JP3806473B2 - Ranging device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a distance measuring device used for a camera, a video, and the like, and more particularly to a technique for providing a device capable of more accurate distance measurement and focusing.
[0002]
[Prior art]
Usually, distance measuring devices used in cameras, etc., often use light, and the “active type” method that projects signal light from the device side and the “passive type” method that uses the luminance distribution image of the object. ”. Both of these are based on the basic principle of triangulation. In the active type, the distance between the light projecting and receiving positions is the basic length (base line length), and in the passive type, the image of the target object is based on the two light receiving positions. The distance to the object is obtained according to the relative position difference. In order to observe the image of the object, it is necessary to measure the amount of light at each light receiving position. Therefore, a line sensor in which a plurality of optical sensors are arranged is used.
[0003]
First, the distance measuring method of this passive type distance measuring device will be described with reference to FIG. Two light receiving lenses 1a and 1b are spaced apart by a base line length B, and light incident from the subject 6 through these lenses has a light pattern such as a curved pattern 3a and 3b on the line sensors 2a and 2b, respectively. However, these patterns depend on the luminance distribution on the subject 6 and the relative positional relationship between the two light receiving lenses, and the relative positional difference x between the two patterns 3a and 3b is the distance between the base line length B and the distance between the lens and the sensor. Depends on f and subject distance L. That is, the following relationship is established.
x = B × f / L (formula a)
A method of calculating the relative position difference by calculation from the output of such a line sensor (hereinafter referred to as correlation calculation) is disclosed in Japanese Patent Publication No. 7-54371.
[0004]
Further, it is an important technique to use which portion of the light pattern to measure the distance. For example, as shown in FIG. 5, even if the person 6a is to be measured, it enters the Rn2 portion of the sensor array 2a. If the added light is taken into account, the luminance of the background mountain 6b is mixed (hereinafter referred to as “mixed perspective”), and correct distance measurement cannot be performed. That is, only the light incident on the sensors R1 to Rn1 in the figure is used, but the sensor signal of Rn2 is not used. For example, the technique disclosed in Japanese Patent Publication No. 5-88445 is necessary for accurate distance measurement. It becomes.
[0005]
Further, according to the focus detection apparatus disclosed in Japanese Patent Laid-Open No. 2-135311, the data change of each data of the line sensor is examined, and a signal corresponding to a place with a small change rate is used for distance measurement. ing.
Furthermore, a technique for switching a distance measurement frame according to the focal length of a camera photographing lens has been widely known for a long time. For example, a distance measurement range display of a camera performing automatic focus adjustment disclosed in Japanese Patent Laid-Open No. 53-99935 is disclosed. There is also a technique for preventing the above-described error from occurring by setting a fixed distance measuring area in the screen of the shooting scene.
[0006]
[Problems to be solved by the invention]
However, first, the technique disclosed in Japanese Patent Publication No. 7-54371 does not touch on the above-mentioned problem of “mixing of perspective”. In the other technique disclosed in Japanese Patent Publication No. 5-88445, this result is obtained after the correlation calculation. If the sensor area is inappropriate, the sensor area is switched, and because of the determination and switching control, it takes a lot of time and the data is processed and processed many times. However, there are problems such as an increase in memory capacity and program capacity. That is, even if these conventional techniques are simply combined, it is difficult to provide a high-speed and high-precision distance measuring device.
[0007]
In addition, in the operation of the conventional apparatus, an error is also caused by the distribution of light and darkness of the subject to be measured, and if this error is not taken into account, not only does the subject have a distance measurement error even at the same distance, Alternatively, a distance measurement error may occur depending on the portion of the subject. Also, in general, it is not always possible to output the distance measurement results as designed due to the "variation" in the components used in the distance measurement device and their assembly. Therefore, an adjustment process is required to deal with this problem during manufacturing. It is. However, even in this adjustment process, an accurate and stable output may not be obtained due to a slight shift in the positional relationship between the adjustment chart and the distance measuring device. This is because correct adjustment is difficult unless the output is stabilized, and as a result, it is impossible to provide an accurate distance measuring device. However, this problem has not been taken into consideration in the above-described prior art.
[0008]
Accordingly, the present invention has been made in view of the above-described problems, and provides a distance measuring device that can always perform a correct distance measurement for any object. To that end, a passive distance measuring device is provided. It is an object of the present invention to provide a distance measuring apparatus that takes measures to remove the dependence on a specific pattern (for example, a chart) possessed by the target object (that is, a specific “characteristic” based on a subject). In addition, the present invention proposes a distance measuring device that is resistant to the above-mentioned errors of mixing near and far, and provides a distance measuring device that can accurately focus on an object when applied for focusing on a camera or video. It is an object.
[0009]
[Means for Solving the Problems]
  To achieve the above object, the first invention is a distance measuring device, wherein a light receiving means for receiving a subject image by a pair of light receiving element arrays having parallax and outputting a pair of subject image signals; Selection means for selecting one of a plurality of areas divided into a predetermined size for each of the pair of subject image signals, sensor signal levels of light receiving elements at both ends in the selected area, and light receiving elements at both ends A change means for obtaining a difference from the sensor signal level of each adjacent light receiving element, and changing the size or position of the selected area so that a calculation result of the difference satisfies a predetermined condition; and Correlation means for performing correlation calculation and interpolation calculation using the sensor signal of the changed area.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
  A plurality of embodiments according to the present invention will be described below with reference to the drawings.
