JPH0814849A - Three-dimensional shape detection method for solder - Google Patents

Abstract

PURPOSE:To accurately detect a three-dimensional solder shape from a single sheet of image information by irradiating solder as a measurement object with illumination light from a light source of multi-stage annular type, and photographing the solder with a camera. CONSTITUTION:A light source 2 is provided in a multi-stage, for example, in four stages, so as to have concentric arrangement and give different irradiation angles along the altazimuth line of a virtual hemisphere about solder 6 as a measurement object. The light source 2 is formed to be annular as a whole. The light source 2 is turned on to irradiate the solder 6, thereby forming circular fringe calescence point images on the surface of the solder 6 under reflected light. A radial density measurement scanning line is determined on the image of calescence points photographed with a camera 4. Furthermore, the inclination angle of a reflection point as a measurement point on the surface of the solder 6 is obtained by establishing correspondence to a density distribution value on the scanning line as well as the graph of preliminarily measured inclination angle-reflection intensity. Thereafter, a distance from the center of the solder 6 to the measurement point and the height thereof are found, on the basis of the inclination angle, thereby detecting the surface contour of the solder 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of detecting a three-dimensional shape of a solder after soldering an electronic component by an image processing technique.

[0002]

2. Description of the Related Art When observing the shape of a solder with image processing technology, the conventional method is to switch the illumination under different illumination conditions several times, capture the image of the solder surface each time, and use the bright spots in each image. The shape was approximated from the position. In the above-mentioned conventional method, when capturing an image, 1/30 seconds are required in the NTSC system, and the reliability of measurement is improved as the number of captured images increases, but on the other hand, it takes time and the number of images is small. Only accurate shape can be obtained.

[0003]

SUMMARY OF THE INVENTION Therefore, it is an object of the present invention to accurately detect the three-dimensional shape of solder from one piece of image information.

[0004]

To achieve the above object, the present invention irradiates a solder to be observed with a multi-stage annular light source turned on to form a circular stripe-shaped bright spot image on the surface of the solder by reflected light. Then, on the captured image of this bright spot image, a scanning line for density measurement in the radial direction is determined, and by associating with the density distribution value on this scanning line and the previously measured inclination angle-reflection intensity graph. Obtain the inclination angle of the reflection point as the observation point on the solder surface, and calculate the distance from the center of the solder to the observation point and the height of the observation point based on this inclination angle to detect the three-dimensional shape of the solder surface. There is.

According to this detection method, the three-dimensional shape of the solder can be inferred from one image, so that the blinking control of the light source,
It is possible to reduce the number of times the solder is photographed, the time required for image processing can be shortened, and the detection method can be speeded up.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an image pickup apparatus 1 used in the solder solid shape detecting method of the present invention. This imaging device 1 is for photographing the surface of the observation target solder 6 attached around the light source 2 for illumination, the shutter 3 for diffusing the light from the light source 2 and the lead 5 on the printed circuit board 18. It is assembled by the camera 4 and the like.

The light source 2 is, for example, a highly directional light emitting diode, and is arranged in multiple stages such as concentric circles and different irradiation angles along the longitude and latitude lines of the virtual hemisphere centered on the installation position of the solder 6 to be observed. It is arranged in four stages, and has a ring shape as a whole. Also, a shutter 3 for light diffusion
Are arranged inside these light sources 2, are made of liquid crystal, and are in an electrically controllable state.

A photographing window 7 is located at the apex position of the shutter 3, that is, an extension of the lead 5 in the axial direction, and a camera 4 and a beam splitter 8 are arranged above the photographing window 7. The beam splitter 8 is for illuminating the leads 5 and the solder 6 from directly above, and corresponds to the side liquid crystal type shutter 9 and the light source 10 for epi-illumination.

