JPH05288534A - Inclination detecting method - Google Patents

Abstract

PURPOSE:To provide the inclination detecting method rarely affected by the noises due to a nozzle, dust, and intensity change, capable of handling various electronic parts in common, and having high flexibility. CONSTITUTION:The inclination detecting method has the first process #1 storing the variable-density image of a recognition object in a memory as a two-dimensional multi-value image, the second process #2 setting multiple sampling points on the two-dimensional multi-value image, the third process $4 determining the deflection angle added with the preset angle in the direction toward the maximum intensity gradient and the size of the intensity gradient, the fourth process #5 generating the deflection angle histogram indicating the maximum intensity gradient occurrence frequency for each deflection angle, the fifth process #7 displacing the deflection angle histogram of the fourth process #5 by the preset angle and accumulating it to generate the accumulated deflection angle histogram, and the sixth process #8, #9 determining the deflection angle having the maximum intensity gradient occurrence frequency in the deflection angle histogram of the fourth process #5 and the accumulated deflection angle histogram of the fifth process #7 as the inclination of the recognition object.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tilt detecting method in an electronic component mounting machine, in which an electronic component is imaged and a mounting correction angle of the electronic component is obtained.

[0002]

2. Description of the Related Art An electronic component mounting machine for mounting electronic components on a substrate has been using a component positioning method to which visual recognition technology is applied. In the conventional visual recognition method in the electronic component mounting machine, a binary processing method for separating the component body and the background by using a threshold unique to the component is mainstream, but recently, a grayscale image has 64 gradations, A multi-valued process of expressing with 128 gradations or 256 gradations is being started. As a conventional technique of the inclination detection method by the multi-valued processing, there is a method by one-dimensional edge detection.

A conventional example of the inclination detecting method by the above-mentioned multi-valued processing will be described below with reference to FIGS.

FIG. 20 shows a two-dimensional multi-valued image 10 of an electronic component of a recognition object, line segments 11 and 12 consisting of points for performing one-dimensional edge detection, and one-dimensional edge detection of the line segments 11 and 12. The illustrated edge points 13 and 14 are shown.

FIG. 21 shows a two-dimensional multi-valued image 10 of an electronic component as a recognition target, line segments 11 and 12 formed by points for performing one-dimensional edge detection, and luminance distribution on the line segments 11 and 12.
Second-order differential value of luminance on the line segments 11 and 12, and point of second-order differential value 0 corresponding to the position of the edge point 13 or 14 of the two-dimensional multi-valued image 10 of the electronic component on the line segment 11 or 12 16
(13, 14) are shown.

Next, the one-dimensional edge detection method will be described with reference to FIGS.

In FIG. 20, a line segment 11 composed of points for which one-dimensional edge detection is performed so that two edge points can be detected on one side of a two-dimensional multi-valued image 10 of an electronic component to be recognized. , 12 are set. For the line segments 11 and 12, edge points 13 and 14 of the two-dimensional multi-valued image 10 are obtained on one side of the two-dimensional multi-valued image 10 of the electronic component of the recognition target by detecting the respective one-dimensional edges. Be done. The inclination of the straight line 15 obtained by connecting the edge points 13 and 14 is the inclination of the electronic component.

FIG. 21 illustrates the one-dimensional edge detection. The two-dimensional multi-valued image 1 of the electronic component of the recognition object is shown in FIG.
At 0, the line segment 11 or 12 is set so as to cross the edge of the two-dimensional multi-valued image 10 of the electronic component in which the brightness changes abruptly. Next, when the luminance distribution on this line segment 11 or 12 is obtained, the distribution becomes like the luminance distribution curve of FIG. 21, and at the position where it crosses the edge of the two-dimensional multi-valued image 10 of the electronic component, The brightness changes rapidly. Further, when the secondary differential value of the brightness distribution on the line segment 11 or 12 is obtained, the secondary differential value becomes like the secondary differential value curve of the brightness in FIG. 21, and the secondary differential value is A point 16 (13, 1, 1) with a secondary differential value of 0 that greatly changes from a negative value to a positive value and becomes 0 in the middle
4) is required. This point 16 is taken as the position of the edge of the two-dimensional multi-valued image 10 of the electronic component of the recognition object, and the inclination of the straight line connecting the two points 16 is defined as the inclination of the electronic component of the recognition object. Is the one-dimensional edge detection method.

