JPH04286914A - Method and apparatus for detecting of signal peak position - Google Patents

Abstract

PURPOSE:To enable fast and correct detection of time when a signal level peak is reached by calculating an intersection of two virtual lines having an inclination corresponding to a rate of change in signal levels before and after the peak value. CONSTITUTION:A signal level is detected by a signal detection part 1 at a predetermined time interval by a clock generator 6. When the maximum detection value B is obtained, this value is used along with respective detection values C,A immediately before and after the value B. On a two-dimensional plot with a signal level and time as axes, a first line AB passing through two points of the maximum value B and a detection value with the larger difference from B, for example the detection value A, and a second line BC passing through the other detection value C in the opposite inclination to the above are obtained. An intersection E of the lines AB,BC is quickly calculated 5 by an arithmetic sequence which is set and stored 4 in advanced, and this time is regarded to be time when the signal reaches the peak.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] This invention is directed to the peak position of a signal,
That is, the present invention relates to a method for detecting the time when the level of an ever-changing signal reaches its peak, and a device for the same.

0002

2. Description of the Related Art It is common practice in many devices to detect the time when a signal reaches its peak and utilize this for various types of control. However, accurately detecting the peak position of a signal is surprisingly difficult and requires relatively complicated processing procedures. As an example, there is a known three-dimensional object measuring device that detects the signal level at regular time intervals and calculates the center of gravity within a predetermined range including the peak of the signal (for example, Japanese Patent Application Laid-Open No. 63-1999) (See Publication No. 279110). As is well known, this type of device irradiates the object to be measured with a slit-shaped laser beam, photographs this with a video camera, converts it into a signal that changes according to the shape of the object, and peaks the signal. The shape of the object to be measured is measured from the position, and it is necessary to accurately detect the peak position of the signal.

0003

[Problems to be Solved by the Invention] In the above-mentioned publication, the subject is an image taken with a video camera, and the peak position of brightness is determined by the center of gravity calculation of Σ (time x brightness)/Σ brightness. However, in order to perform this calculation in real time, the calculation of time x luminance must be performed within one clock of the A/D converter, and this multiplication calculation requires time, so the clock cannot be made faster. In addition, a multiplier is required, and the entire processing circuit becomes quite complex. Furthermore, if the peak width is narrow, data before and after the peak required for calculation may be insufficient, resulting in inaccurate calculation.

The present invention has focused on this point and has been made with the object of accurately detecting the time when the signal level reaches its peak using a relatively simple processing circuit.

[0005]

[Means for Solving the Problems] In order to achieve the above object, in the method of the present invention, two straight lines having slopes corresponding to the rate of change in the signal level before and after the peak value of the signal are imagined, The position of the intersection of these straight lines is calculated to detect the peak position of the signal. Specifically, the signal level is detected at a predetermined time interval, and when the maximum detected value is obtained, this maximum detected value and each detected value immediately before and after it are used to plot the signal level and time as an axis. On the two-dimensional diagram where 1 of the detected value with the smaller difference from
A second straight line passing through the point is determined, and the position of the intersection of these first and second straight lines is calculated and determined as the time when the signal reaches its peak.

[0006] It should be noted that before and after the peak value of a signal, the signal waveform can generally be considered to have a symmetrical shape around the peak, so the slope of the second straight line is the same as the slope of the first straight line. should be treated as having the same inclination angle and opposite slope, but if it is known that the waveform becomes asymmetrical around the peak value due to the characteristics of the device or circuit, the calculation process may be changed depending on the degree of the asymmetrical waveform. An appropriate correction may be made to the slope of the second straight line.

In order to perform such detection, the apparatus of the present invention includes a signal level detection means for detecting the signal level at predetermined time intervals, and a signal level detection means for detecting the signal level at predetermined time intervals, and for detecting the signal level immediately before and after the maximum detected value. A first straight line passing through the two points of the maximum detection value and the detection value with a larger difference from the maximum detection value on a two-dimensional diagram with the signal level and time as axes; , find a second straight line that has an opposite slope to the slope of the first straight line and passes through one point of the detected value that has a smaller difference from the maximum detected value, and find the intersection of the first and second straight lines. and calculation means for calculating the position on the time axis and outputting this as the time when the signal reaches its peak.