Next, FIG. 2 shows the camera 4 and the image processing apparatus 1.
1, the connection relationship between the control computer 12 and each part controlled by the control computer 12 is shown. The camera 4 is connected to the internal memory 13 of the image processing apparatus 11, and C
It is designed to work with the PU 14. Also, the CPU 14
Receives an inspection command or the like from the control computer 12, the printer 15, the monitor 16, and the light sources 2 and 10 and the shutters 3 and 9 via the interface 17.
Control. Note that the shutters 3 and 9 are normally solder 6
Is set to convert the light from the light source 2 and the light source 10 into diffused light when inspecting defective products, and becomes transparent when measuring the three-dimensional shape of the solder, and the light with higher directivity reaches the solder 6. Let

As shown in FIG. 3, irradiation light 19 from a certain light source 2 or light source 10 irradiates the surface of the solder 6 at an irradiation angle φ formed by a surface parallel to the upper surface of the printed circuit board 18, and the surface of the solder 6 is exposed. When reflected and heading for the camera 4, there is only one optical path of the reflected light, and only one reflection point P on the solder surface is determined. Therefore, the inclination angle θ of the solder surface at the reflection point P
Is calculated by the equation θ = (90 ° −φ) / 2.

As shown in FIG. 4, the horizontal axis represents the tilt angle at the reflection point P of the solder surface, and the vertical axis represents the light reflection intensity.
The graph of the reflection intensity at the reflection point P appears as a normally distributed curve. This is because the reflection on the solder surface is not a perfect reflection but includes a diffuse reflection component to some extent.

As shown in FIG. 1, the light sources 2 are arranged in multiple stages along the virtual hemispherical surface and are annular illuminations with different irradiation angles φ. When the graphs of (1) and (2) are combined, a graph of reflection intensity I-tilt angle θ when all the light sources 2 and 10 are turned on is shown in FIG. 5 from the relation of I = f (θ). As seen in this graph, the irradiation angles φ of the light source 2 and the light source 10 are 90 °, 70 °, 50 °, 30 °,
Assuming that the angle is 10 °, when all the light sources 2 and 10 are turned on, the graph shows that the inclination angles θ = 0 °, 10 °, 2
It has peaks at 0 °, 30 ° and 40 °.

In the above graph, the minimum points are the midpoints of each peak, that is, 5 °, 15 °, 25 ° and 35 °. The portion between the peak point and the minimum point corresponds to the inclination of the reflection point P on the solder surface and is given by the reflection intensity I. Where θ _n = n
If x5 ° (n = 0 to 8), the point (θ, f (θ)) when θ = θ _n has a maximum value or a minimum value.

When the surface of the solder 6 is imaged by the camera 4,
As shown in FIG. 6, a plurality of circular stripe-shaped bright spot images are obtained as a set of bright spots, that is, a group of reflection spots P, corresponding to the multi-stage arrangement of the annular light sources 2 and the light sources 10. The three-dimensional shape of the solder is detected by measuring the density distribution on the image along the scanning line of the line segment AB in the radial direction passing through the center of the solder 6, that is, the lead 5.

FIG. 7 shows a graph of the distance x-density v along the scanning line (line segment AB). The maximum and minimum points in the figure correspond to the maximum and minimum points in FIG. For,
Each corresponds to the angle shown in the figure. In FIG. 7, the density value of an arbitrary point is v (x), and its coordinates are (x, v
(X)).

The x-coordinates of the maximum point and the minimum point are defined as x ₀ , x ₁ , ..., X _{8 in} order from the right. Point (x _n , v
(X _n )); When n = 0 to 8, the reflection intensity f (θ _n ) at the angle corresponding to the point is known.

Now, at an arbitrary point (x, v (x)), by associating the position on the graph of FIG. 7 with the position on the graph of FIG. 5, the reflection intensity I at that point becomes clear. Become. For the point x _n ≤ x <x _{n + 1} ,
The association is performed by the following primary conversion formula.

