JPH01134265A - Speed measuring instrument - Google Patents

Abstract

PURPOSE:To measure a speed with high accuracy even when a moving body is at an equal speed or accelerated or decelerated at its start or stop by measuring a frequency only when a signal obtained by rectifying and detecting a narrow-band irregular signal is higher than a threshold voltage. CONSTITUTION:The narrow-band irregular signal outputted by an integrator 19 is branched through an LPF 20 for frequency measurement and signal decision. When the frequency is measured, a zero-cross pulse rectifying device 21 shapes the narrow-band irregular signal into a rectangular wave pulse, and a counter 23 counts the period of the signal and inputs the counted value to a microcomputer (muC) 31. Further, when a signal is decided, the signal from the LPF 20 is rectified 26 on a full-wave basis and detected 27, and then compared by a comparator 28 with a threshold voltage Vref. The output signal of the comparator 28 and the output signal of an FF 29 are inputted to an AND circuit 30, whose output signal is inputted to the muC 31. Then the muC 31 reads in data only when the output of the AND circuit 30 is at 'H' and clears a data latch circuit 24 and an FF 29.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a speed measuring device using an image sensor, and in particular, it is capable of accurately capturing the frequency of a spatially processed signal and measuring speed with high precision. The present invention relates to a speed measuring device suitable for measuring speed.

[Conventional technology]

As described in Japanese Unexamined Patent Publication No. 59-14044, the conventional device measures speed using a spatial filter, and detects the small amplitude portion of a spatially processed narrowband irregular signal (which causes frequency disturbance). In order to eliminate this, threshold levels were set at two points, one for the positive voltage and one for the negative voltage, and the signal was converted into three values and the frequency was measured. However, there is no mention of the distinction between not moving, that is, a stopped state, and moving.

[Problem that the invention seeks to solve]

As mentioned above, the prior art describes signal processing to ternarize a narrowband irregular signal, and the signal acquisition timing and the like are discussed.

Not considered.

The purpose of the present invention is to eliminate the timing of signal acquisition and parts with small amplitude levels of narrow band irregular signals, and to measure with high accuracy not only the equidistant range of moving objects but also the acceleration/deceleration range during startup and stop. This is what I am trying to do.

[Means for solving problems]

The above purpose uses an image line sensor to serially extract the amount of light received by each pixel corresponding to the subject, and electrically skips one or several pixels in a pulse train to configure a spatial filter, which converts narrowband irregular signals into It is divided into a device that measures frequency and a device that determines whether the signal is based on speed.
This determination is achieved by rectifying or rectifying the narrowband irregular signal and comparing it with a threshold voltage that is detected.

[Effect]

The narrowband irregular signal of the spatial filter has no center frequency and has a noise level when the moving object is stationary. Also, when speed occurs, the signal level immediately increases. Therefore, as mentioned above, by comparing the rectified and detected narrowband irregular signal with the threshold voltage, it is possible to determine whether a speed has occurred, and it is also possible to remove the narrowband irregular signal below the threshold voltage. .

Therefore, in such a device, even when a moving object suddenly stops or slips, a signal corresponding to the speed of the moving object can be immediately obtained.

To measure the frequency of a narrowband irregular signal, the narrowband irregular signal is shaped into a rectangular pulse through a zero cross detection circuit, and its period is counted with another stable high frequency pulse.
The method is to calculate it from the count value. When the above-mentioned signal discriminator determines that the signal is not a signal, the count value is taken in, and the low level portion of the stopped state or the narrowband irregular signal is removed. In addition, the accuracy of speed measurement has been improved because the count value, that is, the frequency data, is subjected to frequency distribution processing to determine the center frequency.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a speed measuring device according to the present invention.

Reference numeral 1 denotes an image sensor, such as a CCD line sensor. The line sensor has (1) a small pixel pitch;
(2) Each dimension of the light receiving part can be made with high precision. (3) The signals of each pixel can be read out sequentially using a clock. There are advantages such as 2 is an amplifier.

In the case of a CCD line sensor, it is a differential amplifier that extracts a light reception signal based on the difference between two output signals of the sensor. 3
is an A/D converter that converts the light reception level of each pixel of the line sensor 1 into a digital value.