In order to easily perform the above arithmetic processing, the first and second straight lines are stored in the storage means using two values: the maximum detected value and the relative values between the three detected values immediately before and after the maximum detected value. By storing in advance the calculation procedure for finding the intersection point, and inputting the above two relative values calculated by the subtracter into the address corresponding to each relative value in the storage means, the intersection point of the two straight lines can be calculated on the time axis. The storage means may be configured to output a signal corresponding to the decimal digit in .

FIG. 1 is a block diagram showing the basic configuration for carrying out the present invention, in which 1 is a signal level detection section, 2 is a temporary storage circuit, 3 is a subtracter, 4 is a storage means, and 5 is a storage means. 6 is a clock generator, 7 is an input signal, and 8 is an output signal.

0010

[Operation] The peak position of the signal can be detected by imagining two straight lines having slopes corresponding to the rate of change in the signal level before and after the peak value, and finding the position of the intersection of these straight lines. In addition, by imagining two straight lines using the maximum detection value detected at a predetermined time interval and the detection values immediately before and after it, the location of the intersection can be easily determined using a preset calculation procedure. It can be calculated as follows. In particular, the calculation procedure for finding the intersection of straight lines using the maximum detected value and the relative values between the three detected values immediately before and after the maximum detected value is stored in advance, and the two relative values are stored for each relative value in the storage means. By inputting the signal to the address corresponding to the intersection point and outputting the signal corresponding to the decimal place of the intersection point, the calculation can be performed easily and quickly.

[0011]

[Embodiment] An embodiment shown in the drawings will be described below. This invention was made through improvement research on a three-dimensional object measuring device similar to that of the above-mentioned publication, and this embodiment is also described as an example of application to such a three-dimensional object measuring device. The application of the circuit is not limited to this type of device, and can be appropriately used in various other circuits.

In FIG. 2, reference numeral 10 denotes a video camera of the three-dimensional object measuring device, and an image taken of an object to be measured (not shown) irradiated with a slit-shaped laser beam is output in the form of a predetermined video signal 11. , are converted into digital signals by the A/D converter 12. This digital signal is a series of digital data corresponding to pixels for each scanning line, and is sent to an A register 13, a B register 14, and a C register 15. Registers 16 and 17 are used to shift data sequentially, and the B register 14 receives the previous data, and the C register 15 receives the previous data.

On the other hand, data is also sent from the register 16 to the data selector 18, and compared with the threshold value of the threshold register 19, if the data is greater than this, it is input to the maximum value register 20 and held. . The threshold value is a value set in advance so that peak position detection processing is performed only when the signal level is above a certain value, and when the signal level does not exceed the threshold value, each register 13, 1
4 and 15 remain cleared to 0. These operations proceed in synchronization with the horizontal synchronization signal in the video signal, and when the horizontal synchronization signal ends, the counter 21 starts counting the clock signal 22 and stores the count number in the D register 23.
Enter the information in sequence. The comparator 24 compares the data from the register 16 and the data from the maximum value register 20, which are input one after another, and outputs a maximum value signal 25 when the maximum value is detected.

When the maximum value signal 25 is output from the comparator 23, each register 13 to 15 and 23 is latched, and the subtracter 26 extracts the data B from the B register 14 from the A register 1.
3 data A is subtracted, and the subtracter 27 subtracts the data A from the B register 14.
The data C in the C register 15 is subtracted from the data B in the C register 15, and these results are input to the address of the ROM 28. R.O.
The following two equations t=(B-C)/2(B-A) t=1-(B-A)/2(B-C) are stored in M28 in advance, and the input address The ROM 28 outputs a signal with the number of bits corresponding to , and the D register 23 outputs a count number corresponding to the pixel position at that time.

