JPH07229715A - Edge position detector - Google Patents

Abstract

PURPOSE:To reduce the detection error at edge position as a result of the variation of the level of quantity of bright and dark light for improving measuring precision by providing means for mainly detecting edges and calculating threshold value. CONSTITUTION:The edge detecting mans 5 is provided with a threshold value memory 8 to memorize threshold value Th as the reference to detect edges E1, E2. In accordance with quantity of light received by light receiving elements 11-1n, a threshold value calculating means 12 calculates the threshold value Th. When quantity of light from a light source 100 is varied, the threshold value Th is changed with the variation and so a detection error at an edge position is reduced. Instead that the threshold value Th as the reference to detect the edges E1, E2 is made always constant, the threshold value Th is calculated in accordance with quantity of light received by the receiving elements 11-1n. Therefore, when there is disturbing light or dirt on a window, or when the level of quantity of light from the light source 100 is varied, the threshold value Th can be changed with the variation and so a detection error at the edge position is reduced, resulting in more measuring precision.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an edge position detecting device for detecting an edge by irradiating an object with light.

[0002]

2. Description of the Related Art Conventionally, there is a dimension measuring instrument which detects the position of an edge by using a one-dimensional image sensor and measures the dimension of the object, the interval between the objects, and the position based on the edge position. An example of this type of measuring device is shown in FIG.

In FIG. 6, a light source 100 can be regarded approximately as a point light source and emits a laser beam L such as a visible laser beam. The laser light L becomes parallel light L1 by passing through the collimator lens 101, and is irradiated onto the object 102, and a part (transmission light) of the parallel light L1 is irradiated onto the one-dimensional image sensor 1. The one-dimensional image sensor 1 includes light receiving elements 1 ₁ to 1 _n such as CCDs.
Are arranged in a linear fashion and convert light energy into electric signals. Signal processing circuit and microcomputer not shown (hereinafter referred to as "microcomputer")
Detects the positions of the edges E ₁ and E ₂ of the object 102 based on the outputs from the light receiving elements 1 ₁ to 1 _n , for example, the dimension A of the object 102, that is, two predetermined edges E.
Measure the dimensions based on ₁ and E ₂ .

[0004]

Here, the edge addresses X ₁ and X _{2 in} FIG. 7, which are the reference of the edge position, first indicate the amount of light received by each of the light receiving elements 1 ₁ to 1 _n (FIG. 6). It is converted into a digital value, and as shown in FIG. 7A, it is obtained by comparing a preset constant threshold value with the digital value. It is said that the threshold value is preferably set to a level of about 25% from the level between "bright" and "dark". However, as shown in FIGS. 7A and 7B, the amount of light received by the light receiving element 1 _i , for example, the “bright” level is
It fluctuates according to ambient light, stains on windows, changes in the amount of light from the light source 100 (FIG. 6), and the like. Therefore, when the threshold level is fixed to a constant value as shown by the broken line, the positions of the edge addresses X ₁ and X ₂ are slightly different between FIGS. 7A and 7B, and as a result, the edge An error occurs in the detection position of.

The present invention has been made in view of the above-mentioned conventional problems, and an object thereof is to reduce the detection error of the edge position caused by the fluctuation of the level of the light quantity of bright and dark and improve the measurement accuracy. It is to provide an edge position detecting device capable of performing the above.

[0006]

In order to achieve the above object, the present invention provides an edge detecting means for obtaining the position of an edge based on the amount of light received by a light receiving element and a threshold value serving as a reference for detecting the edge. And a threshold value calculating means for calculating the threshold value based on the amount of light received by the light receiving element.

According to the present invention, the threshold serving as a reference for detecting an edge is calculated based on the amount of light received by the light receiving element. For example, the level of “bright” is shown in FIGS.
If it falls like the above, the threshold value is lowered to the level shown by the chain double-dashed line. Therefore, even if the "bright" level changes as shown in FIGS. 7A and 7B, the edge addresses X ₁ and X
_{Since 2} does not change, the measurement error of the edge position becomes small.

