JP3034967B2 - Distance measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distance measuring apparatus and, more particularly, to a distance measuring apparatus for automatically focusing a camera.

[0002]

2. Description of the Related Art In the flow of camera automation, a camera focusing technique has been developed to enable focusing on a subject which is difficult to focus on. It has been desired to improve the technical means for focusing on the object. That is, 1) a subject located outside the center of the shooting screen, and 2) a moving subject.

Conventional techniques for focusing on these subjects and preventing out-of-focus include: [1] a multi-point ranging technique for measuring a plurality of points on a screen in accordance with the above items. [2] moving object ranging technology that detects a subject speed in accordance with a plurality of distance measurement results measured at staggered times, thereby preventing defocus due to subject movement occurring during a release time lag. . Regarding the latter moving object ranging technology, for example, a detecting device for detecting the speed of a moving object on a belt conveyor is disclosed in Japanese Patent Application Laid-Open No. 62-23257.
Japanese Patent Application Laid-Open No. 63-159817 discloses an automatic focusing device capable of maintaining a focused state irrespective of the movement of a subject.
Respectively.

[0004]

However, the above-mentioned 2)
In two technologies, multi-point ranging technology and moving object ranging technology,
The problem is that the time required for ranging is lengthened. That is, with respect to the above item [1], the time required for distance measurement increases as the distance measurement points increase.
In the above item [2], it is necessary to perform distance measurement many times at different times, and both of these are technical means that sacrifices the release time lag.

That is, in a camera having a multi-point distance measuring function and a moving object distance measuring function, simply combining these techniques results in a long release time lag and misses a photo opportunity.

SUMMARY OF THE INVENTION An object of the present invention is to provide a distance measuring apparatus which solves the above-mentioned problems and has a short release time lag in a camera having both a multi-point distance measuring function and a moving object distance measuring function.

[0007]

A distance measuring apparatus according to the present invention comprises: first distance measuring means for measuring a distance of a subject substantially at the center of a photographing screen; and a subject at a substantially central part of the photographing screen. Moving object detecting means for detecting a moving speed of moving in the optical axis direction, and correcting the output of the first distance measuring means based on the detection result, and measuring object distances at a plurality of peripheral points in the photographing screen The second distance measuring means and the first distance measuring means are operated in response to a distance measurement start instruction signal, and if the subject distance obtained thereby is within a predetermined range, the moving object detecting means is successively operated. And control means for operating the second distance measuring means if it is not within a predetermined distance range.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the illustrated embodiments. First, before describing an embodiment of the present invention, a basic concept of the present invention, a principle of distance measurement of an active triangular distance measuring method, and a principle of detection of a moving object speed detecting method will be described. First, to explain the basic concept of the present invention, from the fact that moving object ranging technology is not very effective at long distances,
An object of the present invention is to effectively utilize the two technologies of the above-described multipoint ranging technology and moving object ranging technology while minimizing the release time lag.

That is, 80% or more of the photos generally taken can take a desired photo even if the focus is achieved only by the result of the center distance measurement. Of the remaining 20%, most of the reasons for out-of-focus are due to the situation where two persons are standing side by side, there is no subject at the center, and a so-called hollow state occurs.

At this time, the result of distance measurement at the center should indicate ∞ or a considerably long distance. Therefore, there is no need to perform multi-point ranging unless the center ranging result indicates a far distance.

The moving object ranging can only be expected to have a remarkable effect at a short distance unless the subject is photographed at an extremely high speed.

In general, the focus position of the photographing lens is proportional to the reciprocal of the subject distance d. For example, when photographing a subject at a speed of 2 m / s, if the time lag is 0.5 seconds for simplicity,

[0013]

Although the focus position needs to be corrected in proportion to the distance, the smaller the subject distance d, the larger the correction amount.

Conversely, when the subject distance d is large, the subject normally enters the depth of field of the lens without performing the focus position correction for the moving object. There is little need for moving object ranging.

When a moving object is photographed, a moving subject is tracked, so that the human eye naturally looks at the center of the screen. Therefore, it is considered that the main subject exists at the center of the screen with considerable probability.

To summarize the above, (a) multi-point ranging is to prevent the hollow phenomenon, so that peripheral ranging can be omitted except when the central subject is at a long distance.