[0009]

However, in the above-described conventional method, as shown in FIG. 22, the suction nozzle 17 that vacuum-sucks the electronic component 10 is on the line segment 12 for performing one-dimensional edge detection. 23, or when dust 19 is attached to a part of the electronic component 10 as shown in FIG. 23, or when a large amount of dust is present on a part of the surface of the electronic component 10 as shown in FIG. In the case where there is the brightness changing portion 21, there is a problem that the edge point obtained by the one-dimensional edge detection method may not match the position of the edge of the electronic component.

In the case of FIG. 22, the edge point 13 obtained by the one-dimensional edge detection method for the line segment 11 coincides with the position of the edge of the electronic component 10, but 1 for the line segment 12.
The edge point 14 ′ obtained by the dimension edge detection method does not match the position of the edge of the electronic component 10 (edge point 14), and the detected inclination of the straight line 18 is different from the inclination of the electronic component 10.

In the case of FIG. 23, the edge point 14 obtained by the one-dimensional edge detection method for the line segment 12 coincides with the position of the edge of the electronic component 10, but 1 for the line segment 11.
The edge point 13 ′ obtained by the dimension edge detection method does not match the edge position (edge point 13) of the electronic component 10, and the detected inclination of the straight line 20 is different from the inclination of the electronic component 10.

In the case of FIG. 24, the edge point 14 obtained by the one-dimensional edge detection method for the line segment 12 coincides with the position of the edge of the electronic component 10, but 1 for the line segment 11.
The edge point 13 ′ obtained by the dimensional edge detection method does not match the edge position (edge point 13) of the electronic component 10, and the detected inclination of the straight line 22 is different from the inclination of the electronic component 10.

In addition, the shape of the edge of the electronic component is different, as shown in FIG.
As shown in FIG. 7, when the edge shape of the electronic component 10 is complicated, the set number and position of the line segment 23 configured by the points for performing the one-dimensional edge detection, and the straight line 25 from the obtained edge point 24. It is necessary to study the calculation method of the inclination of each time according to its shape, and there is a problem that the processing becomes complicated and takes time.

The present invention solves the above-mentioned problems and provides a tilt detecting method which is less affected by noise due to nozzles, dust, brightness changes, etc. and has a high versatility that can handle a wide variety of electronic components in common. That is the issue.

[0015]

In order to solve the above-mentioned problems, a tilt detecting method according to the first invention of the present application comprises a first step of storing a grayscale image of a recognition target object in a memory as a two-dimensional multi-valued image. In the second step of setting a plurality of sampling points on the two-dimensional multi-valued image, and at each sampling point set in the second step, the luminance gradient is in all directions on the two-dimensional multi-valued image. A third step of obtaining a deviation angle obtained by adding a predetermined angle to the maximum direction and a magnitude of the brightness gradient, and from the processing result of the third step, the magnitude of the brightness gradient is used as a frequency weighting for each deviation angle. A fourth step of creating a declination histogram representing the maximum luminance gradient occurrence frequency, a declination angle at which the maximum luminance gradient occurrence frequency is maximum in the declination histogram of the fourth step is obtained, and the declination angle at which this maximum is obtained is recognized. The fifth step, which is the inclination of the object Characterized in that it has.

In order to solve the above-mentioned problems, the inclination detecting method of the second invention of the present application comprises a first step of storing a grayscale image of a recognition object as a two-dimensional multivalued image in a memory, and the two-dimensional multivalued image. In the second step of setting a plurality of sampling points on the upper side, and in each of the sampling points set in the second step, a predetermined direction is set in the direction in which the brightness gradient is maximum among all directions on the two-dimensional multi-valued image. From the processing result of the third step of obtaining the deviation angle and the magnitude of the luminance gradient to which the angle is added, the maximum luminance gradient occurrence frequency for each deviation angle is determined using the magnitude of the luminance gradient as the frequency weight. A fourth step of creating a declination histogram represented, a fifth step of displacing the declination histogram of the fourth step by a predetermined angle and accumulating to create a cumulative declination histogram; and a fifth step of the fifth step. The highest in the accumulated declination histogram Seeking deflection angle luminance gradient occurrence frequency is maximum, and having a sixth step of the deflection angle becomes the maximum and the inclination of the object to be recognized.

In the tilt detecting method of the second invention of the present application, in order to solve the above problems, the predetermined angle in the fifth step is 90.
It is preferably an integral multiple of degrees.

