JP2857240B2 - Flow velocity distribution measurement device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an apparatus for measuring a two-dimensional flow velocity distribution of a flow field based on an image signal of the flow field visualized by a tracer injection method. .

(Prior Art) For example, a method of visualizing a flow field by injecting a tracer into a fluid in order to know a flow velocity distribution of a flow field created by a flow of a fluid in a passage or the like is known. In order to obtain the flow velocity information in the flow field visualized in this way as quantitative distribution data, the state of the visualized flow field is converted into a video signal using an interlaced television camera. For the obtained image data,
Two frame images are created from each field image in the frame image, and information on the inclination angle and length of the thin line obtained by the thinning processing of the flow trace image formed by the same tracer in each image. And a method of measuring the distribution of the flow velocity from the amount of movement between two frame images of the trajectory image formed by the same tracer has been proposed (visualization of flow,
Vol. 9, No. 34, pp. 379-382). Since this method deals with only one frame of television image, it has a feature that the applicable range is wider than that of a method which deals with two or more frames of television image.

However, in the above-described conventional method, a tracking process is performed in which the inclination angle and the length are obtained from the thinning processing result of the trail image, and the trail image is estimated using the result.
If the noise is superimposed on the image, the tracking processing result is greatly affected by the noise, and there is a problem that a magnitude error occurs in the measurement result of the flow velocity.

In order to reduce this error, in order to visualize the flow, a good illumination condition is an essential requirement, and the particle size of the tracer particles must be selected and used within a certain range. This is because the lengths of the fine lines of the trace images of particles having different particle diameters after the thinning process are different even if the lengths of the trace images are equal. Therefore, there is a problem that these points must be overcome in order to put the above-mentioned conventional technology into practical use.

(Problems to be Solved by the Invention) Therefore, as a flow velocity distribution measurement system capable of reducing the influence of noise and measuring the flow velocity distribution state more accurately, interlaced scanning of a visualized flow field without a shutter is performed. Shooting is performed using an image pickup tube type television camera, and a logical product operation of each field image data constituting one frame of the obtained television image signal is performed to correspond to a trail image between each field image data. An arrangement for determining the relationship has been proposed.

However, in the proposed system, the measurement time is limited by the NTSC television image sampling cycle, and in the vidicon TV camera, when the moving speed of the particles in the field angle is fast, the particle flow Due to the fact that an error occurs in the image length, there is a problem that measurement is difficult in the case of a fast flow.

It is therefore an object of the present invention to provide an apparatus for measuring a flow velocity distribution, wherein a fast flow can be measured using the image signal of the flow field without causing the above-mentioned problems. It is in.

(Means for Solving the Problems) A feature of the present invention to achieve the above object is to provide a two-dimensional flow velocity of a flow field based on an image signal of a trace image of the flow field visualized by injecting a tracer. In a flow velocity distribution measuring device that measures the distribution, multiple CCDs for capturing the visualized desired flow field in the same field of view
The shutter type imaging device and the above imaging device have the same start timing of each exposure but different end timings,
A shutter control unit for obtaining from the imaging device a plurality of image data for performing shutter control of the imaging device and causing overlap of the trail images, and a logical product between the plurality of image data obtained from the imaging device A logical operation means for performing an operation and outputting data indicating a correspondence relationship between the flow image indicated by each of the image data, and calculating a centroid position of a corresponding set of flow image according to the operation calculation of the logical operation means And a calculating means for calculating a movement vector of the calculated position of the center of gravity.

(Operation) When the flow field visualized by the CCD shutter type imaging device is photographed, image data indicating the trace of the movement of the tracer while the shutter is open is obtained from each imaging device.

In each of the imaging devices, shutter control is performed by the shutter control unit so that the exposure start timing is the same but the end timing is different, and a plurality of image data that causes overlapping of the trail images are obtained. Therefore, 1
Each trace image represented by one image data and each trace image represented by other image data always overlap at the end of the image by an amount corresponding to the overlap exposure time.