(B) Since a long-distance subject can be covered by the depth of field, even if the long-distance subject is a moving object, moving object ranging can be omitted.

(C) When a moving object is tracked, there is a high probability that the subject is placed at the center, so that the peripheral distance measurement can be omitted in the moving object distance measurement.

Therefore, in the distance measuring apparatus of the present invention, (a) an object located substantially at the center of the photographing screen is measured by the first distance measuring means, and (b) if the result of the distance measurement is a long distance exceeding a predetermined value. It is determined that there is a possibility of a void, and the distance measurement of the peripheral portion, that is, multi-point distance measurement, is performed by the second distance measuring unit that measures the distance of the peripheral object in the shooting screen.

(C) On the other hand, if the result of the distance measurement does not exceed a predetermined value, that is, if the distance is not a long distance, moving object distance measurement is performed next.

The above is the basic concept of the present invention. Next, the principle of known active triangulation will be described with reference to FIG.

In FIG. 3, an infrared light emitting diode (hereinafter abbreviated as IRED) 1 emits infrared light toward a subject 21 via a light projecting lens 2. Subject 21
Is incident on an optical position detecting element (hereinafter abbreviated as PSD) 4 via a light receiving lens 3, and the incident position x depends on the subject distance d and is expressed by the following equation (2). Given to.

X = s · f / d (2) where f is the focal length of the light receiving lens 3, and s is the distance (base line length) between the optical axes of the light projecting and receiving lenses 2 and 3.

Assuming that the distance from the end of the PSD 4 to the point where the optical axis of the light receiving lens 3 crosses the PSD 4 is a and the total length of the PSD 4 is t, the total sum of the photocurrent generated by the incident signal light is ipo.

[0026]

[0027]

[0028]

## EQU1 ## That is, when the ratio of the two signal light currents of the PSD {I1 / (I1 + I2)} is taken, the object distance d
Can be calculated from the constants a, t, s, and f based on the above equation (6).

Further, since the sum ipo of the photocurrent generated by the reflected signal light from the subject located at the subject distance d is inversely proportional to the square of the subject distance d, the standard reflectance chart obtained from the standard reflectance chart located at the subject distance 1 m is obtained. If the total signal light current is Ipo, it is given by the following equation (7).

Ipo = (1 / d) ² · Ipo (7) If Ipo is constant as described above, the sum i of the signal light currents is obtained.
It is also possible to obtain the subject distance d from po.

Therefore, in this specification, the distance measuring method capable of measuring the distance without depending on the reflectance of the object of the above formula (6) is a ratio calculation method, and the distance measuring method according to the above formula (7) is a light amount. These will be referred to as detection methods, respectively.

As described above, the ratio calculation method is not affected by the reflectance of the subject, but involves the calculation of the ratio. For example, when the denominator becomes 0 due to the influence of noise, the distance measurement output becomes ∞. As is clear from the example of
When / N is reduced, the accuracy is extremely reduced.

On the other hand, the light amount detection method simply uses the signal light current I1
This is a method advantageous for S / N as compared with the ratio calculation method, because only the sum of I2 and I2 is detected. Further, in this method, the effect of improving the ranging accuracy is remarkably exhibited by causing the IRED to emit light many times and integrating the output. The principle of the active triangulation is described above.

Next, the detection principle of the moving object speed detection method to which the distance measurement principle based on the above ratio calculation method is applied will be described with reference to FIGS. 4 (A) and 4 (B). FIG. 4A is a block diagram showing the configuration of the technical means. The 1 / d circuit 30 finds the reciprocal 1 / d of the subject distance based on the principle of the above equation (6). This is a circuit for obtaining the subject distance d by conversion. Also, 3
Reference numerals 5 and 36 denote switches, such as analog switches, for switching the output supplied to the integration means together with the sign reversing circuit 32 and the inverter 33 in accordance with the logic level of the signal applied to the terminal 34.

FIG. 4B is a timing chart for explaining the operation. In the case where the IRED emits light eight times, the logic level of the terminal 34 is switched between the first half and the second half, so that the integration of the first half is performed by turning on the switch 35. In the negative direction,
The integration in the latter half is performed in the positive direction when the switch 36 is turned on.