In the present invention, a histogram creating means for calculating the number of data for each digital value from the digital value obtained by digitally converting the light amount received by the light receiving element is provided, and the threshold value calculating means has the number of data for each digital value. It is preferable to obtain the threshold value from

Further, it is preferable that the threshold value calculating means calculates the threshold value based on at least a digital value indicating a peak value of the number of data in "bright".

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, the one-dimensional image sensor 1 is
A CCD is generated by a reset signal, a start signal and a shift signal output from a timing generator (not shown).
(Light receiving element) The photoelectric conversion of the light L1 incident on the light receiving element 1 _i is performed while repeating scanning in a fixed cycle in the array direction of 1 ₁ to 1 _n . The output from the one-dimensional image sensor 1 is input to the A / D converter 3 via the sample hold circuit 2. The A / D converter 3 includes light receiving elements 1 _i to 1 _n.
The amount of light received by is converted into a digital value D, and this digital value D is output to the light amount data memory 4 and the edge detection means (circuit) 5. The light amount data memory 4 has a structure shown in FIG.
As shown in (a), a digital value (light quantity) is stored corresponding to the address of the light receiving element.

The edge detecting means 5 in FIG. 1 comprises a digital comparator 6 and a threshold memory 8. The threshold memory 8 stores a threshold Th that serves as a reference for detecting the edges E ₁ and E ₂ , and the digital comparator 6 reads the threshold Th from the threshold memory 8,
The digital value D and the threshold value Th are compared and binarized.
That is, the digital comparator 6 includes the light receiving elements 1 ₁ to ₁ 1.
The amount of light received by _n in FIG. 4A is compared with the threshold Th, and the level above the threshold Th indicated by the broken line is set to "bright",
The lower level is "dark". The edge detecting means 5 of FIG. 1 successively compares the binarized information (bright or dark information) of the i-th address and the (i + 1) -th address, and detects the edge ( The boundary between light and dark) E
_It is determined that _i exists, and the address X _{i of the} edge E _i is determined.
Are stored in the edge address memory 7.

The digital value D from the A / D converter 3 is output to the histogram creating means 30.
As shown in FIG. 2B, the histogram creating means 30 creates the histogram of FIG. 4B by calculating the number of light receiving elements (the number of data) that have received the light amount corresponding to each digital value D. The histogram memory 31, the one increment adder 32, and the latch circuit 33 of FIG. 1 are provided. The histogram creating means 30 creates a histogram according to the flowchart of FIG. 3 described later. Further, in the signal processing circuit configured as described above, illustration and description of various amplifiers are omitted.

Information of the light amount data memory 4, the edge address memory 7 and the histogram memory 31 of FIG. 1 is read by the CPU 10. CPU 10
Includes a dimension calculation means 11, a threshold value calculation means 12, and a peak position calculation means 13.

The dimension calculating means 11 reads the edge addresses X ₂ and X ₁ which are the reference for dimension measurement from the edge address memory 7 and calculates the dimension A of the object 102, for example, according to the following equation (1). A = (X ₂ −X ₁ ) · P (1) where P is the pitch of the light receiving elements

The peak position calculating means 13 is shown in FIG.
The number M of light-receiving elements is read from each address of the histogram memory 31 in (b), and the address having the largest number M of light-receiving elements in the “bright” and “dark” regions, that is, “bright” in FIG. 4B. , And “dark”, digital values (peak positions) D _P1 and D _{P2 at} which the number M of light receiving elements reaches a peak are obtained.

The threshold value calculating means 12 of FIG. 1 obtains a threshold value Th from the above two peak positions D _P1 and D _P2 according to the following equation (2) for each scan of the one-dimensional image sensor 1, for example. Th = D _P2 + (D _P1 −D _P2 ) / 4 (2) Thus, the threshold value calculation means 12 calculates the threshold value Th based on the amount of light received by the light receiving elements 1 ₁ to 1 _n, and the calculated threshold value Th. Th is written in the threshold memory 8 described above. Therefore, the threshold Th is rewritten each time the one-dimensional image sensor 1 is scanned.