[0018]

In the tilt detecting method of the first invention of the present application, in the first step,
A grayscale image of the recognition object is stored in a memory as a two-dimensional multi-valued image, a plurality of sampling points are set on the two-dimensional multi-valued image in the second step, and set in the second step in the third step. At each of the sampled points, a deviation angle obtained by adding a predetermined angle to the direction in which the brightness gradient is maximum in all directions on the two-dimensional image multi-valued image and the magnitude of the brightness gradient are calculated.
In the step, a deviation angle histogram representing the maximum frequency of occurrence of the maximum brightness gradient for each deviation angle is created from the processing result of the third step using the magnitude of the brightness gradient as a frequency weight, and in the fifth step, the fourth step. When the argument having the maximum frequency of occurrence of the maximum luminance gradient is obtained in the argument histogram, the argument having the maximum is the inclination of the recognition object. The reason is as follows.

In a two-dimensional multi-valued image of an object to be recognized such as an electronic component, the direction in which the brightness gradient of the two-dimensional multi-valued image is maximum at a point corresponding to the edge of the electronic component is usually the edge of the electronic component. Since the direction is perpendicular to, the direction obtained by adding an odd multiple of 90 degrees to the direction in which the brightness gradient is maximum is the tangential direction of the edge of the electronic component, that is, the direction in which the edge extends.

Therefore, the predetermined angle in the third step is 90
If it is an odd multiple of 90 degrees (the predetermined angle is an angle formed by the direction in which the brightness gradient is maximum and the tangential direction of the edge of the electronic component, it may be other than an odd multiple of 90 degrees). For each sampling point set in two steps, the declination obtained by adding this predetermined angle to the direction in which the brightness gradient is maximized indicates the tangential direction of the edge of the electronic component near each sampling point. Therefore, the declination direction in which the maximum luminance gradient occurrence frequency obtained in the fourth step in the declination histogram is maximum indicates the direction of the edge of the electronic component. Even if the electronic component has a complicated shape, since there is a linear edge at some portion, the direction can be detected with this linear edge.

In this case, there is a nozzle image in the two-dimensional multi-valued image, dust adheres to a recognition target such as an electronic component, or there is a partial change in brightness of the recognition target such as an electronic component. However, since all of them are partial, the number of data obtained from them is very small compared to the number of data obtained from the whole, so there is almost no effect and the above-mentioned predetermined The versatility is high because a wide variety of electronic components can be handled in common by simply determining the appropriate angle.

In addition to the operation up to the fourth step of the tilt detecting method of the first invention of the present application, the tilt detecting method of the second invention of the present application, in addition to the operation of the tilt angle of the fourth step of the fifth step in the fifth step, The accumulated deviation angle histogram is created by displacing by minutes (the angle between a plurality of straight line portions on the two-dimensional multi-valued image) and accumulated, and in the sixth step, the maximum is obtained in the accumulated deviation angle histogram of the fifth step. When the declination that maximizes the frequency of occurrence of the brightness gradient is calculated, this declination that is the maximum is the inclination of the recognition target.
The reason is as follows.

Usually, the outer shape of an object to be recognized such as an electronic component is
The rectangular parallelepiped is used as a reference, and even if a complicated shape is added to a part, the direction in which most of the edges of the outer peripheral portions extend coincides with any of the four sides of the rectangular parallelepiped. Therefore, in the deviation histogram showing the maximum luminance gradient occurrence frequency for each deviation obtained in the fourth step, there are four deviation angles corresponding to the directions of the four sides of the rectangular parallelepiped that are different by integer multiples of 90 degrees. In, the frequency of occurrence of the maximum luminance gradient for each declination is large. Therefore,
If, in each argument histogram, the argument having the maximum frequency of occurrence of the maximum brightness gradient is not clear, a predetermined angle (angle between a plurality of straight line portions on the two-dimensional multi-valued image) is determined in the fifth step.
Is an integral multiple of 90 degrees, and the deviation angle histogram is displaced by an integral multiple of 90 degrees to create an accumulated deviation angle histogram. In this accumulation deviation angle histogram, the maximum luminance gradient occurrence frequency is the maximum. The declination angle becomes clearer than each declination histogram before accumulation, and the inclination of the recognition object can be clearly determined.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the inclination detecting method of the present invention is shown in FIG.
~ It demonstrates based on FIG.