In other words, the output of the integrating means for one IRED emission depends on the subject distance d. Thus, by switching the integration direction in the middle of the IRED emission, the output of the integrating means is switched after the IRED emission ends. Travel speedυ
Is output as VOUT from the integrating means. The above is the detection principle of the moving object speed detection method to which the distance measurement circuit based on the ratio calculation method is applied. Next, an embodiment will be described.

FIG. 1 is a block diagram of a distance measuring apparatus according to a first embodiment of the present invention. Reference numeral 101 denotes a first distance measuring apparatus for measuring a distance to a subject located at the center of a photographing screen. Means. This output result is determined by the CPU 104, which also functions as a determination unit and a control unit, and the next operation command is determined by the second distance measurement unit 1 that measures the distance to the subject located at the periphery of the screen.
02 or via the gate 106 to the moving object detecting means 103 which outputs a signal depending on the speed of the subject.

Then, based on an output result of the second distance measuring means 102, the moving object detecting means 103 or the first distance measuring means 101, the CPU 104 focuses on a focusing lens, an actuator, a position detecting encoder and the like. Control means 105. The reason why the multi-point ranging or the moving object ranging is switched in accordance with the output of the first ranging unit 101 is as described in the basic concept.

FIG. 2 is a flow chart of the distance measuring sequence in FIG. 1. When a distance measuring start signal is input to the CPU 104 by operating a release switch (not shown) in step S0, as shown in step S1. Then, the distance measurement at the center of the screen is performed. This distance measurement result d
s is compared with a predetermined object distance, for example, 4 m in step S2, and if ds ≦ 4m is satisfied, correction when the object is a moving object is required. Therefore, the process proceeds to step S3 to perform moving object detection.
Obtain the subject speed υ.

Since this speed υ is small and, for example, no correction is necessary for a still subject, this is determined in step S4.
Proceed to step S8. In this step S8, the distance measurement result d
s is stored in the memory area dx of the CPCU 104, and this d
The photographing lens is driven to the point x to perform focusing.

Returning to step S4, the object speed 被 写 体
Is large and correction is required, step S5, step S5
As shown in FIG. 6, focusing is performed at a distance of ds−Δt. Here, t is a time lag from ds ranging to exposure start.

Returning to step S2, if the distance measurement value ds is longer than 4 m, there is a possibility of a void, as described above. Therefore, the distance measurement of the peripheral subject is performed by the second distance measurement means.
Here, distance measurement is performed on two points on the left and right in step S9, and among the three points including the center in step S10,
Select the closest ranging data. Then, step S1
The lens is driven to the position indicated by the closest distance measurement data among the three points after 1 to perform focusing.

The above is the description of the first embodiment of the present invention.
Next, a second embodiment of the present invention will be described. The second embodiment is significantly different from the first embodiment in that the distance measurement accuracy is improved at a long distance where the subject distance exceeds a predetermined value, for example, 4 m. ,
This will be described in detail with reference to FIG.

In the second embodiment, as shown in FIG.
A known signal light for distance measurement by the RED 1 is projected through a light projecting lens 2, reflected light from a subject (not shown) is received by a PSD 4 via a light receiving lens 3, and a subject distance is calculated based on the incident position. Active triangulation is adopted.

However, the IRED 1 and PSD 4 are divided into three in order to measure not only the center of the photographing screen but also the left and right, and are divided into 1S, 1L, 1R and 4 respectively.
S, 4L, 4R.

That is, the light for distance measurement projected from the IRED 1L is, as indicated by dL1,
3 and an angle θ from the optical axis of a photographic lens (not shown). If there is a subject at the light projection position, the reflected light from the subject enters the light receiving lens 3 through the optical path of dL2.

The incident angle of the dL2 is inclined with respect to the direction connecting the principal points of the light transmitting and receiving lenses 2 and 3, ie, the base line length direction, depending on the subject distance. Therefore,
The position of the signal light spot incident on the PSD 4 is different in the base line length direction of the PSD 4 depending on the subject distance. That is, the object distance can be obtained by knowing the incident position of the signal light on the PSD 4.

The function of the PSD will now be described.
The element SD has a function of outputting the incident position of the signal light as two current signals. Therefore, when the signal light enters the center of the PSD 4, the ratio of the two current signals of the PSD becomes 1: 1. When the signal light enters the point where the PSD is internally divided into 3: 1, the current signal ratio becomes 1: 3.