  First, as a premise,FIG.Next, a method for calculating the relative position difference of the light pattern in the above-described passive type distance measuring device will be described.
  Figure1The relative position difference x of the light distribution incident on the sensor arrays 2a and 2b changes depending on the subject distance L due to the difference B (base line length) between the positions of the two light receiving lenses 1a and 1b. If the focal length of each light receiving lens is f, the subject distance L is obtained by the following equation.
          L = B × f / x (Formula b)
  Since each sensor of the sensor array outputs a current signal according to the amount of incident light, if these are converted into digital signals by the A / D converter 4, the relative calculation is performed by the correlation calculation by the calculation means 5 for calculating the image shift amount. The position difference x can be detected. The subject distance L is obtained by inputting this result into a calculation control means (CPU) 10 comprising a one-chip microcomputer or the like and calculating based on the above equation b. The above is the basic principle and general apparatus configuration of the passive type triangulation system.
[0011]
The above deviation amount calculation function is generally composed of two processes as will be described later, but these may be built in the CUP 10 as a control program. When the camera is focused using such a technique, the CPU 10 controls the operation of the camera, and an automatic focus (AF) function can be achieved by appropriately controlling the photographing focus lens and the like via an actuator such as a motor. Can be provided with a camera.
[0012]
In order to calculate the image shift amount, a calculation step (that is, correlation calculation) for checking how much the image is shifted in units of sensor pitch in both line sensors is required. Then, a calculation step (hereinafter referred to as interpolation calculation) for calculating the shift amount more accurately with a finer resolution is required.
[0013]
  Figure on sensor array 2a1When the light is incident in a pattern like the waveform 3a shown in FIG. 2, the output magnitudes of the sensors R1 to R6 have a distribution 3a as shown by a bar graph in FIG. Here, “R” indicates a right sensor, “L” indicates a left sensor, and subscripts 1 to 6 attached thereto are, for example, optical axis reference of the light receiving lens. Suppose that the absolute position of the sensor position at is indicated. The same signals as the outputs R1 to R6 are output from the outputs L1 to L6 of the left sensor.OutputIn this case, since the relative position difference x is 0, the subject distance L to be obtained is “infinity”.
[0014]
If the subject is present at a "finite distance", the left sensor L shifted by the number S of sensors determined from the x and the sensor pitch SP has the outputs R1 to R6 as shown in FIG. A similar value of the output signal is obtained.
The value FF (i) on the vertical axis in the graph of FIG.
[0015]
FF (i) = Σ | R (i) −L (i) | (Formula c)
In other words, the result FF obtained by subtracting the output of the corresponding L sensor from the output of a certain R sensor and adding the absolute value for each sensor may be used. That is, first, Li is subtracted from Ri to take its absolute value, and i is changed by a predetermined width and added.
Next, if one of Ri or Li is shifted by one unit and the difference is taken in the same manner as the adjacent sensor which has previously taken the difference, FF (i + 1) can be expressed by the following equation.
[0016]
FF (i + 1) = Σ | R (i + 1) −L (i) |
In this way, the FF can be obtained while sequentially changing the shift amount (hereinafter referred to as the SIFT amount), but the SIFT amount where the FF, which is the sum of the differences between R and L, becomes the minimum value (Fmin) is the most. Since it is considered that the position is well matched, the SIFT amount in this case is obtained as S. The above is the outline procedure of the process related to the correlation calculation.
[0017]
Further, when the output distribution of both sensor arrays is illustrated with the above S taken into account, as shown in FIG. 2 (b), it is the same as the R side sensors with corresponding subscripts from the sensors shifted by S on the L side. Is obtained.
[0018]
Next, the “interpolation calculation” process will be described in detail with reference to FIGS. The actual image shift amounts on the two sensor arrays are not exactly shifted by the sensor pitch, and for accurate distance measurement, the image shift amount must be detected with a finer accuracy than the pitch. Therefore, an interpolation operation is performed. R and L in FIGS. 2B and 2C respectively represent a part of sensor outputs constituting the sensor arrays 2a and 2b in FIG.
FIG. 2 (d) shows a graph that has been shifted to the above-described S for which the “correlation calculation” has already been completed, and has been made easier to compare. That is, L0 to L4 should be accurately described as Ls to Ls + 4, but this S is omitted so as not to become complicated in description.
Here, it is assumed that light shifted by x with respect to the R reference is incident on the L sensor after shifting by S. At this time, for example, the light incident on R0 and R1 is mixedly incident on the L1 sensor, and similarly, the light shifted by x on the R reference is sequentially incident on each L sensor. It can be seen that the outputs (L1 to L3) are expressed as shown in [Equation 1] in FIG.
[0019]
FF values F-1 and F + 1 obtained by shifting the shift amount from Fmin and Fmin in the plus and minus directions are expressed by using the outputs of the respective Rn and Ln. It is expressed as follows. Further, when [Expression 2] is expanded using [Expression 1], values Fmin, F-1, and F + 1 are expressed as [Expression 3] in FIG.