FIG. 1 is a flow chart of a method of an embodiment of the inclination detecting method of the present invention.

In the flowchart of FIG. 1, in step # 1, the grayscale image of the recognition object is stored in the memory as a two-dimensional multivalued image, and the process proceeds to step # 2.

In step # 2, a plurality of sampling points are set on the two-dimensional multi-valued image, and the process proceeds to step # 3.

In step # 3, it is determined whether or not the processing of the next steps # 4 and # 5 is completed for all the sampling points set in step # 2. If completed, the process proceeds to step # 6. If not, step # 4
Proceed to.

In step # 4, at each sampling point set in step # 2, a deviation angle obtained by adding a predetermined angle to the direction in which the brightness gradient is maximum among all directions on the two-dimensional multi-valued image. And the magnitude of the brightness gradient are obtained, and the process proceeds to step # 5.

In step # 5, from the processing result of step # 4, a deviation angle histogram representing the maximum frequency of occurrence of the brightness gradient for each argument is created using the magnitude of the brightness gradient as a frequency weight, and the process returns to step # 3. ..

In step # 6, if the deviation angle histogram created in step # 5 clearly indicates the deviation angle at which the maximum luminance gradient occurrence frequency is maximum, it is judged that accumulation of the deviation angle histogram is unnecessary. Proceeding to step # 8, if the deviation angle histogram created in step # 5 does not clearly indicate the deviation angle at which the maximum luminance gradient occurrence frequency is the maximum, it is determined that accumulation of the deviation angle histogram is necessary, and step # 8 is executed. Proceed to 7.

In step # 7, the deviation angle histogram created in step # 5 is displaced by a predetermined angle and accumulated to create an accumulated deviation angle histogram, and the process proceeds to step # 8.

In step # 8, the argument that maximizes the frequency of occurrence of the maximum luminance gradient is found from the argument histogram created in step # 5 or the accumulated argument histogram created in step # 7, and step # 9 Proceed to.

In step # 9, the inclination of the recognition object such as an electronic component is obtained from the declination that maximizes the maximum luminance gradient occurrence frequency obtained in step # 8.

2 to 5 show a two-dimensional multi-valued image 10 of a grayscale image of an electronic component or the like stored in the memory in step # 1 of the flowchart of FIG. 1 of this embodiment. These two-dimensional multi-valued images 10 are composed of edges composed of several straight line portions. Then, the luminance of the image changes rapidly at this edge. Further, the inclination of a recognition target such as an electronic component can often be represented by the direction in which the edge of the electronic component extends.

6 and 7 show a plurality of sampling points S set on the two-dimensional multi-valued image in step # 2 of the flowchart of FIG. 1 of the present embodiment. As shown in FIGS. 6 and 7, the plurality of sampling points S are set in a lattice point shape in the whole or a part of the electronic component.

8 to 10 show, in step # 4 of the flow chart of FIG. 1 of the present embodiment, for each sampling point S set in the previous step, in all directions on the two-dimensional multivalued image, A method of obtaining the declination obtained by adding a predetermined angle in the direction in which the brightness gradient is maximum and the magnitude of the brightness gradient will be described.

In FIG. 8, eight points A to H in the x and y directions are provided around each set sampling point S. For example, if the brightness at the point A is expressed as Z (A), The brightness gradient Z _x of the sampling point S in the x direction is Zx = Z (C) + Z (D) + Z (E) −Z (A) −Z
(H) −Z (G) The luminance gradient Z _y of the sampling point S in the y direction is Z _y = Z (G) + Z (F) + Z (E) −Z (A) −Z
(B) -Z (C).

If Zx and Z _{y are} regarded as a two-dimensional vector (Zx, Z _y ), the direction θ that maximizes the brightness gradient is obtained as shown in FIG.
In terms of degrees, the deviation angle = θ + 90 °, and the magnitude of the brightness gradient is given by equation (1).

[0040]

[Equation 1]

The upper part of FIG. 10 shows all sampling points S.
Is the direction θ that maximizes the brightness gradient and the magnitude of the brightness gradient. The direction of the bar of each sampling point S indicates the direction θ in which the brightness gradient of each sampling point S is maximum. The length of the bar at each sampling point S indicates the magnitude of the brightness gradient at each sampling point S. The direction θ where the brightness gradient of the sampling point S near the side of the rectangular two-dimensional multi-valued image 10 is maximum is the direction perpendicular to the side, and near the apex angle of the rectangular two-dimensional multi-valued image 10. The direction θ in which the luminance gradient of the sampling point S at 1 is maximum is inclined in the direction of the bisector of the apex angle.