PSD having such a position detecting function
The shorter the length, the higher the resolution when detecting the minute displacement of the signal light. This is because, since the position information is output as a ratio, if the length of the PSD is long, the amount of change in the signal with respect to the minute displacement becomes small, making it difficult to detect. However, on the other hand, the longer the PSD length, the larger the displacement range of the signal light, so that it is suitable for wide-range ranging. Returning to FIG. 5 again, the amplifiers 5 to 9 are preamplifiers that absorb and amplify the output signal of the PSD 4 with low input impedance, and are selected by the CPU 104 via the input terminal 14 and the decoder 16 based on Table 1 below.

[0051]

That is, the unselected preamplifier does not perform an amplifying operation because its input and output are open.

The signal current of the PSD 4 amplified by the preamplifiers 5 to 9 is compressed by the compression circuits 10 and 11. Then, the ratio calculation is performed by the expansion calculation circuit 12, and the output terminal 1
A signal depending on the subject distance is output to 3.

The CPU 104 controls the I
One of the three divided 1L, 1S, and 1R of the RED1 is made to emit light, and the signal output to the output terminal 13 is A / D converted and input to be used as distance information.

In this embodiment, an intermediate electrode is provided in the PSD 4S for center distance measurement in order to utilize the above-described switching of the distance measurement accuracy depending on the PSD length.

The operation of the present embodiment configured as described above will now be described.
Referring to the flowchart of FIG. 6, first, center distance measurement is performed in step S21. In this case, the distance measuring light projected from the IRED 1S is received using the full range of the PSD 4S. That is, since the intermediate electrode of the PSD 4S is not used, the preamplifiers 5 and 7 are selected as shown in the section 3 of Table 1 above, whereby the range from the nearest to ∞ is the range where the distance can be measured.

On the other hand, in the center (far-distance) distance measurement mode, as described later, it is assumed that the distance to the subject at the substantially central part in the photographing screen is about 10 m or more, so that the IRE is used.
The ranging light projected from D1S is received using the upper half range of PSD4S in FIG. That is, P
Using the intermediate electrode of SD4S, the above-described ratio calculation operation is performed by the preamplifiers 5 and 6 as shown in section 4 of Table 1 above. This intermediate electrode has a PSD4 of 1:
Since it is provided at a position internally divided into 1, the distance measurement accuracy is doubled as described above. However, in this mode, 1
Since the distance can be correctly measured only over a distance longer than 0 m, the camera cannot be established only in this mode even if the accuracy is high.

Proceeding to step S22, the program proceeds to step S22.
If it is determined that the distance measurement result ds in step 21 is 4 m or less, the moving object ranging mode is enabled, so that the moving object is detected (step S23).

This moving object detection is performed by measuring the distance twice at a predetermined time interval. The IRED is 1S and the PSD is 4S.
Use the full range of The first distance measurement result is d1, the second distance measurement result is d2, and the distance measurement time interval is t.
If 1, the object speed 被 写 体 in the lens optical axis direction can be obtained as υ = (d1−d2) / t1.

This object speed す ぎ る is too slow, for example,
If it is 1 m / sec or less, the subject can be regarded as a stationary object substantially. Therefore, it is determined that no correction is necessary in step S24, and the process proceeds to step S28, where the CPU focuses on the distance measurement result ds. On the other hand, if moving object correction is necessary, the CPU performs an operation as in step S26 to focus on dx.

Returning to step S22, the object distance d
If s is longer than 4 m, the left and right distance measurement is performed in step S29. This is, first, 1L of IRED and 4L of PSD
, And then the distance is measured by 1R of IRED and 4R of PSD. The preamplifier used at this time is, as shown in sections 1 and 2 of Table 1 above,
8, 9 are selected. In addition, both ends of the PSDs 4L and 4R are connected to the amplifiers 8 and 9, respectively.

By the way, the PSD 4S may be connected to the amplifiers 8 and 9 in the same manner. However, if the light receiving area is increased, the influence of external light noise is increased. PSD4S is dedicated to S /
N is improved to improve accuracy.

