L Gaishuku processing, but in recursion and series source processing, base 4i! As the length of the information becomes longer, and as the difference in word length between image information becomes larger, the time required to complete the processing becomes longer, and in the end, it becomes difficult to perform rapid distance measurement. I'm worried that being too proud of my camera will cause inconvenience when taking pictures.

In the case of a long-range weapon V, which has just been disembarked, lengthening the length of each throw of the reference information will lead to improved ability and discrimination of objects, and also, as you, the head of the information cabinet, Making it larger will lead to subdivision of the detection distance zone and expansion of the detection distance range to the short distance #M koji. Therefore,
↑ - When introducing serial processing of information, it was found that the performance of the fibrillation device was immediately affected by the inability to set the length of the information and the length of the ki between the images due to processing time. This can lead to a decline in the energy consumption.

The present invention was made in view of the above circumstances, and as a fibrillation device, it does not require the configuration of the electric circuit system to become complicated and bulky, and the processing time is also significantly shortened. The object of the present invention is to provide a distance measuring device ii of the present invention with the aim of significantly expanding the range of distance to be detected and improving the distance measuring performance. divides the detection target distance m range into a number of chain plates, and each
By setting up multiple distance measuring systems corresponding to each of these, and operating them in parallel, the distance to the object can be determined based on which position in which area the object is located. It is characterized by the fact that it is also V.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

First, referring to FIG. 1, in the same figure, MR.

and M n, jt i & force direction predetermined baseline length Vrfi
! A pair of mirrors arranged diagonally at an elevation angle of 45 degrees, L
B, and Li3. tj, respectively, a pair of distance measuring lenses that receive each reflected light from mirrors MR and MR, and form an image, PRn-respectively lenses X
LS, and L3. The optical system has 45 mirror surfaces MR and MR for reflecting the image light from IS is the original thread MR, -LB, -Ml, and MR, -XaB, -M, respectively.
A linear image sensor such as a 00D or BjjD7 sensor or a Mouse OS image sensor is used to generate a time-series scanning electrical signal by scanning the detected image for distance measurement obtained through R. , in this example, the optical system M
For the whale image obtained through R, -LB Bi1R1, the M-bit sensor element is indicated by 8 in the figure, and for the image obtained through optical JMa, -18t -MR4, it is also indicated by R. The recommended sensor elements (where L s N =M+2L) are arranged for both optical systems. Here MIL, −
The optical system consisting of LS, = MIL is applied as a reference (sighting) optical system to always aim at an object on its front axis in this embodiment. The system fires the image on one sensor area s in such a way that its center coincides with the center of the area 5q)cFJ, and thus a reference visual field signal is obtained from the area 7+i area #c8.

On the other hand, the optical system consisting of MR, -L8ffi-MR4 is
As a reference optical system for aiming the target object using the above-mentioned basic optical system MR, -LB, -MR, and moving it at a position offset by a predetermined baseline length J with respect to the reference optical system. Applied, the spatial image containing the object is placed on the center area by the reference optical system according to the object distance, and the closer the distance is, the reference i! Shape r & g away from the statue of Shu.
It will be done. In ζζ, the sensor area shaped by the 1st to Mi1th element fψ of the sensor area is the same as the relationship of the sensor area B to the reference optical system Ml, -L8□-MljK, and the reference optical system Midl. -18, - The optical axis of MR4 is arranged with respect to the reference optical system so that it coincides with its center, and the position of the sensor element with title l, which is the end of the senna region R, is located at the detection distance. Determine the nearest limit of range. 41 is image φ sensor 8
Quantum for performing analog processing and quantization of the scanning output from 11. is the quantized signal corresponding to sensor area 8 and the l-M tenth (
Accept the quantized signals corresponding to the sensor area formed by M+L) sensor elements, and calculate the O distance-range from the finite distance to infinity based on these two quantized signals. A distance measurement calculation circuit that performs distance measurement calculation for the target, RF, tj, and a quantized signal corresponding to the 7th area 8, and a sensor? 0 out of the area R. L+l~1t(=mouse+2'
Accept the quantized signal corresponding to the sensor area - formed by the L)th CM+L) sensor elements, and calculate from the above distance range based on both of these quantized signals. This is a distance measurement calculation circuit that performs large distance measurement calculations targeting the distance range on the short distance side. Here, the distance measurement calculation circuit RIF
, and iiy, a, e.g. JP-A-Sho b pertaining to one applicant.
5-115023, and as illustrated in No. 21iI, the sensor area mouse! and reference visual image m and l by 1 mK
For each (M+L) number of pixels, the continuous pixels t"q of Mil in the reference visual field image by one sensor area 8 are shifted by a predetermined number of primes every K18th order, and the pit, pie, Detect matches and mismatches using (1-prime pair l-volume), and finally, store the maximum-dispersion in the maximum matching pixel number memories MOO and MOO, respectively, and generate this maximum matching pixel number. The positions of the M continuous pixels in each sensor region R8 and - are stored in the best-number position memories MOL and MO, respectively.
L, and the position memory MOL.