The register 29 stores this count number in the ROM.
The signal is shifted by the number of bits outputted from 28 and is synthesized. This signal corresponds to the decimal place value of the time counted by the clock signal 22, and the peak position of the signal is calculated by the value expressed on the time axis by the clock signal 22. The above processing is performed while being controlled by the CPU 31 via the interface 30.

FIGS. 3 and 4 are explanatory diagrams of this calculation procedure. FIG. 3 shows B-A≧B-C, and FIG. 4 shows B-C>B-
Case A is shown. The horizontal axis is the time axis t in units of clock signals, and the vertical axis is the signal level, that is, the brightness of the image taken by the video camera. Straight lines are shown as solid lines, and signals are shown as broken lines. In FIG. 3, the straight line AB is the maximum value B latched in the B register 14 and the A register 1.
Straight line passing through the two points of value A immediately after it is latched to 3, straight line BC
passes through the point of the previous value C latched in the C register 15,
It is a straight line that has the same negative slope as the slope of straight line AB, and E is the intersection of the two straight lines. For convenience, each point A
, B, and C are 1.5, 0.5, and -0.5, respectively, and the time of intersection E is assumed to be T. The explanation will proceed.

Since a straight line is expressed by the formula y=at+b, the straight line AB and the straight line BC are respectively y=a1t+b1...(1) y=a2t+b2...(2) However, since a1=-a2, ( 2) The formula is y=-a
It is expressed as 1t+b2...(2)'. Therefore, B=a1×0.5+b1...(3) A=a1×1.
5+b1...(4) B-A=a1×(0.5-1.
5) a1=-(B-A)...(5) Also, C=-a1×(-0.5)+b2...(6) B-C
=a1×0.5+b1-a1×0.5-b2=b1-b
2...(7)

From formulas (1) and (2)', the intersection E is E=a1
T+b1...(1)'E=-a1T+b2...(2)''a1T+b1=
-a1T+b2 2a1T=b2-b1 T=-(b1-b2)/2a1...(8)(5)(
From formula 7) T=-(B-C)/-2(B-A
)=(B-C)/2(B-A)...(9) The time T at the intersection E can be found by equation (9).

Further, in the case of FIG. 4, the situation is as follows. The straight line BC is a straight line passing through the two points of the maximum value B and the previous value C, and the straight line AB
is a straight line that immediately passes through the point of value A and has the same negative slope as the slope of straight line BC. B=a1×0.5+b1 C=a1×(-0.5)+b1 B-C=a1×(0.5+0.5)=a1...(1
0) A=-a1×1.5+b2 B-A=a1×(0.5+1.5)+b1-b2b1-
b2=B-A-2a1...(11) In equation (8), (1
0) Substituting (11), T=-(b1-b2)/2a
1 =-(B-A-2a1)/2a1=1-(B-A)/2
a1=1-(B-A)/2(B-C)...(12)
Therefore, the time T at the intersection E can be found using equation (12). Note that the above equations (9) and (12) are calculated at point A,
This is an example of a formula for calculating the intersection E by finding two straight lines from the coordinate positions B and C, and it is of course possible to process using a formula summarized in another form.

[0020] Examples of specific numerical values of the calculation results obtained by the above processing are shown below. For example, in FIG. 4, A=28, B=
30, assuming C=13, T=1-(30-28)/2×(30-13)=0.9
It becomes 4. On the other hand, when calculating the center of gravity according to the conventional technology using the same numerical values, T = (1.5 x 28 + 0.5 x 30 - 0.5 x 13) /
(28+30+13)=0.71, which is a considerable difference from the calculation result of the example. This is because the range of data used in center of gravity calculations is inappropriate; it is difficult to determine the range of data to be used in center of gravity calculations, and results vary depending on the range.