The lower part of FIG. 10 shows the processing result of the previous process (FIG. 10) in step # 5 of the flowchart of FIG. 1 of this embodiment.
Angle (θ + predetermined angle) obtained by adding a predetermined angle to the direction θ in which the brightness gradient of the sampling point S is maximized
On the other hand, a declination histogram representing the maximum frequency of occurrence of the luminance gradient for each declination, which is created by accumulating the magnitude of the luminance gradient as a frequency weight, is shown.

The method of creating the deviation histogram shown in the lower part of FIG. 10 is as follows.

This deviation angle histogram means that 360 degrees is n.
This is an equal division (n is an arbitrary natural number), and the maximum luminance gradient occurrence frequency of each of the n sections (1 to n) is obtained. Now, let us say that the deviation angle obtained by adding 90 degrees to the direction θ in which the brightness gradient is maximum is m degrees, the frequency at which the brightness gradient in the kth argument section is maximum is H (k), and the brightness gradient at the sampling point S is D.
Then, the processing is H ([mn / 360] +1) ← H ([mn / 360] +
1) + D (however, [] indicates integerization, and the processing in this case indicates that D is added to the ([mn / 360] +1) th declination section).

In this embodiment, since the two-dimensional multi-valued image 10 in the upper part of FIG. 10 is a rectangular parallelepiped, a predetermined angle is set in the direction θ where the brightness gradient of the sampling point S near each side of the rectangular parallelepiped is maximum. The added declination forms a group for each side, and the declination at which the maximum luminance gradient occurrence frequency of the declination histogram for each group is maximum is as shown in the lower part of FIG. It is displaced by a multiple of 90 degrees.

11 and 12 are step # 7 in the flow chart of FIG. 1 of the present embodiment, the deviation angle histogram of step # 5 is displaced by a predetermined angle and accumulated to create an accumulated deviation angle histogram. Here's how to do it. This predetermined angle is
It is the angle difference between the straight line portions of the two-dimensional multi-valued image 10.

In the present embodiment, as shown in the upper part of FIG. 10, in FIG. 11, since the two-dimensional multi-valued image 10 is a rectangular parallelepiped, in the deviation angle histogram of step # 5, 360 degrees is a rectangular apex angle 90. The range divided into four equal to each side corresponds to each side of the rectangular parallelepiped, and the frequency distribution is similar within the range divided into four. Therefore, in FIG. 11, the range of 0 degrees to 90 degrees is left unchanged, the range of 90 degrees to 180 degrees is added with 90 degrees, and the range of 0 degrees to 90 degrees is added, and the range of 180 degrees to 270 degrees is 180 degrees. Add degrees and add to the range of 0 degrees to 90 degrees, 2
In the range of 70 degrees to 360 degrees, 270 degrees is added and 0 degrees to 90 degrees
When added to the range of degrees, a cumulative declination histogram as shown in FIG. 12 is created.

In this case, in this embodiment, 360 degrees is 90 degrees.
However, even if the entire 360-degree range is added 90 times each time and accumulated four times, a cumulative deviation angle histogram similar to that shown in FIG. Needless to say, it can be achieved.

FIG. 13 shows that in steps # 8 and # 9 of the flow chart of FIG. 1 of the present embodiment, the maximum luminance gradient occurrence frequency is found in the deviation angle histogram of step # 5 or the accumulation deviation angle histogram of step # 7. Find the maximum declination,
A method of setting the maximum deflection angle as the inclination of the recognition target will be described.

As shown in FIG. 13, continuous (2j-
1) The barycenter of the section where the sum of the frequencies of the number of sections (j is an arbitrary natural number) is maximized is the declination angle that maximizes the frequency of occurrence of the maximum luminance gradient.

That is, first,

I = -j + 1.

Next, let this t be the central value (2j-1).
When the center of gravity of the frequency of each section is obtained, this center of gravity becomes the inclination of the component.

FIGS. 14 and 15 show that the present embodiment is less affected by noise due to nozzles, dust, brightness changes, and the like.

FIG. 14 is a deviation angle histogram in the case of an ideal two-dimensional multi-valued image that is not affected by nozzles, dust, brightness changes, etc., and its shape is centered on the inclination angle θ of the component. It is a mountain type with small dispersion.