In addition, if {| R0−R1 | + | R1−R2 | + | R2−R3 |} in [Equation 3] is expressed as (ΣΔR), the previous deviation amount x does not depend on (ΣΔR). Is obtained by the calculation shown in [Equation 4] in FIG. The above is the “interpolation calculation”.
[0020]
These calculations are performed by the calculation unit 5 in FIG. 1, but may be performed by a calculation control means (CPU) 10 such as a one-chip microcomputer according to a predetermined program.
[0021]
FIG. 3 is a conceptual diagram of the configuration of a camera according to the main embodiment of the present invention. This camera is equipped with a distance measuring unit 100 having a basic configuration including the lenses 1a and 1b and the sensor arrays 2a and 2b described above as a distance measuring device. The unit built-in correlation means 5 is constituted by a high-speed hard logic circuit (not shown).
The A / D conversion circuit 4 is a circuit that converts the analog output of the sensor array into digital data. Here, the A / D conversion circuit 4 also has a function of storing the converted data. The output circuit 7 outputs this stored data to the CPU 10 by serial communication. As shown in the figure, the CPU 10 includes a port 19 composed of each port element for serial communication and a terminal for controlling each function of the camera, an arithmetic register 14 for controlling the port, a general-purpose register 18, and an arithmetic unit. 17, a RAM 15 for temporary storage of data, and a ROM 16 storing a program for controlling these with a predetermined algorithm.
[0022]
Through the port 19 from the CPU 10, the focusing lens 13 is fed out by the feeding means 12 that performs feeding control by predetermined control, and the shutter 8 and the like are driven and controlled.
Also, a ROM (EEPROM) 9 is provided via the port 19, and this EEPROM 9 corrects errors caused by variations in camera parts and assembly accuracy, as in an embodiment described later. This is an electrically writable ROM that is used to store a unique correction coefficient. The correction value is detected and written to the EEPROM in the manufacturing line adjustment step (FIG. 14).
As described above, the CPU 10 performs overall control of the camera including distance measurement control at the time of photographing while referring to the correction data.
[0023]
(First embodiment)
First, FIG. 4 is a flowchart showing a schematic processing procedure such as arithmetic processing according to the embodiment of the present invention. In the distance measuring apparatus of the present invention, processing is executed in the following procedure. That is, first, data based on incident light of the sensor array, correlation values in sensor pitch units, and the like are read (S101). Subsequently, data for distance measurement in a predetermined area is read from the left and right sensors by one characteristic method described later. Schematically, the left and right data are not read indefinitely from a certain area and are interpolated. For example, the data of the right sensor at the end of the area | R1 −R0 | and | Rn + 1−Rn | (ie, | R4−R3 |) is compared (S102), and the area is changed by shifting the difference to an area satisfying a predetermined condition (S103), while the other left side The sensor data output is also obtained, and then correlation calculation and interpolation calculation are performed (S104). From this result, the correct distance to the subject can be obtained and used for focusing.
[0024]
In the flowchart of FIG. 7, the processing operation of the camera distance measuring apparatus having the above-described configuration is illustrated in detail as the first embodiment of the present invention.
First, a correlation area is specified. For example, here, A / D conversion is performed on the data based on the light pattern incident on the sensor array by designating correlation areas as shown as R1 to R3 (see FIG. 1D) (see FIG. 1D). S1).
[0025]
The correlation unit 5 reads the calculated correlation result S in sensor pitch units (S2), and reads data for distance measurement in a predetermined area from the R side sensor (S3). As described above, as a technique that is one of the features of the present invention, sensor data | R1 -R0 | (in FIG. |) And | Rn + 1-Rn | (in FIG. 7, | R4 + n-R3 + n |) are compared (S4 to S7). , S13, the calculation is performed after the shift, and with this processing method, F + 1 shown in [Expression 3] in FIG. If it is determined in step S6 that the data difference on the end side of the area is large, the shift is made to the smaller number in the area with the smaller data difference (S12).
[0026]
If it is determined in step S7 that the data difference on the first side of the area is large, the shift is made to the larger number in the area with the smaller data difference (S13). By returning to step S3 by these shifts and repeating the same processing, the operation and effect described with reference to FIG. 5C can be obtained.
Subsequently, as described above, the output of the L-side sensor for the selected region is also obtained (S8), calculation based on [Equation 2] in FIG. 17 is performed (S9), and based on [Equation 4] in FIG. Interpolation is performed (S10), x is obtained, the distance L is obtained using this result and S, and focusing is performed (S11).
Note that the distance measurement control according to this flowchart is performed instantaneously prior to the shutter operation of the camera.
[0027]
(Operation effect 1)
As described above, according to the distance measuring apparatus of the present embodiment, the calculation method based on the expression shown in [Expression 3] is used to minimize the approximation error and calculate from the result value without erroneous distance measurement. Accurate focusing can be performed based on the obtained value. Further, since the shift amount S is obtained by using the correlation calculation means configured by the hard logic circuit in the distance measuring unit, the calculation speed can be increased.