4.5 is a buffer with a gate, and this gate is a reference clock φ1 for reading out the light reception level of each pixel of the line sensor.
5 and the scan signal 5H17, only the number of dummy pixels is counted (6, 8) t, and the clock TA7 with the period of the pitch of the spatial filter. It is turned on and off by TB9. This signal will be described later. 10.11 is clock TA.

An adder that adds the digital values of pixels passed by TB performs subtraction 12 using a scan clock when the sum of two series (for convenience, A and B series) of one scan is determined. The output of the subtracter 12 is a signal obtained by bit shifting, that is, amplifying 13 and delaying the scan clock by 18.
The A converter 14 converts it into an analog signal. The output of the D/A converter is a value for each period of the scan clock 17, and is integrated 19 times to form a continuous waveform. At this time, when the latch time of the D/A converter approaches the scan time, the integrated signal level increases. This integrated signal presents a narrowband irregular signal.

There is also a method of reading the subtracted digital value for each scan clock into a computer and using software such as FFT (Fast Fourier Transform) to find the center frequency.

For narrowband irregular signals, a high frequency reference clock1
Since signals such as 5 and scan clock 17 are superimposed, LPF (Low Pa5s Filter)
20 to remove high frequency components. Although not shown in the figure, DC components are also removed using a capacitor or the like. The low frequency range is passed through in order to measure the frequencies in the low speed range.

The signal from which noise has been removed is branched to a frequency measuring device and a signal discriminating device. For frequency measurement, a narrowband irregular signal is waveform-shaped into a rectangular wave pulse by a zero-cross detection and comparator circuit 21, and the period of this signal is counted by another stable high oscillation frequency 22 pulse using a counter 23. This counted value is connected to the microcomputer 31 via the latch circuit 24.

Further, for signal discrimination, the signal after the LPF 20 is passed through a buffer 25 and subjected to after-wave 26 or half-wave rectification and detection 27 to obtain an envelope component signal. This detected signal is compared with a threshold voltage V r e f by a comparator 28 . The level of this threshold voltage V r e f is experimentally set to a level higher than the noise level when a moving object is stopped, and slightly higher than the voltage level at which the frequency waves of the narrowband irregular signal are disturbed. F, F, (flip-flop) circuit 2 that captures and latches the output signal of the comparator 28 and the rise of the pulse waveform signal of the irregular signal.
The output signal of 9 is input to the AND circuit 3o.

The output signal of the AND circuit 30 is sent to the microcomputer 31.
Enter.

The microcomputer 31 reads the data of the counter by looking at the output of the AND circuit 30, that is, the output is 41.
When the data is “H”, the data is read and the data latch circuit 24
and clear F, F, circuit 29. If data is not read when the signal is L'', data will not be read when the moving object is stationary or when the level of the irregular signal is low, improving the reliability of the data.

Next, we will briefly explain the principle of a non-contact speedometer using a spatial filter. FIG. 2 shows that when the light receiving elements 32 arranged in a row are moving at a speed V with respect to the subject 34, the lens 3
This is a schematic representation of the relative relationship with 3. This principle is that random reflection unevenness on the surface of the subject 34 is optically detected, spatial processing is performed, and the velocity V of the moving object can be measured in the form of frequency. Here, the spatial processing means arranging the photoelectric conversion elements 32 in a line at regular intervals as shown in FIG. 2, and calculating the difference in the total sum of photocurrents connected every other element. This signal presents a narrowband irregular signal. Assuming that the frequency of this narrow band irregular signal is Hl, the speed of the moving object is V, the magnification of the lens is m, and the pitch of the light receiving element 32 (pitch of the spatial filter) is P, the following relational expression is obtained.
Below margin ■=-・fc −(1)
That is, since the pitch P and the magnification m are known, the velocity V can be measured by finding the center frequency fc.

Next, the configuration of a spatial filter using the CCD line sensor 35 shown in FIG. 3 will be described.

As described above, the line sensor 35 has a small pixel pitch of several tens of μm, and a very narrow gap between pixels of several μm. Therefore, when an object is observed using an optical system, the uneven pattern of the object can be captured with high precision due to its high resolution, and the moving speed of objects with small unevenness can also be measured.

The configuration of the spatial filter is to take out pixel signals arranged in a - column, and use a pulse train to skip every other pixel signal or every few pixel signals as shown in FIG. 3 to form a filter. Third
The figure shows a case where the pitch P of the filter is 4 pixels.