As described above, in one set of images obtained by the CCD shutter type imaging device, each of the trajectory images showing the movement trajectory of the same tracer particle is superimposed on the image at the end corresponding to the overlapping exposure time. It will have an overlap. Therefore, by finding this characteristic overlap between the trace images in each image, it is possible to associate the trace images with the same tracer particle between different image data. This association is performed, for example, based on the result of a logical product operation of two image data, and calculates the position of each center of gravity of a corresponding set of trajectory images. Etc. can be measured. By performing the above measurement at an appropriate place in the flow field, the distribution of the flow velocity in the flow field can be measured.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows the flow field of the fluid 2 visualized by injecting appropriate tracer particles 3 such as nylon 12 particles into the fluid 2 in a water tank 1 made of transparent glass. 1 shows a configuration of a flow velocity distribution measuring device 10 according to the present invention. The flow velocity distribution measuring device 10 controls the shutter of the visualized flow field so as to generate a common exposure timing in the same field of view and a shift in the exposure timing. The first imaging device 11 and the second imaging device Has 12

Each of these imaging devices 11 and 12 is a CCD shutter type television imaging device, and includes first and second imaging devices 1 and 2.
In order to make the field of view with respect to the flow field captured by 1 and 12 the same, in this embodiment, the optical device 14 using the half mirror 13 is used.
Is used. The optical device 14 is configured by arranging the half mirror 13 in the dark box 15 as shown in the drawing, and the entrance window 15a of the dark box 15
Are arranged so as to face the water tank 1. The first and second imaging devices 11 and 12 are connected to the dark box 15 via light paths 15b and 15c, and a half mirror
The light A transmitted through 13 is incident on the first image pickup unit 11, while the light B reflected by the half mirror 13 is incident on the second image pickup device 12.

The optical system of the optical device 14 is adjusted so that the field of view seen from the first image pickup device 11 and the field of view seen from the second image pickup device 12 are the same, and therefore the second image obtained by the first image pickup device 11 is obtained. If one image and the second image obtained by the second imaging device 12 are images obtained at the same timing, they are mutually inverted left-right inverted or vertically inverted images, but have the same visual field image.

Therefore, in order to facilitate comparison between the first image and the second image, one of the images needs to be inverted by a mapping operation, and the mapping operation for this inversion is performed in the data processing unit 17 described later. It is performed in.

In practice, the adjustment for matching the visual field viewed from the first imaging device 11 with the visual field viewed from the second imaging device 12 is performed in addition to or in addition to mechanical adjustment in the optical system. In addition, this may be performed by electrically correcting the signal from the imaging device as described later.

Reference numeral 16 denotes the first and second imaging devices 11 and 12
For controlling the opening and closing of each shutter (not shown)
And a shutter control unit for outputting the second control signals S1 and S2. Examples of the waveforms of these control signals S1 and S2 are shown in FIGS. 2 (a) and 2 (b), respectively.

The first control signal S1 is a signal that becomes “H” level at time t = t0 and becomes “L” level at t = t1 after a lapse of time T1, and the first imaging device only during the period of t0 <t <t1. The shutter 11 is controlled to the open state. On the other hand, the second control signal S2 attains the “H” level at time t = t0, maintains the “H” level state at t = t1 after the elapse of the time T1, and the time t2 after the time T2 elapses from t = t1. = Signal at "L" level at t2. As a result, the shutter of the second imaging device 11 is opened only during the period of t0 <t <t2. As can be seen from the above description, the time T1 from t0 to t1
Is that both the first and second imaging devices 11 and 12 are in the exposure state, but the second imaging device 12 further has a time period from t1 to t2.
T2 is also exposed. That is, each exposure timing in the imaging devices 11 and 12 is set to include a common exposure timing (t0 <t <t1) that is staggered in time and that the corresponding images always overlap at the ends. Have been.

Returning to FIG. 1, the shutter control as described above allows the first and second imaging devices 11 and 12 to obtain the first
And the second television image signals V1 and V2 are supplied to a data processing unit 17 constituted by a known microcomputer.
, Where image data processing for flow velocity distribution measurement is performed according to a predetermined processing program.

FIG. 3 is a functional diagram showing the contents of image data processing executed in the data processing unit 17.
The image data processing executed in the data processing unit 17 will be described based on this functional diagram. The first television image signal V1 and the second television image signal V2 are
Analog / digital conversion means (A /
D) The data is converted into corresponding first and second digital data VD1 and VD2 by 171 and 172.

The first digital data VD1 further includes an inversion processing unit 1
The data is input to 73, where processing for left / right or vertical inversion is performed, and the inverted first digital data VD1 'is output. As a result, the image represented by the inverted first digital data VD1 'is matched with the image represented by the second digital data VD2. The reversing means 17
It is needless to say that a required inversion processing means may be provided on the output side of the analog / digital conversion means 172 instead of 3.

FIG. 5A shows the inverted first digital data VD1 '.
Is shown, while FIG. 5 (b) shows a trail image y according to the second digital data VD2. FIG. 5 (c) is a diagram in which FIG. 5 (a) and FIG. 5 (b) are superimposed. As can be seen from FIG. 5C, the trajectory image x according to the inverted first digital data VD1 'and the trajectory image y according to the second digital data VD2 are displaced along the flow direction of the fluid 2, and The parts obtained at time T1 overlap.