[0028]
(Function 1)
In addition, the error of “mixing of perspective”, which has been a problem in the prior art, determines the above-mentioned | R1 −R0 | and | Rn + 1 −Rn | (in FIG. 1, | R4 −R3 |), which is also a feature of the present invention However, it can also be mitigated by a technique for reselecting the area where the approximation described above is established. For example, in a scene as shown in FIG. 5B, when the distance measuring unit 100 measures the person 6a, if the bright tree 6b enters the sensor array, the two graphs in FIG. The data as shown in Fig. 5 is obtained from the two left and right sensor arrays (right (R), left (L)). Here, the horizontal axis indicates the position n of each sensor constituting the sensor array, the vertical axis indicates the output values Ln and Rn of each sensor, and the width K corresponds to the face width of the person 6a. In such a scene, a sensor output in which the face and the background are mixed may be generated. However, for example, when ranging is performed in the area E as shown in FIG. 5C, such a mixed output is added to the calculation. Can result in false ranging.
However, if the method of the present invention is used, the difference between the outputs of the first sensor 1 and the preceding sensor 0 and the output difference of the last sensor E and the adjacent sensor E + 1 are compared. Even in the case of E, as shown in FIG. 5D in “area change”, the difference between the outputs of the first sensor 1 and the previous sensor 0, the last sensor E in the area, and the adjacent sensor E + 1 It is understood that correct distance measurement is possible because an area where the output difference is small is selected and interpolation calculation is performed at a place where it is difficult to mix the perspective.
[0029]
(Second Embodiment)
In the first embodiment described above, the high-speed calculation can be realized by using the correlation calculation means or the like. However, in the subsequent processing, the above-mentioned “area change” is performed by performing the approximate determination of the data difference at the end of the area. In the meantime, the distance measurement point of the subject image may change, and the already obtained shift amount S may change.
Therefore, in the second embodiment shown in the flowchart of FIG. 8, in order to deal with the above-described problems, an approximate determination of the data difference at the end of the area is performed before the correlation calculation, and the area used during the correlation calculation and the interpolation calculation. The “change” is eliminated. Here, the correlation calculation is also executed by the CPU according to the algorithm shown in FIG. 11, and is provided as a subroutine “correlation calculation” in step S27 in FIG. For simplicity of explanation, the correlation is assumed to be performed in an area in which three sensors are combined in accordance with FIG. 1 and [Equation 3].
[0030]
First, it is a step for selecting which part of the sensor array (line sensor) is used for distance measurement. Here, three steps from the n0th sensor are used. Since n0 and n4 at both ends are used, n0 is input as a reading start point (S20).
Then, the R side data is read (S21), and in subsequent steps S22 to S25, and steps S35 and S36, as in the case of FIG. Area shift is performed, and in accordance with the result, in the subsequent steps S21 to S26, substantial data reading of the selected areas of the left and right R and L sensor arrays is performed.
Thereafter, the above-described correlation calculation is performed (S27). Subsequently, Fmin, F-1, and F + 1 are calculated in the same manner as in step S9 of FIG. 7 (S28 to S30), interpolation calculation is performed, and x is calculated. The distance L is calculated (S31), and the distance L is calculated based on the values S and x thus obtained (S32). Optimum focusing is performed according to the distance measurement value (S33).
[0031]
(Operation effect 2)
As described above, in the present embodiment, determination is performed not only at the time of interpolation processing but also at the time of correlation processing to reduce the approximation error at the end of the area, and the area considering the result is used. No correlation error occurs when switching between. Therefore, according to the present embodiment, it is possible to provide an AF camera capable of accurate focusing without an approximation error during interpolation.
[0032]
(Third embodiment)
Although the above embodiment is relatively simple, there is one problem. That is, in the above-described embodiment, the sensor area is changed and moved to the place where the approximation is established, and therefore, there is a concern that a side area that the area to be measured cannot be measured can be measured. FIG. 9 illustrates a flowchart of the third embodiment as an example of countermeasures against this problem. Only the minimum necessary area can be measured, and distance measurement with little error is intended.
[0033]
For example, in the present embodiment, a distance measurement method using only data of an area with the smallest approximate error is performed by appropriately shifting the area within a predetermined area during distance measurement according to the flowchart of FIG. 9 described in detail below. ing. That is,
First, similarly to step S20 in FIG. 8 described above, the sensor number n0 immediately preceding the head of the area is designated as n, and the constant Rc4 for approximation error comparison is set to 100 (S40). The R-side sensor data thus selected is read and input to the CPU (S41), and in the next steps S42, S43, and S44, the difference Rc3 of the data difference at both ends of the area is calculated (S41 to S44).
[0034]
Rc3 thus obtained is compared with a comparison constant Rc4 (S45). If Rc3 is smaller, n in this case is set to n1 (S46), and Rc3 at this time is set as a new comparison constant Rc4 ( S47) After n is incremented (S48), it is determined that n does not increase by more than 5 sensor units from n0 (S49). If it is within this range, the process returns to step S41 again to loop and end error Rc3. Find n that minimizes. After this loop, n is obtained as n1 when n is between n0 and n0 + 5 and the error is minimum. Therefore, in this embodiment, since the lowest error area is obtained in the vicinity of the area designated first, the target point can be correctly measured.
Note that steps S50 to S53 from the reading of the corresponding L sensor data after step S50 to the interpolation calculation and focusing are the same as those after step S26 in FIG.