Next, a method of constructing a spatial filter using a pulse train will be explained using FIG. 4.

In FIG. 4, (a) the SH signal 36 is a scan clock signal, and (b) the φ signal 37 is a clock for reading out pixel signals of the line sensor, which is the reference clock in FIG. (
a) SH multiplied signal, count 1 for the number of pixels of the line sensor
6 and created it. (c) The signal 38 is an output signal of the line sensor, and is a signal whose level changes depending on the amount of light received by the subject. In the case of a CCD line sensor, -
In general, the signal (c) 38 and (not shown in Fig. 4)
c) By synchronizing with signal 38, taking a differential with a constant level pulse signal, removing offset noise, amplifying and integrating, it is possible to observe the line unevenness pattern of the object shown in (d) 39.

Further, generally, a CCD line sensor or the like has dummy pixels, that is, insensitive pixels at both ends.

(e) The TA signal 40 is counted by 6 for one dummy pixel from the scan clock SH, and 4 pulses from the beginning of the signal pixel (
One pulse is a pulse signal every other (corresponding to one pixel). (
f) The TB signal 41 is a pulse signal that is counted by 8 (dummy pixels + 2 pixels) from the scan clock SH and every 4 pulses. (g) 42 represents a pixel of the line sensor, and the number of signal pixels is 128.

If the signal pixels are S□l sat S31..., S1□8, the pixels corresponding to the TA times signal are 8198m9...
・、S1m! Then, the pixel corresponding to the TB multiplied signal is S3.
S...

S□, 7. Therefore, in FIG. 1, when the A/D-converted digital values of the line sensor are passed using the TA and TB pulses, the signals of the pixels 5ztS4tSst, . . . , S, etc. are not taken in, and a filter is formed.

In FIG. 4, the pitch P of the spatial filter is set to 4 pixels, but since the (e) and (+) pulse trains can be arbitrarily selected, the pitch P can be easily changed.

Furthermore, in FIG. 1, all pixels of the line sensor are A/D converted, but there is no problem if the A/D conversion is performed and added at the timing of the TA and TB signals.

The above is the configuration of a spatial filter using line sensors.

Next, we will discuss narrowband irregular signals.

FIG. 5(h) 43 shows the waveform of the narrowband irregular signal after LPF when the moving object is moving at a constant speed. Irregular signals include areas where the vibration level is low and the frequency is disturbed. Further, the line sensor is installed so that the line direction coincides with the moving direction and is parallel to the subject surface. If this happens, the irregular signal becomes asymmetrical, that is, it becomes a waveform in which low frequency components are superimposed, making it impossible to accurately measure the frequency.

In FIG. 6, a method for measuring the frequency of a narrowband irregular signal will be described using a pulse waveform. (i) 44 is an irregular signal (
j) 45 is a pulse signal shaped into a rectangular wave at zero cross. (k) 46 is an oscillation frequency pulse for counting, and 9 a crystal oscillator or the like is used to stabilize the oscillation frequency. (j
As an example of a method of counting the period of the pulse waveform of )45 using the signal of (k)46, a method of counting using two series of counters will be described.

First, the first series of counters count from the rising edge ■ of the (j) 45 signal to the next rising edge ■ ((1) 47 signal).

The second series counter counts from the next rising edge ■ to rising edge ■ ((m) 48 signals). In other words, in two series (
j) A method of counting 45 cycles alternately. With such a method, it is possible to capture all cycles.

Further, the frequency is expressed as count value/oscillation frequency.

Next, using Fig. 7, determine whether the signal is based on speed or not,
A system for removing low level narrowband irregular signals will be described.

The (h)49 signal is a narrowband irregular signal, and the (0)50 signal is a full-wave rectified signal of the (h) signal.

The (p)51 signal captures the envelope line using the signal waveform obtained by detecting the (0) signal. It also shows the level relationship between this (p) signal and the threshold voltage V r e f , and when the moving object is stationary, the noise level is almost zero level because DC is cut. Therefore, as shown in (p) 51, the threshold voltage Vref is set to a level slightly higher than where the amplitude level of the narrow band irregular signal is small. (p
) signal and Vref are input to a comparator, and the output waveform is (g)52. Therefore, the irregular signal (h)4
9 is pulse-shaped at the zero cross, and the period read into the computer 31 is (q) with the signal rising, (
r) The signal shown in 53 is obtained. By doing this, it is possible to remove frequencies where the amplitude level is small and where the frequency is disturbed by the narrow band irregular signal. It is also possible to determine the presence or absence of irregular signals.

