JPH0623644B2 - Edge position detector - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for optically detecting an accurate position of an edge of an object.

(Prior Art) Generally, there is a strong demand for knowing the position of an edge of an object from optical imaging, and it is used in many technical fields for an assembly process or automation of work.

FIG. 3 is a diagram showing a conventional configuration of such an optical edge detecting device. Reference numeral 1 denotes an object to be detected whose edge is to be detected. When the object to be measured reflects light emitted from a light source, the reflected light is sensed and imaged by a line sensor 2 such as a CCD, and its output is a multiplexer (hereinafter ,
After being multiplexed by an MPX 3 and sample-held by a sample-hold (hereinafter referred to as S / H) circuit 4, analog-digital (hereinafter referred to as A / D)
The output of the line sensor 2 is digitally converted. That is, the output of the line sensor 2 including the detection information of the edge of the DUT 1 is taken out as a digital signal. Hereinafter, this digital signal is referred to as the output (or digital output) of the line sensor 2.

The digital output of the line sensor 2 is applied to the subtractor 6 and stored in the first memory 7 for each frame. 8 is a second memory, in which the DUT 1
The output of the line sensor 2 (referred to as a background signal) when there is no signal is stored in advance, and the output is applied to the subtractor 6 and subtracted from the digital signal of the line sensor 2. Therefore, the output of the subtractor 6 is a signal containing the complete edge information of only the DUT 1 from which the background signal of the DUT 1 has been subtracted, which is stored in the third memory 9. An averaging circuit 10 divides the total digital output of the line sensor 2 for one frame by the number of bits for one frame to calculate an average value. Reference numeral 11 denotes an arithmetic unit that calculates a threshold value for edge detection by adding, subtracting, multiplying, and dividing a certain fixed value to the output of the averaging circuit 10. Reference numeral 12 denotes a comparator, which compares the output signal of the third memory 9 with the threshold output of the arithmetic unit 11, and uses the output to associate the digital output of the line sensor 2 with the scanning position to determine the edge position. It is detecting. Reference numeral 13 denotes a microcomputer (hereinafter referred to as CPU), which calculates the position of the edge of the DUT 1 from the output of the comparator 12. 14 is a data bus.

FIG. 4 shows the result of edge detection by the configuration of FIG. 3 described above, where a is the output of the subtracter 6, which is the output of the line sensor 2 after subtracting the background signal. b is a threshold value of the output of the calculator 11. Therefore, the edge position detected by the DUT ₁ becomes the bit position m ₁ at which the output exceeds the threshold value b.

(Problems to be Solved by the Invention) However, in the conventional optical edge detection device, since the edge position is determined by whether or not a certain threshold is exceeded, the cross section of the edge of the DUT 1 is curved. In the case of a spherical surface, there is a problem that the detected edge position lacks accuracy. That is, as shown in FIG. 5, although the correct edge position is the bit position m ₃ , the cross section of the edge changes gently, so that the digital output gradually increases and exceeds the threshold value b. m ₂ is detected as the edge position.

The present invention aims to provide an excellent optical edge position detecting device which eliminates the problems of the conventional device.

(Means for Solving the Problems) In order to achieve the above-mentioned object, the present invention is provided with a differentiator and an integrator in addition to the configuration of FIG. The difference between the adjacent bits of the output is taken, and whether the difference is positive or negative is judged. When this difference takes a positive value, the difference value is integrated by an integrator, the bit position P is temporarily stored, and a negative value is stored. When taking the value of, the integrator subtracts a constant value for integration, determines whether this integrated value has reached a constant first threshold value, and if it has not reached the first threshold value. To release the memory, first
When the threshold value is reached, the bit position is further changed and integration is performed. When the integrated value reaches a second threshold value which is larger than the first threshold value, the bit position P is set to the edge. The position is determined.

(Operation) According to the present invention, when the difference between the digital outputs of adjacent bits is taken and it is determined that this digital output has increased, this can be detected as an edge position, and the digital output is constant. After the value is exceeded, the detected bit position is determined as the edge position even if the digital output suddenly decreases suddenly.

(Example) Hereinafter, the Example of this invention is described in detail, referring drawings.

FIG. 1 is a diagram showing the configuration of an embodiment of the present invention, and the same reference numerals as those in FIG. 3 denote the same parts.

Reference numeral 1 is an object to be detected whose edge is to be detected. When the object to be measured reflects the light emitted from the light source, the reflected light is CC
It is sensed and imaged by the line sensor 2 such as D, and the output is multiplexed by the MPX3 and S /
It is sampled and held by the H circuit 4 and is digitally converted by the A / D conversion circuit 5. That is, the output of the line sensor 2 including the detection information of the edge of the DUT 1 is
It is converted into a digital signal, applied to the subtractor 6, and stored in the first memory 7. This first memory 7
Stores the digital output of the line sensor 2 for each frame. In the subtractor 6, the background signal of the DUT 1 stored in advance in the second memory 8 is A /
After subtracting from the output of the D conversion circuit 5, the edge information signal of only the DUT 1 is output and stored in the third memory 9. Reference numeral 15 is a differentiator, which controls the data of the contents of the third memory 9 by the CPU 13 to sequentially detect the difference between adjacent bits of the digital output of the line sensor 2, and the difference value is detected by the integrator 16. Integrated. The integration at this time is performed by directly integrating when the difference value is positive and by subtracting a constant value when the difference value is negative. The values thus integrated are compared with the values in the fourth memory 19 and the fifth memory 20 by the comparators 17 and 18, respectively. This fourth memory 19
The fifth memory 20 stores a threshold value 1 for determining an edge and a threshold value 2 higher than the threshold value, and the CPU 13 controls the data bus 14 and the memory 9 and calculates data. The algorithm control shown in FIG. 2 is performed to detect the edge.