[0035]
(Operation effect 3)
As described above, in the present embodiment, by shifting the area within the determined area and performing distance measurement using the data of the area with the smallest approximate error, a part other than the area to be measured is measured. In addition, it is possible to improve the accuracy of distance measurement while minimizing the approximation error at the end of the area.
[0036]
(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described with reference to FIGS. 10 and 11. In this embodiment, first, when a state in which a part of the distance measurement area as shown in FIG. 10B is deviated from the person 6 is detected, the distance calculation is performed by removing the deviated sensor. This is an embodiment that takes measures against the “perspective mixed error” (see FIG. 6) described in the above. In this example, since the area used for distance measurement can be narrowed according to the case of shooting, it is more resistant to mixed perspective errors. In addition, if such an error does not occur, the area is not reduced, so that it is possible to measure a distance with higher accuracy for a subject larger than a human face even at a long distance. it can.
[0037]
In addition, in the shooting scene as shown in FIG. 10B, a sensor that is out of the subject has a poor correlation. Therefore, as shown in the graph of FIG. 10A, the difference | Rn−Ln + s | The value is different from that of the sensor that hits the subject. Using this principle, in the example of FIG. 10, the sensor output exceeding the difference Fc between the sensors R2 and Ls + 2 at the center is removed and interpolation calculation is performed.
[0038]
In order to realize the above functions in the CPU, an algorithm as shown in the flowchart of FIG. 11 may be adopted as a program. In this flowchart, for the sake of simplicity of explanation, the maximum width of the area is assumed to be 4 sensors from R0 to R4, and the above-mentioned mixed error is calculated by omitting any sensor at both ends of the area. . Accordingly, calculation processing is normally performed on four sensors or on three sensors when an error is likely to occur. In actual design, in order to cancel the noise component and reduce the quantization error, the calculation is performed in a wider area using the largest number of sensors. In this case, the above gist is expanded and applied. be able to.
[0039]
Steps S1 to S3 are equivalent to the steps with the same numbers in FIG. From the subsequent step S60, the corresponding L side sensor data is read (S60), and the first sensor 0 in the area described above is designated (S61). The difference Fi between R and L shown in FIG. 10 is calculated (S62), the designated sensor is incremented (S65), and this is repeated by returning to step 62 according to the determination (S66) until becoming the sensor 4. Since the sensor 2 is just the center sensor, the process branches to step S64 in step S63, and the difference F2 at this time is set as a threshold value Fc for determination as shown in FIG.
[0040]
Subsequent steps S67, S68, and S69 are a group of determination steps for the area edge that is one feature of the present invention, and there are three cases of sensors 0 to 4, 0 to 3, and 1 to 4, respectively. Approximate determination of the end is performed for (S67 to S69). In this case, a wider area is more preferable in terms of accuracy. First, the widest area is determined, but in this embodiment, in order to further determine the difference F described with reference to FIG. 10, the process proceeds to steps S70, S71, and S72. Branch to select an area with little difference from Fc and further narrow down the area. That is, the control is performed so that the difference Fi between Ri and Li greatly deviates from the reference value Fc by these determinations.
Then, using the area selected in such a process, interpolation is performed in subsequent steps S73, S74, and S75. Note that the calculation algorithm for calculating Fmin, F-1, and F + 1 for interpolation calculation is the same as in FIG.
[0041]
(Operation effect 4)
As described above, in the present embodiment, the size of the area is changed to deal with the perspective error, and when the target object is a person, the area is narrowed to eliminate the influence of the long distance, and in the case of a subject other than the person, A wide ranging area can be set, and a ranging sensor that is most suitable for the subject is selected to enable more accurate focusing.
[0042]
(Fifth embodiment).
Further, a flowchart as shown in FIG. 13 is also illustrated as a fifth embodiment based on the same technique.
In step S80, as in the flow up to step S66 in FIG. 11, the difference F between the left and right L and R is obtained for the sensor in the center of the area, and is set as the reference difference value Fc. Next, an area that does not deviate significantly from Fc thus obtained is detected (S81), sensor numbers at both ends thereof are obtained, and these are set as i1 and i2 (where i1 <i2). The subsequent step S82 is a step of performing an approximate error determination process at the end of the area which is one of the features of the present invention. Among them, the sensor indicating the value of | Ri + 1-Ri | having almost the same output difference as | Ri1-1-Ri1 | is detected, and its number i is set to i3 (S82). Then, by using the sensor between the numbers i3 and i1 obtained as described above, the image position difference between the correlation calculation and the interpolation calculation is detected, so that the distance measurement according to the principle of the passive type distance measurement can be performed. it can.
[0043]
(Operation effect 5)
As described above, according to the present embodiment, the size of the area is appropriately changed to take measures against mixed errors, and the distance measuring sensor part that is most suitable for the subject is appropriately selected to achieve high accuracy of focusing. ing.
[0044]
(Sixth embodiment)
The technique of the present invention is also useful for reducing errors other than the distance measurement error that occurs during shooting with a camera, for example. Therefore, as a sixth embodiment, an example in which the gist of the present invention is applied in a camera assembly process will be described in detail.