It has a function of outputting the memory contents of and MOL as distance 1 information. In addition, position memory MOL.

and moL, 6 (the data to be stored is given by counting the reference visual field image 1 and the amount of the above shift-prime number with respect to the standard visual field image, but the memory MOL in the ranging calculation circuit 1 has The data obtained by adding a constant @V to the counted and shifted pixel number is stored. That is, this constant ``''r is calculated by adding the constant @V to the number of pixels shifted.In other words, the constant ``''r is calculated by adding the constant @V to the number of pixels counted and shifted. It is easy to understand that ゛ corresponds to the target distance axis III and the distance between the two distances.
．． and a size discrimination circuit D8 for determining the magnitude relationship between the contents of each maximum matching pixel in each wing, Mool and MOO, in response to the discrimination output of the size discrimination circuit Do, The distance measuring calculation circuit ity or RN which is determined to be larger.

Best-number position memory MOL, or Mol, O in
a data selector DTlj that selects output data;
Although it is by no means essential, the data selector DB
The position memory MOL selected from the MOL, or MOL,
Since the distance information given by the contents of - A distance S signal is output through the table DT. In addition, 8Q is a control circuit that controls the sequential plowing of the signal generation and processing system explained above, and special distance calculation - path my and ... IF□ perform distance measurement calculation operations in parallel in time. It will be controlled as you do it.

Thus, here, two distance measuring systems (B `` ``1t ``t '' -4 肱4
) are operated in parallel and the detection results of both systems are judged to determine the object distance.

Incidentally, when determining the size 111 between the contents of the maximum matching pixel number memories MOO1 and MOO, by the size discrimination circuit no, if the same result is obtained, the object is
Location memo 'J M
Even if the contents of the OL and MO-○ deviations are used as distance information and are used as distance information, there will not be much difference.

The configuration of the quantization circuit QZ is also based on the above-mentioned Japanese Patent Laid-Open No. 56
-115023 Publication, regarding the details, it is possible to avoid making the configuration of the electric circuit system complicated or large in any way, and the processing time can be significantly shortened, and the distance range to be detected is This is extremely useful for distance measuring devices, as it allows for a large expansion of the field of view and an improvement in distance measuring performance.

In addition, as an example, only the case where the number of divisions of one detection target distance range is 2 is shown, but this number of divisions can be selected as necessary and within a feasible range. #I don't want to talk to you.

[Brief explanation of drawings]

#! Figure 1 is a block diagram showing an embodiment of superimposition, and Figure 2 is a block diagram showing the operation contents of the two distance measurement calculation paths in Figure 1.
11r is a schematic diagram. MR, ~MR, are mirrors or mirror surfaces, Lli, and LS! is an imaging lens,
7>t, StR; TL, B, /4 sensor area, Quit
Quantization path, RIF, and RF are ranging calculation circuits, DO
ij size discrimination circuit, DB is the data selector, 8-kasu, RF, the distance measuring thread of cram school 1, 8-1! , RF, is a patent featuring a second ranging thread.Gi Marushima, representative of Canon Co., Ltd. -

Claims (1)

[Claims] It is a device that detects the distance to an object, and it divides the detection target distance range into areas of the number of wheels, sets a distance measuring system of m number corresponding to each area, and parallel to this. 1. A distance measuring device characterized in that the distance measuring device is operated to specify a distance to an object based on which position in which region the object is located.