The set of data VD1 'and VD2 obtained in this manner is supplied to AND operation means 174, where each stream image according to the inverted first digital data VD1' and each stream image according to the second digital data VD2 are obtained. An AND operation is performed with the image. Then, combination data indicating a combination of flow trajectory images having a logical product image according to the logical product operation result is provided to the flow velocity vector calculating means 175.

The flow velocity vector calculation means 175 is further provided with the inverted first digital data VD1 'and the second digital data VD2, and calculates the center of gravity of each flow image based on the input data. Further, a movement vector of each position of the center of gravity of the pair of trace images based on the combination data indicating the combination of the trace images is calculated as a flow velocity vector at that point. Based on the flow velocity vector data indicating the calculated flow velocity vector, the two-dimensional flow velocity distribution of the fluid 2 in the visual field captured by the first and second imaging devices 11 and 12 is measured.

Next, the image processing operation will be described in more detail based on the flowchart of the image processing program stored in the memory 17a in the data processing unit 17 shown in FIG.

When the execution of the image processing program 20 is started, first,
In step 21, an image correction coefficient is calculated.
The image correction coefficient is necessary for matching the two visual fields when there is a shift between the visual field of the image obtained by the first imaging device 11 and the visual field of the image obtained by the second imaging device 12. This is a coefficient indicating the correction amount.

In the illustrated embodiment, the degree of displacement between the two visual fields is first detected using a predetermined pattern (not shown), and a correction coefficient is calculated according to the detection result. In the next step 22, two image data are input based on the first and second television image signals V1, V2. In step 23, using the image correction coefficient calculated in step 21, image correction for equalizing the fields of view of these two trail images and inversion processing on image data based on the first television image signal V1 are performed. Thus, first trail image data D1 based on the output of the first imaging device 11 and second trail image data D2 according to the output of the second imaging device 12 are obtained.

These stream image data D1 and D2 correspond to the inverted first digital data VD1 'and the second digital data VD2 shown in FIG. 3, respectively, and the first and second stream image data D1 and D2 Figure 5 (d),
(E) respectively.

Next, the process proceeds to step 24, where the flow image data D
Data processing for binarizing each image indicated by 1, D2 is performed. In the data processing for the binarization, each data D1, D2
To a binary image, which results in 2
Binary data Da and Db representing the digitized flow image are obtained.

In the next step 25, unnecessary information is removed based on the binarized data Da and Db, that is, a noise removal process is performed. This process removes the data of the trail image connected to the frame of the image at all from each trail image.

Noise removal processing of the trail image periphery due to expansion and contraction.

A process of removing stream images having a number of images equal to or less than a predetermined number as noise.

Is included. However, it goes without saying that other appropriate noise removal processing may be performed in addition to or in place of some of them.

Next, in step 26, labeling 1a, 1b, 1c,..., 2a, 2b, 2c... Is performed on each of the trail images in the two trail images left as a result of the noise removal processing in step 25 ( With reference to FIGS. 5 (d) and 5 (e)), the calculation of the center of gravity and the calculation of the area are performed for all labeled trajectory images.

Thereafter, the process proceeds to step 27, where the position of each of the trail images in the trail image obtained by the first imaging device 11 and the trail image in the trail image obtained by the second imaging device 12 is determined. The logical product is calculated. Next, in step 28, combination data indicating the combination of the labeling of the trail image in the one trail image and the labeling of the trail image in the other trail image having the result value of the logical product operation is obtained. . For example, the label
When the trail image of 1a and the trail image of the label 2a are in a correspondence relationship and in the positional relationship shown in FIG. 6, the result of the logical product of these trail images is indicated by hatching in FIG. It represents the area of the superimposed portion, and when this area is larger than zero, it is assumed that both trace images are trace images formed by the same tracer particle. Therefore, in this case, combination data indicating one set of trail images indicated by the labels 1a and 2a is obtained. This calculation is performed for all the combinations, and the combinations of the trail images are determined.

In step 29, it is checked whether the following conditions are satisfied in order to determine whether each combination of the trail images obtained as described above is valid.

(A) Each of the trail images is combined with only one of the trail images at the other time.

(B) A straight line connecting the centers of gravity of one set of trace images passes through a logical product image obtained by these.

Only the combination data that satisfies each of the conditions (a) and (b) is taken as correct combination data, and the combination data that does not satisfy any one of the conditions (a) and (b) is removed.

In step 30, the center of gravity of the stream image 1a obtained by the first imaging device 11 is selected from among the set of stream images indicated by the combination data for which the validity has been recognized, as shown in FIG. 6, for example. Nagareato image 2a obtained from W ₁ using the second image capturing apparatus 12
The vector VC is calculated toward the gravity center position W _2, the vector VC is the velocity vector at this point of the fluid 2.
In this way, the flow velocity vector is calculated as described above for the flow image pairs indicated by all the combination data.