In the present invention, only the data immediately before and after the peak value are used, so in the case of a three-dimensional object measuring apparatus such as the embodiment, accurate results can be obtained with a low possibility of errors caused by virtual images. In addition, in the embodiment, processing is performed assuming that the signal waveform has a symmetrical shape in the front and back with the peak as the center, but for example, if the front and back become asymmetrical with a certain tendency due to the characteristics of the video camera 10, etc. , it is sufficient to add a procedure to correct this. Furthermore, when asymmetry before and after the peak occurs randomly, instead of using three detected values as in the example, two detected values before and after the peak are used to find the slope of each straight line. do it.

[0022]

[Effects of the Invention] As is clear from the above explanation, the present invention imagines two straight lines having slopes corresponding to the change rate of the signal level before and after the peak value, and finds the position of the intersection of these straight lines. This allows the peak position of the signal to be detected. Specifically, for example, using the maximum detection value detected at a predetermined time interval and each detection value immediately before and after it, a straight line passing through two points, the maximum detection value and the detection value with a larger difference from this, is drawn. , find a straight line that has an opposite slope to the slope of this straight line and passes through one point of the detected value that has a smaller difference from the maximum detected value, calculate the position of the intersection of these straight lines, and determine the point where the signal peaks. This is the time when .

[0023]Therefore, a multiplier, which is required when calculating the center of gravity, is not required, and the calculations are easy, and the configuration of the entire processing circuit is simplified, and the load on the CPU is also reduced. In addition, since there is no need for time-consuming multiplication calculations, the clock speed can be increased, and even if the peak width is narrow, there will be no problems such as insufficient data before and after the peak necessary for calculation, resulting in inaccurate calculations. . Furthermore, since only data immediately before and after the peak value is used, there is an advantage that errors due to noise are reduced. In addition, the calculation to find the intersection of two straight lines can be performed using any appropriate method, but the calculation process can be easily performed by organizing the calculation procedure in an easy-to-use format and storing it in a storage means. can.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of an apparatus for implementing the present invention.

FIG. 2 is a block circuit diagram of one embodiment.

FIG. 3 is an explanatory diagram of calculations in the embodiment.

FIG. 4 is an explanatory diagram of calculations in the same embodiment.

[Explanation of symbols] 10 Video camera 12 A/D converter 13, 14, 15 register 21 Counter 24 Comparator 26, 27 Subtractor 28 ROM 31 CPU A Immediate value B Maximum value C Last value T Peak position

Claims (4)

[Claims] [Claim 1] Detecting the peak position of the signal by imagining two straight lines having slopes corresponding to the rate of change in the signal level before and after the peak value of the signal, and calculating the position of the intersection of these straight lines. A signal peak position detection method characterized by: [Claim 2] Detect the signal level at a predetermined time interval, and when the maximum detected value is obtained, use this maximum detected value and each detected value immediately before and after it to axis the signal level and time. On the two-dimensional diagram where and a second straight line that passes through one point of the detection value that has the smaller difference from the first and second straight lines, calculate the position of the intersection of these first and second straight lines, and take this as the time when the signal reaches its peak. A signal peak position detection method characterized by: 3. Signal level detection means for detecting a signal level at predetermined time intervals; a storage circuit for temporarily storing each detection value immediately before and after the maximum detection value when the maximum detection value is obtained; On a two-dimensional diagram with the axis of and time as the axes, the maximum detected value and the detected value with a larger difference from this two.
A first straight line that passes through the point and a second straight line that passes through one point of the detected value that has an opposite slope to the slope of the first straight line and has a smaller difference from the maximum detected value, and and arithmetic means for calculating the position on the time axis of the intersection point of the second straight line and outputting this as the time when the signal reaches its peak. 4. The storage means is preliminarily stored with a calculation procedure for determining the intersection of the first and second straight lines using the maximum detected value and the relative value between the three detected values immediately before and after the maximum detected value. By inputting the two relative values calculated by the subtracter into the address of the storage means, the storage means outputs a signal corresponding to the decimal place on the time axis at the intersection of the two straight lines. 4. The signal peak position detecting device according to claim 3.