{V (x) -v (x _n )} / {v
_{(X n + 1) -v (} x n)} = {f (θ) -f (θ n)}
/ {F (θ _{n + 1} ) −f (θ _n )}

If the position of v (x) (distance x from the apex A) is determined by the above equation, the density v is determined, f (θ) is obtained by the above equation, and I = f (θ) from FIG. Thus, the inclination angle θ is obtained with θ = f ⁻¹ (I). However, when 40 ° <θ, the above formula cannot be used.

When the inclination angle θ is obtained in this manner, the height h at the reflection point P is calculated by the equation h = ΣΔx · tan.
Calculated by θ. However, the calculation is sequentially performed pixel by pixel from point B to point A.

The actual measurement is performed in the following order. It is assumed that the inclination angle-reflection intensity graph and the maximum and minimum points (θ _n , f (θ _n ) on the graph are known in advance.

(1) All the light sources 2 and 10 are turned on. (2) In the fully lit state, the camera 4 images the surface of the solder 6, and the center of the lead 5 is detected by image processing. (3) Scanning line (line segment A for measuring density distribution from captured image)
B) is distributed at equal angles with respect to the circumference, and the search direction is determined. (4) Obtain the distribution of the density values of the reflection points P as the observation points on the scanning line, and extract the maximum points and the minimum points (x _n , v (x _n )). (5) Distribution value v of density at a distance x of an arbitrary reflection point P
I = f (θ) is obtained from (x) and the above equation, and θ = f ⁻¹ (I) is obtained from this. (6) From the obtained tilt angle θ and the distance x, the height h at the observation point (reflection point P) is sequentially obtained from the point B to the point A by the above formula. (7) From the height h, the distance x, and the inclination angle θ, the three-dimensional shape of the solder 6 to be observed is restored on the scanning line by image processing.

[0023]

According to the present invention, since the three-dimensional shape of the solder can be inferred from one image, it is possible to control the blinking of the light source, the number of times the solder is shot, and the time required for image processing can be shortened. It is possible to speed up the detection method.

[Brief description of drawings]

FIG. 1 is a side view of an imaging device.

FIG. 2 is a block diagram of a connection part of a camera, an image processing device, a control computer, and the like.

FIG. 3 is an explanatory diagram showing a relationship between irradiation light and a tilt angle.

FIG. 4 is a graph of tilt angle-reflection intensity.

FIG. 5 is a graph of tilt angle-reflection intensity.

FIG. 6 is an explanatory diagram of a captured image (bright spot image).

FIG. 7 is a distance-density graph.

FIG. 8 is an isometric view of a tilt angle.

[Explanation of symbols]

 1 Imaging Device 2 Light Source 3 Shutter 4 Camera 5 Lead 6 Solder 7 Shooting Window 8 Beam Splitter 9 Shatter 10 Light Source 11 Image Processing Device 12 Control Computer 13 Memory 14 CPU 15 Printer 16 Monitor 17 Interface 18 Printed Circuit Board 19 Irradiation Light P Reflection Point φ Illumination angle θ Inclination angle I Reflection intensity

Claims (1)

[Claims] 1. Multi-stage annular light sources arranged concentrically and having different irradiation angles are all turned on, all the irradiation light from these light sources irradiates the solder to be observed, and the solder surface has circular stripes at reflection points. -Shaped bright spot image is formed, the bright spot image is photographed by a camera and converted into an image signal, and then the bright spot image of the reflection spot is distributed over the circumference on the photographed image to measure the density. A scanning line is determined, and the inclination angle θ of the reflection point as the observation point on the solder surface is obtained by associating the density distribution value on this scanning line with the previously measured inclination angle-reflection intensity graph, and each scanning line The distance x from the solder center to the observation point corresponding to the inclination angle θ obtained by
A three-dimensional shape detecting method for a solder, which obtains a three-dimensional shape of a solder by obtaining an inclination angle θ, a height h and a distance x of an observation point of a bright spot image from an expression h = ΣΔx · tan θ.