Let us consider the case where the system shown in FIG. 1 is mounted on, for example, a self-propelled robot.

(1) When a self-propelled robot moves straight, it takes a short time from when the command to "go straight" is output until the travel drive motor actually starts moving. Therefore, as mentioned above, if the pulse count frequency measurement method is used, the frequency will be measured from the noise signal within the above-mentioned time.

(2) When a self-propelled robot stops, it takes a certain amount of time for the robot to come to a complete stop after the ``stop'' command is output.Therefore, if the spatial frequency data is not read at the same time as the ``stop'' command, an error will occur. Become.

(3) When a self-propelled robot is traveling straight ahead, in an extreme case, if all the vehicles slip and the robot itself does not move, the output of the spatial filter will be at a noise level and the frequency will be measured in this state.

In this way, in cases (1) to (3), reliable data can be obtained by looking at the level of the output (irregular signal) of the spatial filter and capturing the spatial frequency.

Next, the spatial frequency data obtained when moving using the method described above will be described.

Although reliable spatial frequencies can be obtained using the system shown in FIG. 1, when spatial frequencies are measured experimentally, there are variations as shown in FIG. 8. Therefore 1
Rather than finding the speed from the spatial frequency data of each individual,
As shown in the figure, data reliability is improved by taking a frequency distribution and setting its peak frequency as the center frequency jc.

In addition, we experimentally varied the moving speed and the center frequency fc according to each frequency distribution when moving a certain distance.
Find the speed using equation (1). Compared to the accuracy between this speed and the actual average speed, the accuracy was slightly improved by setting the center frequency to the average value of fc±Δfc.

Therefore, the first step is to measure the speed of moving objects such as self-propelled robots.
As shown in Figure 0, the center frequency fc is initially determined by the frequency distribution.
is determined, and from now on, the average value within fc±Δfc is newly set as the center frequency fc.

In such an algorithm, even if the speed changes, it can be followed and the speed can be measured with high accuracy.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to measure the spatial frequency only when an object is moving, and it is also possible to eliminate the case where the level of the narrow band irregular signal is small, thereby improving the reliability of the spatial frequency in one circuit. In addition, the center frequency was determined using software, such as frequency distribution processing of spatial frequency data. As a result, highly accurate speed measurement was possible.

[Brief explanation of the drawing]

Figure 1 is a system block diagram of one embodiment of the present invention, Figure 2 is a system block diagram of an embodiment of the present invention.
The figure is a conceptual diagram of an optical system using a spatial filter, Figure 3 is a configuration diagram of a line sensor and a spatial filter, Figure 4 is an explanatory diagram of input/output signals of the line sensor and pulse signals of the filter configuration, and Figure 5 is a narrowband irregular signal waveform diagram, Figure 6 is a signal waveform diagram of one frequency measurement, Figure 7 is a signal waveform diagram that removes the low signal level part of Figure 5, and Figure 8 is an experimentally obtained signal waveform diagram. FIG. 9 is a frequency distribution diagram thereof, and FIG. 10 is a flowchart of speed measurement in a moving vehicle. 1... Image line sensor, 3... A/D converter. 10.11...Adder, 12...Subtractor, 14...
・D/A converter, 23... Frequency measurement counter, 2
6...Full wave rectification, 27...Detection, 28...Comparator, 31...Microcomputer, 43...Narrowband irregular signal. Representative Patent Attorney Masaru Ogawa Akira Ogawa 2nd Eye 5th Akira 50 Mitsuotsume High 7th Departure 6 Kuni M Review Office'1 Figure No C Pa Yumizui Rainbow 70th

Claims (1)

[Claims] 1. In a device that introduces the reflected light from the object to be measured through a lens to an image sensor, configures a spatial filter with an electric pulse signal, and measures the frequency from the output signal of the spatial filter to obtain the speed, Compare the full-wave or half-wave rectified and detected signal with the threshold voltage,
A speed measuring device characterized in that the frequency is measured only when the detected signal is equal to or higher than a threshold voltage.