As described above, the EEPROM 9 (FIG. 3) is provided for storing a unique correction coefficient in order to correct errors caused by variations in camera parts and assembly accuracy. The correction value is detected and stored in the EEPROM in the manufacturing line adjustment process as shown in FIG. 6), and the CPU performs predetermined control while referring to the correction data. Specifically, a camera 23 that adjusts the camera ranging function during manufacture that is conveyed by the belt conveyor 22 by the operator 21 in this adjustment process is sequentially held by the fixed base 24 and connected to the personal computer 20 so as to be communicable. Then, by measuring the predetermined charts 25 and 26, a deviation from the design value is detected with respect to the distance measurement error caused by "variation" in the parts and assembly of the camera. The personal computer 20 sequentially calculates correction data for correcting the distance measurement error deviation, and writes the correction data in the memory (EEPROM 9) built in the camera.
[0045]
Note that, in the normal assembly process, for example, a slight attachment error of the light receiving lens occurs, and as a result, a graph showing the relationship between the image position difference (S + x) and the amount of extension of the focusing lens shown in FIG. The middle straight line often generates an H tilt error between the design value and the broken line having a predetermined tilt. Therefore, in the manufacturing process to which the present invention is applied, this inclination error is detected from the distance measurement results at two points, and this value is stored in the memory of the camera before shipment. This camera can correct correctly with the accuracy within the allowable range as originally designed by using the stored inherent error value.
[0046]
One of the two charts to be used is placed at a distance L1, and the other chart is at a distance L2. The chart of L1 can be moved up and down by the control of a personal computer. Then, the distance chart of L2 can be measured. As shown in the figure, there are black and white patterns on each chart, and distance measurement is performed by detecting the image position based on this light and dark distribution.
[0047]
The sensor data obtained by this light / dark distribution has the waveform of FIG. 14B, but the distance measurement data differs depending on which region of the obtained data is used for the interpolation process. Even if the same distance measurement area is used, when the width E1 of FIG. 14B is selected, the above-mentioned approximation of the end portion is difficult to be established, whereas when the width E2 of FIG. 14C is selected. The above-described end approximation is easily established. In other words, it can be said that the distance can be correctly measured by using E2 rather than E1, but if correct distance data is not obtained at the time of adjustment, the correct correction coefficient cannot be calculated, and the distance is measured using an incorrect correction coefficient. A camera cannot focus accurately. Therefore, in the present embodiment, the personal computer 20 performs the adjustment of the distance measuring device of the camera 23 according to the following procedure in accordance with the flowchart shown in FIG.
[0048]
First, a distance chart of L1 is placed in the distance measuring field (S201), and data from the distance measuring unit is read (S202).
In step S203, a condition determination based on the equation shown in the figure is performed on the approximation error at the end of the area according to the present invention (S203). If the condition is satisfied, the process branches to step S205. On the other hand, if the above equation is not satisfied, the area is changed (S204), and the process returns to step S203 to search for an area that satisfies this condition while repeating the same condition determination.
Thereby, the error in the area as shown in FIG. 14B is improved and shifted to the area as shown in FIG. 14C, and the position difference calculation of the image with less error is performed in step S205.
[0049]
Subsequently, the L2 chart is set so that the chart of the L1 chart can be measured from the ranging field of view (S206). Select an area and perform image position difference calculation with little error. Based on the difference in image position difference between L1 and L2 obtained in this way, the completeness of the distance measuring device is inspected, and the result is compared with the design value so that the correction coefficient represented by the inclination error of the above-mentioned graph straight line. H is calculated (S211), this value is stored in the EEPROM of the camera (S212), and this adjustment process is terminated.
[0050]
(Operation effect 6)
According to the present embodiment, the distance measurement error varies as shown in FIGS. 13B and 13C due to the error in the delicate positional relationship between the chart and the camera generated during adjustment. Therefore, it is possible to always calculate a correct correction coefficient and perform accurate adjustment. The camera adjusted in this way can be accurately focused without having an algorithm for approximate correction of the edge of the area during the distance measurement sequence due to the ROM capacity of the CPU built in the camera and the time lag.
(Modification)
In the embodiments exemplified above, as a means for obtaining the correlation amount S in sensor pitch units, the correlation means 5 built in the unit 100 constituting the distance measuring apparatus (FIG. 3) of the present invention is constituted by a high-speed hard logic circuit. However, the present invention is not limited to this configuration, and the CPU 10 is in charge of this equivalent function, and the arithmetic control procedure shown in the flowchart of FIG. 12 is realized by software, and the shift amount S is calculated. May be. The processing procedure here is a process for obtaining the correlation amount S. The minimum value of the FF described in FIG. 2 is represented as “FF1”, the shift amount on the horizontal axis of the graph is represented as “SFT”, and In the following description, n is an integer representing a sensor number.
[0051]
First, FF1 is initialized to 100 (S251), and SFT, n, and FF are initialized to 0 (S252). The subsequent steps S253 to S256 are a difference calculation processing loop. That is, the difference FF1 is taken (S253), and the obtained difference is added to the FF (S254). After the sensor number n is incremented by 1 (S255), the repetition of this processing loop is determined by comparison with a predetermined limit number nc (that is, the number of target areas) (S256).