FIG. 15 is a deviation angle histogram of a two-dimensional multi-valued image when it is affected by nozzles, dust, brightness change, etc., and its shape is a mountain shape with a large variance, but it is a two-dimensional multi-valued image. Even if there is an image of the nozzle in the image, dust is attached to the recognition target such as an electronic component, or there is a partial change in brightness on the recognition target such as an electronic component, these are all partial. Since the number of data obtained from these is very small compared to the number of data obtained from the whole, there is almost no effect and the declination angle at which the height of the mountain is maximum hardly changes. , The accurate inclination can be detected stably.

Two examples of combinations of the reflector 2, the illumination 3, the TV camera 5, and the visual recognition device 6 using the method of this embodiment will be described with reference to FIGS. 16 to 19.

16 and 17 show a transmission type case in which the TV camera 5 captures a shadow picture of the electronic component 4 by irradiating the light of the illumination 3 from behind the electronic component 4 sucked by the suction nozzle 1. .. In the case of the transmissive type, the two-dimensional multi-valued image has a shape according to its outer contour, and the inclination of the electronic component is detected by the straight line portion of this outer contour. The visual recognition device 6 is
It has an A / D conversion circuit 9, an image memory 8 and a CPU 7.

18 and 19, the light of the illumination 3 is directly emitted from the lower surface of the electronic component 4 sucked by the suction nozzle 1.
A case of a reflection type in which a reflection image from the electronic component 4 is captured by the TV camera 5 is shown. In the case of the reflection type, the two-dimensional multi-valued image of the electronic component does not form a recognizable image because the brightness is almost the same as the background in portions other than the electrodes of the electronic component, and the two-dimensional multi-valued image of the electronic component is Irrespective of the outer contour, the electrode portion has a large reflection coefficient and the like, and the inclination of the electronic component can be detected by the linear portion of the electrode shape. The visual recognition device 6 includes an A / D conversion circuit 9, an image memory 8,
It has a CPU 7.

According to the inclination detecting method of the present invention, the inclination of the electronic component can be detected by the linear portion as long as there is a linear portion in the two-dimensional multi-valued image of the electronic component, whether the transmission type or the reflection type. In particular, in the case of the reflection type, even if the shape of the electronic component is complicated or there is no linear portion in the outer shape, it can be detected if the linear portion is present in a portion having a large reflection coefficient such as an electrode.

If some straight line portions of the two-dimensional multi-valued image of the electronic component have a known angle difference such as an integral multiple of 90 degrees, another straight line of the two-dimensional multi-valued image of the electronic component is obtained. No matter how complicated the part is, the inclination can be easily detected. Therefore, the inclination can be detected for many types of electronic components by the same method.

The inclination detecting method of this embodiment is not limited to the above-mentioned embodiment, and various modes are possible. For example:

A bias angle histogram representing the maximum brightness gradient occurrence frequency for each argument is created using the magnitude of the brightness gradient at the sampling point S as a frequency weight, and in this deviation angle histogram, the maximum brightness gradient occurrence frequency is the maximum. If the angle can be obtained and the maximum deviation angle can be used as the inclination of the recognition object, the sampling point S setting method, the brightness gradient magnitude and direction determination method, the deviation angle histogram creation method, The calculation method, the deviation angle histogram or the accumulated deviation angle histogram is used to find the deviation angle that maximizes the frequency of occurrence of the maximum brightness gradient, and the calculation method that uses this maximum deviation angle as the inclination of the recognition target can be freely designed. ..

[0064]

According to the inclination detecting method of the present invention, the maximum luminance gradient for each declination is determined by using the magnitude of the luminance gradient at a plurality of sampling points on a two-dimensional multi-valued image of a recognition object such as an electronic component as a frequency weight. A declination histogram that represents the frequency of occurrence is created, and the declination that maximizes the frequency of occurrence of the maximum luminance gradient in this declination histogram or cumulative declination histogram is obtained, and this maximum declination is used as the inclination of the recognition target. Since the tilt of the recognition target such as the electronic component is detected, the influence of noise due to the nozzle, dust, brightness change, etc. is reduced, and the accurate tilt can be detected. It is possible to improve the reliability of the electronic component mounting device used.