In step S30, the closest result is selected from ds and the left and right distance measurement results. Step S32
If it is determined that the distance is longer than 10 m, the length of the central PSD 4S is halved as described above, and a more accurate distance measurement is performed (step S33). The preamplifier uses 5 and 6.

Here, the reason why the distance is measured only at the center at a distance of more than 10 m will be explained. Usually, it is extremely rare to photograph a person at a distance of 10 m or more, for example, a building is often photographed. In such a situation, it is considered that the hollow phenomenon is unlikely to occur.

In step S32, dx is 10
If it is determined that the distance is less than m, focusing is performed on the object with the highest probability of being the main subject, that is, the closest one of the three distance measurement results. The above-described determination, calculation, selection of the amplifier, and focusing operation are performed by the CP.
Let U do it. The above is the description of the second embodiment of the present invention.

Next, a third embodiment of the present invention will be described with reference to FIGS. In the first and second embodiments, the subject distance is measured by the ratio calculation method. On the other hand, in the third embodiment, the ratio calculation method and the light amount integration method are used to measure the subject distance with high accuracy. The difference is that they are used together. By using the integrating means also for moving body speed detection, speed detection not affected by noise is enabled.

FIG. 7 is a block diagram of a distance measuring apparatus according to a third embodiment of the present invention, and FIG. 8 is a flowchart thereof. In the figure, a light projecting means 29 can select and project light at the center, left, and right of the photographing screen, respectively. Can be respectively received.
Each of them has an active triangulation system as shown in FIG.

When the release operation is performed by pressing the shutter button, the CPU 104 turns on the switch 22 in the direction A, the switch 23 in the direction C, and the switch 26 in the direction E. The IRED for distance is emitted.

The central sensor 4S receives the reflected signal light from the object (not shown), and the ratio calculating means 24 measures the distance based on the above formula (6) (step S42).
And the integrating means 28 performs distance measurement based on the above equation (7) (step S43).

The CPU 104 reads these distance measurement results and determines data having higher reliability as the center distance measurement data LS (step S44). Then, this LS is compared with, for example, distance data of L0 corresponding to 3 m (step S45).

As a result of this comparison, if LS is shorter than L0, as described above, the effect of the moving object distance is obtained, and the peripheral distance measurement by the left and right sensors 4L and 4R is unnecessary. Side, the switch 23 is turned off, and the switch 26 is turned on to the F side, and the moving body speed detection described with reference to FIG. 4 is performed while projecting the central IRED (step S48). At the same time, left and right I
The RED is sequentially emitted, and the left and right distance measurement results LR and LL are input to the CPU 104 using the ratio calculation means 24 (step S46).

At this time, the focusing distance Lx is determined differently depending on the output result υ of the speed detecting means comprising the speed detecting means 27 and the integrating means. That is, if υ is υo, for example, 0.2 m or less per second, it is determined that the object is not a moving object (step S49), and no moving object correction is particularly performed. Then, the peripheral distance measurement results LR and LL and the central distance measurement result L
The closest distance measurement value is selected from S (step S4)
7).

On the other hand, if υ is greater than υo, it is determined that the main subject is the center subject, and the time lag Δt until the exposure is reached.
Then, the focusing distance Lx is determined in accordance with the following formula (8): Lx = LS−υ · Δt (step S50).

Returning to step S45, LS becomes L0
If farther, the CPU 104 sets the switch 22 to the B side,
The switch 23 is switched to the D side and the switch 26 is switched to the E side, and the flow proceeds to steps S51 and S52. That is, the center IRED no longer emits light, and the left and right IREDs
Sequentially emit light by the number of light intensity integrations, and perform ratio calculation distance measurement and light intensity integration distance measurement for the left and right subjects in the same manner as in steps S42 and S43.

Step S53 is a subroutine for selecting the most reliable distance measurement value LR or LL from the ratio calculation result and the light amount integration result, and is the same routine as step S44. This is to select the ratio calculation result on the assumption that the S / N of the distance measurement signal is sufficiently high, for example, when the light quantity integration result is sufficiently large, for example, near 5 m.

The left distance measurement result L obtained in this way is
Based on L, the right-side distance measurement result LR, and the center-side distance measurement result LS, the closest distance measurement value is selected in step S54. The selected distance measurement value is set to the focusing distance Lx. And

After that, in step S55, Lx is focused, and after an exposure sequence (step S56),
This flow ends. Here, it is assumed that the CPU 104 performs the control, the determination, the calculation, and the like.