Next, a method of creating the above-mentioned histogram will be described with reference to the flowchart of FIG. First, in step S1, the histogram creating means 30 in FIG. 1 clears the data in the histogram memory 31, and then the process proceeds to step S2 in FIG. In step S2,
The address of FIG. 2B corresponding to the digital value D of the A / D converter 3 (FIG. 1) is searched, and step S3 (FIG. 3) is searched.
Then, the process proceeds to step 1 to read the data (the number of light receiving elements) a of the address. Subsequently, the process proceeds to step S4 (FIG. 3), the data a is input to the one-increment adder 32 of FIG. 1, and the one-increment adder 32 outputs (a + 1) to the latch circuit 33 in step S5 (FIG. 3). . The latch circuit 33 holds (a + 1) for a minute time in step S6 (FIG. 3), proceeds to step S7,
Write (a + 1) in the histogram memory 31 and
Then, the process proceeds to step S8. In step S8, it is determined whether or not one scan of the one-dimensional image sensor 1 of FIG. 1 is completed. If one scan is not completed, the process returns to step S2 of FIG. 3, while if it is completed, Proceeding to step S9, the creation of the histogram is completed.

The other structure is the same as that of the conventional example shown in FIG. 6, and the same portions or corresponding portions are designated by the same reference numerals and the description thereof will be omitted.

Next, a method of detecting the edge addresses X ₁ and X ₂ will be described. In FIG. 1, for example, when an object 102 is transferred in the direction of arrow B, the light source 100
The laser beam L1 is emitted to the one-dimensional image sensor 1, and the one-dimensional image sensor 1 repeats scanning at a constant cycle. The digital value D is output from the A / D converter 3 to the light amount data memory 4, the edge detecting means 5, and the histogram creating means 30 for each scan. With this output,
The histogram creating means 30 creates a histogram (FIG. 2B) by the method described above, and the peak position calculating means 13 reads the number M of light receiving elements from the histogram memory 31 and calculates the peak positions D _P1 and D _P2 . After this calculation, based on the peak positions D _P1 and D _P2 , the threshold value calculation means 12 calculates the value shown in FIG.
The threshold value Th in (a) is calculated, and the threshold value Th is input to the threshold value memory 8 in FIG. 1 and rewritten.

The digital comparator 6 compares the rewritten threshold value Th with the digital value D input in the next scanning, and binarizes the digital value D. After the binarization, the edge detection means 5 determines that the edge is at the position where the binarized signal changes, and stores the edge address X _i in the edge address memory 7. The dimension calculation means 11 is
The dimension A of the object 102 based on the edge address X _i
Is calculated and the dimension A calculated when the object 102 is conveyed to a predetermined measurement position is output to a display (not shown) or the like.

Here, the threshold Th in FIG. 4A is
It is preferable to set a level of about 25% from the level between "bright" and "dark". Therefore, as shown in FIGS. 7A to 7B, for example, when the level of “bright” changes, the level of the threshold Th also needs to be changed as shown by the chain double-dashed line.

On the other hand, in this dimension measuring device, the threshold value calculating means 12 calculates the threshold value Th in FIG. 4A based on the amount of light received by the light receiving elements 1 ₁ to 1 _{n in} FIG. Therefore, when the amount of light from the light source 100 in FIG. 1 fluctuates, the threshold Th is changed in accordance with the fluctuation, so that the edge position detection error is reduced. As a result, the accuracy of dimension measurement is improved.

By the way, in the above embodiment, the threshold Th is
Although the calculation is performed for each scan of the one-dimensional image sensor 1, the threshold Th may be calculated for each of a plurality of scans in the present invention. The threshold Th may be calculated only in the scan immediately before the object 102 arrives at the measurement position shown in FIG. Alternatively, the threshold Th may be calculated for each scanning, a plurality of past thresholds Th may be stored, the average value Th _AVE may be calculated, and binarization may be performed based on the average value Th _AVE. .

By the way, in the above-mentioned embodiment, the threshold value Th is obtained by making the histogram of FIG. 4A, but the present invention does not necessarily need to make the histogram. For example, the center C in “bright” and “dark” in FIG.
1, C2, C3 are obtained, and a large number of light receiving elements 1 _i before and after that are obtained.
It is also possible to detect the levels of “bright” and “dark” from the average value D _AVE of the light amount and calculate the threshold Th based on the detected values.