FIG. 2 is a flow chart for explaining the edge detection algorithm of the embodiment of the present invention shown in FIG.
(However, i is an arbitrary positive integer, the same applies hereinafter) is the bit position of the digital signal output of the line sensor 2, D _i is the i-th output of the line sensor 2, K is the difference value between adjacent bit outputs, and TH ₁ and TH ₂ are the first and second thresholds, respectively, and Me is the memory.

The operation of this embodiment will be described with reference to FIG. First, the difference value K between adjacent bit outputs is calculated from K = D _{i + 1} −D _i from the first bit, which is the end of the line sensor 2, to determine whether the difference value K is positive or negative.

Here, when the difference value is positive, the difference value K is integrated to obtain the integrated value S as S = S + K, and it is determined whether the value of the flag Fl is 0 or 1. When the value of the flag Fl is 0, the bit position i is stored in the memory Me, the value of the flag Fl is set to 1, and the bit position i is incremented by 1. On the other hand, when the value of the flag Fl is 1,
Increase the bit position by one.

That is, when the adjacent bit output increases, the bit position i is stored.

When the difference value is not positive, the difference value K is integrated to obtain the integrated value S as S = S−a (where a is a constant value), and the integrated value S and the first threshold value TH ₁ are obtained. Judge the relationship between When the integrated value S is smaller than the first threshold value TH ₁ , the integrated value S is set to 0, the flag Fl is set to 0, and the bit position i is incremented by 1. On the other hand, when the integrated value S is not smaller than the first threshold value TH ₁ , the bit position is incremented by 1. That is, when the adjacent bit output decreases or is the same and the integrated value S obtained by subtracting a constant value does not exceed the threshold value TH ₁ , the flag Fl becomes 0 and the memory stored before that is updated. In the active state and the adjacent bit outputs decrease or are the same,
When the integrated value S obtained by subtracting a constant value exceeds the threshold value TH ₁ , the bit position is updated as it is.

By the above operation, when the integrated value S is larger than the threshold value TH _{1, the} bit position stored in the memory Me is unchanged, but when it is smaller than the threshold value TH ₁ , the memory Me stores the difference value. When K goes positive, it is updated to the new bit position.

Further, it is judged whether the integral value S is larger or smaller than the threshold value TH ₂ , and if it is smaller, the difference value K of the next bit position is obtained, and if it is larger, the bit position stored in the memory Me. Finally, i is detected as an edge position.

The present invention has been described above. In the second flowchart, the bit position i is described starting from the first bit.
If similar edges are detected in the opposite direction from the maximum bit position of the line sensor, both edges of the edge can be detected.

(Effects of the Invention) As described above, according to the present invention, the position where the change in the line sensor output appears is determined to be the edge, and therefore even when the cross section of the edge gradually changes like a sphere. The advantage is that the position of the edge can be detected accurately and there is no error. In addition, it is possible to prevent the line sensor output from abruptly changing and erroneously detecting an edge even if an object exists.

[Brief description of drawings]

FIG. 1 is a block diagram of an edge position detecting apparatus showing an embodiment of the present invention, FIG. 2 is a flow chart of an edge detecting algorithm, FIG. 3 is a block diagram of a conventional edge position detecting apparatus, FIGS. The figure shows the relationship between the line sensor output and the determination threshold value. 1 ... DUT, 2 ... Line sensor, 3 ... Multiplexer (MPX), 4 ... Sample and hold circuit (S /
H circuit), 5 ... Analog-digital conversion circuit (A / D conversion circuit), 6 ... Subtractor, 7, 8, 9, 19, 20 ... Memory, 13 ... Microcomputer (CPU), 14 ... Data bus, 15 ... differentiator, 16 ... integrator, 17, 18 ... comparator.

Claims (1)

[Claims] 1. A line sensor for scanning an object to be measured, a differentiator for sequentially detecting a difference between outputs at adjacent scanning positions on the line sensor along a scanning direction, and an integral or a constant value of the difference. Is provided with a storage means for storing the scanning position determined to be an edge, scans the DUT from one end of the line sensor to the other end, and When the differential result is positive, this scanning position is stored in the storage means as the edge, and the difference is integrated in the integrator,
On the other hand, when the result of differentiation by the differentiator is negative or 0, the integrator subtracts a constant value for integration, and if the result of integration is less than or equal to the first threshold value, the stored contents are retained. If the integration result is less than or equal to the first threshold value, the integration value is set to 0, and a new scanning position where the differentiation result is positive is stored in the storage means again, and the integration result by the integrator is An edge detection position, wherein a scanning position stored in the storage means is set as the edge when a second threshold value larger than the first threshold value is exceeded.