FIGS. 9 (A) and 9 (B) show each distance measurement operation and IRED.
FIG. 9 is a timing chart showing a state of light emission selection of FIG.
FIG. 9A shows a case where the process branches to Y in step S45 of FIG. 8, and FIG. 9B shows a case where the process branches to N.

In the present embodiment, as is apparent from the chart of FIG.
Since the left and right distance measurement calculations 42 and 43 are performed in a superimposed manner during 1, the time lag can be prevented from becoming long.

In this embodiment, as described with reference to FIG. 4, the integration means 28 is switched in a time-sharing manner.
It is possible to provide a highly accurate multi-AF for moving objects without increasing the number of components such as capacitors, that is, in a form that is advantageous in terms of cost and space.

As described above, another operation is rationally combined while performing the calculation in the time-sharing manner, so that a camera which does not have a long time lag and which is resistant to a photo opportunity can be provided. it can.

As described above, according to each of the above embodiments, even if the subject is not in the center of the screen or a moving object such as a running child, the subject does not become out of focus and has a time lag. It is possible to provide a short distance measuring device that does not cause a risk of missing a photo opportunity.

[0083]

As described above, according to the present invention, the distance of the subject substantially at the center of the photographing screen is measured by the first distance measuring means, and if the measured distance is within a predetermined range, The moving object detection means outputs an operation signal depending on the moving speed of the subject in the optical axis direction at a substantially central portion in the photographing screen. On the other hand, if the distance measurement value is not within the predetermined range, the second distance measurement is performed. Since the distance of the peripheral subject in the shooting screen is measured by the means, a remarkable effect that the release time lag in a camera having both the multi-point distance measuring function and the moving object distance measuring function can be shortened is exhibited.

[Brief description of the drawings]

FIG. 1 is a block diagram of a distance measuring apparatus according to a first embodiment of the present invention.

FIG. 2 is a flowchart of a distance measurement sequence in FIG. 1;

FIG. 3 is a diagram illustrating a principle of active triangulation.

FIG. 4 is a block diagram and a timing chart illustrating a detection principle of a moving body speed detection method.

FIG. 5 is a block diagram of a distance measuring apparatus according to a second embodiment of the present invention.

FIG. 6 is a flowchart of a distance measuring sequence in FIG. 3;

FIG. 7 is a block diagram of a distance measuring apparatus according to a third embodiment of the present invention.

FIG. 8 is a flowchart of a distance measurement sequence in FIG. 7;

FIG. 9 is a timing chart of distance measurement calculation and light emission selection in FIG. 7;

[Explanation of symbols]

 101 first distance measuring means 102 second distance measuring means 103 moving object detecting means 104 CPU (determining means, control means)

Claims (3)

(57) [Claims] A first distance measuring means for measuring a distance of a subject at a substantially central portion in the photographing screen; and a moving speed of moving the subject at a substantially central portion in the photographing screen in the optical axis direction , Based on the detection result,
Moving object detecting means for correcting the output of the first distance measuring means, second distance measuring means for measuring subject distances at a plurality of peripheral points in the photographing screen, and the first distance measuring means in response to a distance measuring start instruction signal; Activate distance measuring means
And the subject distance obtained thereby falls within a predetermined range.
If there is, the moving object detecting means is operated continuously, and a predetermined distance range is set.
If it is not within the range, the control means for operating the second distance measuring means
And a step. 2. If the moving speed is lower than a predetermined value as a result of further operating the moving object detecting means, the control means sets the distance measured by the first distance measuring means to the final distance measured value. The distance measuring apparatus according to claim 1. 3. A center distance measuring means for outputting a signal depending on a distance of a subject at a central portion in a photographing screen, and detecting a moving speed of the subject at a substantially central portion in the photographing screen in the optical axis direction. , Based on the detection result
Comprising a moving object detecting means for correcting the output of the central distance measuring device, and a peripheral distance measuring means for outputting a signal depending on the distance to the subject at a plurality of points of the peripheral portion within the imaging screen, a distance measurement start instruction Activate the center distance measuring means in response to the signal.
And operating the moving object detecting means or the peripheral distance measuring means according to the result .