[0052]
When the sum of the differences for each sensor is obtained as FF by the above processing loop, FF is compared with FF1 (S257). If FF1 is larger, FF is substituted into FF1 (S258), SFT is substituted into S (S259). After incrementing SFT by one (S60), the process returns to step S53 and repeats the above steps until a predetermined limit value Smax is reached. When Smax is reached, the process proceeds to step S3 in FIG.
[0053]
(Operation effect 7)
As described above, the means for obtaining the correlation amount can also be realized in software by the program in steps S251 to S261. As is clear from this, if the target area number nc for adding the correlation is small, the processing loop The number of repetitions can be reduced to increase the speed. In addition, since the correlation unit for each pitch can be obtained with an accurate S value even if it is somewhat rough, higher-speed processing is possible than calculation processing without area switching.
[0054]
(Other variations)
In addition, various modifications can be made to the embodiments of the present invention without departing from the gist of the present invention.
[0055]
The present invention includes the following invention as described above.
[1] Light receiving means for receiving a subject image by a pair of light receiving element rows having parallax and outputting a pair of subject image signals;
Block selecting means for sequentially selecting one block out of blocks divided into a predetermined size for each of the pair of subject image signals;
Change means for obtaining a difference between sensor signal levels at both ends in the selected block and adjacent sensor signal levels adjacent to each other and changing a predetermined calculation procedure related to the selected block according to a calculation result of the difference; ,
A distance measuring device comprising:
[2] The ranging apparatus according to [1], wherein the calculation procedure is changed by changing a block size or a block position.
[3] The two line sensors that output two light pattern signals according to the luminance distribution of the subject image observed from different fields of view having parallax, and the selection means that selects a block from the outputs of the line sensors, An adjuster used for adjusting a distance measuring apparatus having means for comparing the signals of the obtained blocks and calculating a relative position difference of the subject image based on parallax and a storage means for storing adjustment data. And
The adjuster calculates a signal difference between a sensor located at both ends of the block selected by the selection unit and a sensor adjacent thereto, and changes the selected block according to the calculation result. A ranging device having an adjuster.
[0056]
In addition, the following inventions are also included.
(1) Two line sensors that output two light pattern signals corresponding to subject images observed from different fields of view having parallax, selection means for selecting a block from the output of the line sensor, and the comparison of the compared blocks A distance measuring apparatus having means for comparing signals and calculating a relative position difference of the subject image based on parallax;
The distance measuring device characterized in that the selection means obtains a signal level difference between a sensor located at both ends of the selected block and a sensor adjacent thereto, and changes the selected block according to the result. .
(2) The selection means includes the absolute value of the output difference between the first sensor of the selected block and the last sensor of the previous block, the first sensor of the selected block and the first of the next block. The distance measuring device according to (1), wherein the absolute value of the output difference of the sensors is compared.
(3) The selection unit performs the comparison for a plurality of blocks adjacent to the selected block, and the calculation unit performs the relative position difference calculation according to the result. The described distance measuring device.
(4) The selection means performs the comparison for the selected block, and if the comparison result does not satisfy a predetermined relationship, performs the comparison again for the second block obtained by dividing the block, Accordingly, the distance measuring device according to (2), wherein the calculation means calculates the relative position difference.
(5) outputting a pair of subject image signals corresponding to the subject image observed by the pair of light receiving element arrays having parallax;
The subject image signal is divided into blocks of a predetermined size and sequentially selected one by one,
The difference between the sensor image signal level at both ends in the selected block and the signal level at each adjacent position adjacent thereto is obtained, and a predetermined calculation procedure related to the selected block is changed according to the result, and A distance measuring device that calculates a relative position difference of the subject image based on parallax by selecting a block.
[0057]
Further, the present specification includes the following inventions in addition.
(1 ′) two line sensors that output two light pattern signals according to the luminance distribution of the subject image observed from different fields of view having parallax;
Selecting means for selecting one area portion of the output of the line sensor;
A calculating means for comparing two light pattern signals from the compared portion and calculating a relative position difference based on the parallax of the subject image;
Ranging device having
The selection means has a function of calculating and comparing a difference between a sensor located at both ends of the selected area portion and an adjacent sensor, and changing the selected portion.
(2 ′) In (1 ′), the selection means calculates the absolute value of the output difference between the first sensor and the previous sensor in the selected area portion, and the output difference between the last sensor and the next sensor. Has a function of comparing absolute values.
(3 ′) In (2 ′), the selection means performs the comparison for a plurality of area portions adjacent to the selected area portion, and the calculation means performs the relative position difference calculation according to the result. Features.
(4 ') In (2'), the selection means performs the comparison for the selected area portion, and if this does not satisfy a predetermined relationship, the selection means compares the second area portion obtained by dividing the area portion. According to the result, the calculation means calculates the relative position difference.
(5 ′) two line sensors that output two light pattern signals according to the luminance distribution of the subject image observed from different fields of view having parallax;
Selecting means for selecting one area portion of the output of the line sensor;
A calculating means for comparing two light pattern signals from the compared portion and calculating a relative position difference based on the parallax of the subject image;
Storage means for storing adjustment data;
In an adjusting machine for adjusting a distance measuring device having
The adjustment means calculates and compares the difference between the sensor located at both ends of the area portion selected by the selection means and the adjacent sensor, and changes the selected portion so as to change the selected portion. It has a switching function to control.