As a result, the flow velocity vector at each position in the image becomes apparent, and in step 31, based on each flow velocity vector,
The flow velocity distribution is measured, and data indicating the result is output in an appropriate form. This data may be shown as an image on a display tube, for example, or the data may be printed out.

As described above, the device shown in FIG. 1 uses two CCD shutter type television image pickup devices, and as described in FIG. 2, these image pickup devices have different timings for opening the shutter, and Since exposure control is performed so as to have an overlapping portion within the exposure time, each of the trail images indicated by one image data and each of the trail images indicated by other image data are overlapped exposure time (T1) Are obtained as two images that overlap by an amount corresponding to.

Therefore, by appropriately controlling the opening and closing of the shutter, regardless of the moving direction and the moving speed of the tracer,
The trail image of the trail image obtained by the first imaging device 11 and the second
The correspondence between the trajectory image of the trajectory image obtained by the imaging device 12 and the trajectory image can be found very easily by calculating the logical product between them. In this case, the first and second
Since the way of shifting the exposure in the imaging devices 11 and 12 is set to overlap at the end of the trail image, it becomes easy to grasp the correspondence of the trail images between the respective image data. Therefore, since the measurement time is not restricted by the TV image sampling period of the NTSC system, the flow velocity distribution of the flow field can be measured very simply and quickly with a simple process.

In the above embodiment, in order to obtain the same visual field in the first imaging device 11 and the second imaging device 12,
Although the optical device 14 was used, the optical device 14 is not always necessary. For example, these two imaging devices 11 and 12 are juxtaposed at a considerable distance from the water tank 1, and the input optical axis of each imaging device 11 and 12 is A configuration may be adopted in which the fields of view are the same by being regarded as substantially parallel. In this configuration as well, as described above, it is of course possible to perform correction for making the same field of view for each output signal as necessary, but in this case, it is not necessary to perform reverse transfer. It has the advantage that.

(Effects of the Invention) According to the present invention, as described above, the shutters of a plurality of CCD shutter-type imaging devices are controlled such that the exposure timing of each imaging device is shifted in time but has overlapping exposure timing. Since an overlapping portion is generated at the end of a set of flow images corresponding to the flow images obtained from the respective imaging devices, the correspondence between the flow images is determined by a logical AND operation of the data indicating each flow image. In addition to being able to easily determine, the trajectory image in each trajectory image can be accurately imaged without being affected by the TV image sampling period, the moving speed and direction of the tracer. In addition, the present invention has an excellent effect that the flow velocity distribution can be measured with extremely high reliability.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a flow velocity distribution measuring system according to the present invention, and FIGS. 2 (a) and 2 (b) show control signals for controlling respective shutters of the image pickup apparatus of FIG. Waveform diagram,
FIG. 3 is a functional diagram showing a configuration of a main part of the flow velocity distribution measuring system of FIG. 1, FIG. 4 is a flowchart showing an image processing program executed by the computer of FIG. 1, and FIGS. FIG. 5 (e) is a view showing a trail image for explaining image processing in the system of FIG. 1, and FIG. 6 is a logical AND operation and a velocity vector calculation of the trail image executed in the system of FIG. FIG. 2 fluid 3 tracer 10 velocity distribution measuring device 11 first imaging device 12 second imaging device 13 half mirror 14 optical device 16 shutter control Apparatus 17 Data processing unit 174 AND operation means 175 Flow velocity vector calculation means V1 First television image signal V2 Second television image signal VD1 First Digital data VD2: Second digital data

Claims (1)

(57) [Claims] The present invention relates to a method for measuring a flow field of a flow field, which is visualized by injecting a tracer.
In a flow velocity distribution measuring apparatus adapted to measure a three-dimensional flow velocity distribution, a plurality of CCD shutter type image pickup apparatuses for photographing a visualized desired flow field in the same field of view, and a start timing of each exposure of the image pickup apparatus Is the same, but the shutter control means for obtaining a plurality of image data from the imaging device for performing shutter control of the imaging device and causing overlap of the trail images so that the end timings are different, and obtaining the image data from the imaging device. Logical operation means for performing a logical product operation between the plurality of pieces of image data obtained and outputting data indicating a correspondence relationship between the trail images indicated by the image data, and a corresponding one of the logical operation means according to the operation result of the logical operation means Calculating means for calculating the position of each center of gravity of the set of trajectory images and calculating a movement vector of the calculated position of the center of gravity. Measuring apparatus.