However, in such a method, the calculation is generally troublesome and the accuracy may be lacking. Therefore, as in the present embodiment, a histogram is prepared and the threshold values are calculated from the peak positions D _P1 and D _P2. It is preferable to determine Th.

Further, in this embodiment, "bright" and "dark"
Both peak positions D _P1 and D _P2 are obtained, but it is not always necessary to obtain both peak positions D _P1 and D _P2 , and one peak value may be obtained. For example, the "dark" level may be stable in a place where ambient light is not likely to enter, and in such a case, the "dark" level is considered to be a constant value and the "bright" level is considered to be constant. The threshold value Th may be calculated by obtaining only the peak position D _{P1 of}

When a laser is used as the light source 100 in FIG. 1, the laser light L1 diffracted by the edges E ₁ and E ₂ of the object 102 causes interference. Therefore, the received light waveform of the one-dimensional image sensor 1 is disturbed as shown in FIG. 5A and spreads in the range of “bright” level as shown in FIG. 5B. Here, in this embodiment, since the peak position D _P1 is obtained from the histogram, the “bright” level can be obtained accurately. Therefore, the edge position can be obtained more accurately.

The present invention can also be applied to a scanning type size measuring device for measuring the size of a shadow from time by scanning a thin laser beam or a linear laser beam with a galvano mirror or a polygon mirror.

Further, although the transmission type edge position detecting device has been described in the above embodiment, the present invention can be applied to a reflection type edge position detecting device in which reflected light from an object is received by a one-dimensional image sensor. . Further, it goes without saying that the present invention can be applied not only to the distance and position between objects but also to the dimension measurement of a specific portion of the object.
Further, the light receiving element is not limited to CCD but may be CPD or the like, and the signal processing circuit including the edge detecting means 5 and the histogram creating means 30 in FIG. On the contrary, the threshold value calculation means 12 and the peak position calculation means 13 of the CPU 10 may be configured by a signal processing circuit.

[0030]

As described above, according to the present invention,
The threshold value that is the reference for detecting edges is not always set to a constant value, but the threshold value is calculated based on the amount of light received by the light receiving element.Therefore, ambient light, window stains, or the amount of light from the light source When the level changes, the threshold value can be changed according to the change, so that the detection error of the edge position becomes small. Therefore, the measurement accuracy is improved.

If a histogram is created as in the second aspect of the invention, the threshold value can be calculated easily.

Further, as in the third aspect of the present invention, if the threshold value is calculated based on at least the digital value indicating the peak value in "bright", the threshold value can be accurately set even if the received light waveform is disturbed by interference. Since it can be obtained, the detection error of the edge position becomes small.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a dimension measuring device according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram showing a memory map of a light amount data memory and a histogram memory.

FIG. 3 is a flowchart showing an operation of a histogram creating means.

FIG. 4 is a conceptual diagram showing an edge detection method.

FIG. 5 is a conceptual diagram similarly showing an edge detection method.

FIG. 6 is a schematic configuration diagram of a general dimension measuring device.

FIG. 7 is a conceptual diagram showing a conventional edge detection method.

[Explanation of symbols]

1: One-dimensional image sensor 12: Threshold calculation means 13: Peak position calculation means 5: Edge detection means 30: Histogram creation means 102: Object D: Digital value Th: Threshold values D _P1 , D _P2 : Peak position

Claims (3)

[Claims] 1. An edge position detecting device, which irradiates an object with light, causes the light-receiving element to receive the transmitted light or reflected light, and detects the edge position based on the output from the light-receiving element. An edge detection unit that obtains the position of the edge based on the light amount received by the light receiving element and a threshold value that serves as a reference for detecting the edge, and a threshold value calculation unit that calculates the threshold value based on the light amount received by the light receiving element. Edge position detector. 2. The method according to claim 1, further comprising a histogram creating means for calculating the number of data for each digital value from a digital value obtained by digitally converting the amount of light received by the light receiving element, and the threshold value calculating means for each digital value. An edge position detecting device, characterized in that the threshold value is obtained from the number of data. 3. The edge position detecting device according to claim 2, wherein the threshold value calculating unit calculates the threshold value based on the digital value at which the number of data in “bright” has a peak.