[0058]
【The invention's effect】
As described above, according to the technology of the present invention, the distance measurement range of the passive type distance measuring device is appropriately switched for the subject, and the approximation error of the interpolation calculation is minimized and more accurate distance measurement is performed. It is possible. Therefore, by applying this technique, it is possible to provide a camera and a video camera that can be accurately focused.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing a basic structure of a distance measuring apparatus according to an embodiment of the present invention.
2 (a) to 2 (d) show the calculation principle of distance measurement according to the present invention;
(A) is a graph showing the relationship between the focus and the shift amount;
(B) is a graph showing the signal output for each sensor portion of the right sensor R;
(C) is a graph showing the signal output for each sensor portion of the left sensor L,
(D) is a graph showing the relationship between the left and right sensor outputs and the outputs of adjacent sensor portions.
FIG. 3 is a block diagram showing in detail the structure of the distance measuring device of the present invention.
FIG. 4 is an outline processing flowchart of arithmetic processing according to the embodiment of the present invention.
FIGS. 5A to 5D show the relationship between the sensor output and the width of the subject, and FIG. 5A shows two graphs showing the outputs of the left and right sensor arrays;
(B) is a conceptual diagram showing the positional relationship between the camera, the subject and the background;
(C) is an output waveform when ranging in area E,
(D) is an output waveform after area change.
FIG. 6 is a conceptual diagram showing a positional relationship between a line sensor and a subject.
FIG. 7 is a flowchart showing a calculation processing procedure of the first embodiment of the distance measuring apparatus.
FIG. 8 is a flowchart showing a calculation processing procedure of the second embodiment of the distance measuring apparatus.
FIG. 9 is a flowchart showing a calculation processing procedure of the third embodiment of the distance measuring apparatus.
FIGS. 10A and 10B show the sensor output and the positional relationship between the subject and the sensor.
(A) is a graph showing left and right sensor signals;
(B) is a figure which shows the corresponding position of the to-be-photographed object and the right sensor area in the imaging | photography scene.
FIG. 11 is a flowchart illustrating a calculation processing procedure according to the fourth embodiment of the distance measuring apparatus.
FIG. 12 is a flowchart showing a control procedure corresponding to the correlation means of the distance measuring apparatus.
FIG. 13 is a flowchart showing a calculation processing procedure of the fifth embodiment of the distance measuring apparatus.
FIGS. 14A to 14C show application examples to which the sixth embodiment of the distance measuring device is applied,
(A) is explanatory drawing which shows the manufacture and inspection line of a camera,
(B) is a waveform diagram of sensor data output when the width is E1,
(C) is a waveform diagram of sensor data output when the width is E2.
FIG. 15 shows a sixth embodiment of the distance measuring device,
(A) is a graph showing the relationship with the design value of the feed amount for focus adjustment;
(B) is a flowchart showing a calculation processing procedure as the sixth embodiment.
FIG. 16 is a diagram illustrating [Equation 1] representing an output of each L sensor portion when light of a deviation amount x is incident on the L sensor with the right sensor R as a reference.
FIG. 17 is a diagram illustrating [Expression 2] based on a difference between the left and right sensors;
FIG. 18 is a diagram illustrating [Expression 3] in which [Expression 2] is further developed.
FIG. 19 is a diagram illustrating [Expression 4] representing the relationship between the minimum value of F obtained by interpolation and the shift amount.
[Explanation of symbols]
1a, 1b ... lenses (for sensors),
2a, 2b ... line sensors,
3a, 3b ... signal output waveform,
4 A / D converter,
5 ... operation part (correlation means),
6 ... Subject,
7 ... Output section,
8 ... Shutter
9… EEPROM,
10: Control means (CPU),
12 ... Feeding means,
13 ... Zoom lens,
14: Register for calculation,
15 ... RAM,
16 ... ROM,
17 ... calculation means,
18: Register (general purpose),
19 ... Port.
S1 to S13 ... processing steps as the first embodiment of the present invention,
S20 to S36 ... processing steps as the second embodiment of the present invention,
S40 to S53: processing steps as a third embodiment of the present invention,
S60 to S75: processing steps as a fourth embodiment of the present invention,
S80 to S83 ... processing steps as the fifth embodiment of the present invention,
S101 to S104 ... Schematic processing steps of the embodiment of the present invention,
S201 to S212 ... Processing steps as the sixth embodiment of the present invention,
S251 to S261: Processing steps as one modification of the present invention.

Claims (1)

Light receiving means for receiving a subject image by a pair of light receiving element rows having parallax and outputting a pair of subject image signals;
Selecting means for selecting one of a plurality of areas divided into a predetermined size for each of the pair of subject image signals;
The difference between the sensor signal level of the light receiving element at both ends in the selected area and the sensor signal level of the light receiving element adjacent to each of the light receiving elements at both ends is obtained, and the calculation result of this difference satisfies the predetermined condition. Changing means for changing the size or position of the selected area ;
Correlation means for performing correlation calculation and interpolation calculation using the sensor signal of the area changed by the changing means,
A distance measuring device comprising:

Images