Further, the tilt detecting method of the present invention can be applied regardless of whether the electronic component mounting apparatus using the tilt detecting method of the present invention is a transmissive type component recognition or a reflective type component recognition. Since the inclination can be detected by setting the same calculation processing method for many kinds of electronic components, the inclination detection work is simple and easy, and the electronic components using the inclination detection method of the present invention This has the effect of improving the work efficiency of the mounting device.

[Brief description of drawings]

FIG. 1 is a flowchart of a method of an embodiment of a tilt detecting method according to the present invention.

FIG. 2 is an example of the two-dimensional multi-valued image of FIG.

FIG. 3 is an example of the two-dimensional multi-valued image of FIG.

4 is an example of the two-dimensional multi-valued image of FIG.

5 is an example of the two-dimensional multi-valued image of FIG.

6 is an example of a two-dimensional multi-valued image and sampling point distribution chart of FIG. 1. FIG.

7 is an example of the two-dimensional multi-valued image and the sampling point distribution diagram of FIG. 1. FIG.

8 is a calculation diagram of a brightness gradient at the sampling point in FIG.

9 is a calculation diagram of a brightness gradient at the sampling point in FIG.

FIG. 10 is a distribution diagram of the sampling points of FIG. 1 and their luminance gradients, and a deviation angle histogram thereof.

11 is a deflection angle histogram of FIG.

FIG. 12 is a cumulative declination histogram of FIG.

13 is an argument histogram showing the maximum frequency calculation method of FIG. 1. FIG.

FIG. 14 is a noise-free argument histogram of FIG.

15 is a declination histogram with noise of FIG.

16 is a configuration diagram of the reflection type of FIG.

17 is a two-dimensional multi-valued image of FIG.

FIG. 18 is a configuration diagram of the transmission type of FIG. 1.

19 is a two-dimensional multi-valued image of FIG.

FIG. 20 is a configuration diagram of a conventional example method.

FIG. 21 is a configuration diagram of a conventional method.

FIG. 22 is a configuration diagram of a conventional method.

FIG. 23 is a configuration diagram of a conventional method.

FIG. 24 is a configuration diagram of a conventional method.

FIG. 25 is a configuration diagram of a conventional method.

[Explanation of symbols]

 S Sampling point θ Brightness gradient maximum direction 1 Adsorption nozzle 2 Reflector 3 Lighting 4 Electronic component 5 TV camera 6 Visual recognition device 7 CPU 8 Image memory 9 A / D conversion circuit 10 Two-dimensional multi-valued image

Claims (3)

[Claims] 1. A first step of storing a grayscale image of a recognition object as a two-dimensional multivalued image in a memory, a second step of setting a plurality of sampling points on the two-dimensional multivalued image, and a second step of the second step. 2 at each sampling point set in the process
From all the directions on the three-dimensional multi-valued image, the third step of obtaining the magnitude of the brightness gradient and the deviation angle obtained by adding a predetermined angle to the direction in which the brightness gradient is maximum, and from the processing result of the third step ,
A fourth step of creating a declination histogram representing the maximum frequency of occurrence of the luminance gradient for each declination using the magnitude of the luminance gradient as a frequency weight, and the maximum frequency of occurrence of the maximum luminance gradient is the maximum in the declination histogram of the fourth step. And a fifth step in which the maximum deviation angle is determined as the inclination of the recognition target object. 2. A first step of storing a grayscale image of a recognition object as a two-dimensional multivalued image in a memory, a second step of setting a plurality of sampling points on the two-dimensional multivalued image, and a second step of the second step. 2 at each sampling point set in the process
From all the directions on the three-dimensional multi-valued image, the third step of obtaining the magnitude of the brightness gradient and the deviation angle obtained by adding a predetermined angle to the direction in which the brightness gradient is maximum, and from the processing result of the third step ,
A fourth step of creating a deviation angle histogram representing the maximum occurrence frequency of the brightness gradient for each deviation angle using the magnitude of the brightness gradient as a frequency weight, and the deviation histogram of the fourth step is displaced by a predetermined angle and accumulated. The fifth step of calculating the cumulative declination histogram and the declination that maximizes the maximum luminance gradient occurrence frequency in the cumulative declination histogram of the fifth step are obtained, and the maximum declination is recognized. Sixth to be the inclination of the object
An inclination detecting method, comprising: 3. The tilt detecting method according to claim 2, wherein the predetermined angle in the fifth step is an integral multiple of 90 degrees.