JP2600320B2 - Parts inspection method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a component inspection method for determining that a component or the like to be inspected exists at a specific position in an inspection process during a manufacturing process.

2. Description of the Related Art Conventionally, as a method of this kind, a lighting device 3 is installed diagonally above an inspection target component 2 on an inspection substrate 1 in FIG. A histogram area for obtaining a luminance histogram is set in a range in which the component 4 can be formed, and a lighting device 5 is installed at a position opposed to the lighting device 3 with the component 2 interposed therebetween. Make shadow 6 of part 2
When a histogram area for obtaining a luminance histogram is set in a range in which a histogram can be obtained, the image is captured by an image recognition device 8 through a TV camera 7 installed right above the component 2, and when these two histogram areas are illuminated by the respective illuminations described above. There is a component inspection method for inspecting the existence of an inspection target component based on a difference between two types of obtained histograms.

However, in the above method, however, the histogram is determined according to the color of the inspection board, its surface and state, the shape (size, height) of each component, the position where the component is placed, and the like. It was difficult to set parameters because of the large difference, and setting time was required. Therefore, since the parameters are set depending on the experience and feeling, the set values vary from person to person, and a stable inspection cannot be performed.

Means for Solving the Problems Therefore, in the component inspection method of the present invention, before the component inspection, each component of the plurality of substrates to be inspected is sampled by a lighting device installed diagonally above the component to be inspected. A first step of obtaining a luminance histogram in a range where shadows can be formed and a luminance histogram in a range not forming shadows;
A second step of statistically processing each luminance histogram of each component of each board obtained in the first step to obtain a feature amount, and a statistical processing of the feature amount of each component of each board obtained in the second step for each component The third value is a set value for determining whether or not the part has a shadow.
This is a component inspection method that pre-processes up to the process.

The present invention compares a set value obtained by statistically processing two types of luminance histograms obtained in advance for each component of a plurality of sample substrates with a value obtained by statistically processing two types of luminance histograms for each component of an inspection substrate. When there is variation in the reflection state due to the material behind the inspection component, when the shadow of the inspection component becomes thin due to reflected light from other inspection components near the inspection component, or when the background image and shadow are made of a material whose background is difficult to reflect In the case where it is difficult to find a difference, the influence of the variation of each part is also taken into consideration, and it is easy to find the setting for judging the presence / absence of a part, and since this set value has numerical support. A highly reliable component presence inspection has been made possible.

Embodiment Hereinafter, an embodiment of the present invention will be described. 1 and 2 are flowcharts of a component inspection method according to an embodiment of the present invention,
FIG. 3A is a view showing histogram areas 9 and 10 when the shadow 6 of the component 2 is formed by the lighting device 5, and FIG. 3B is a graph A of a luminance histogram of the histogram area 9 in FIG. 3A.
₁₂ and the feature quantity values alpha _A12, Figure 3 c is a third diagram a histogram domain plot B ₁₂ of the luminance histogram of the 10 and the feature quantity values alpha _B12, the Figure 4 d shadow part 2 by the illumination device 3 4
Shows a histogram region 9, 10 when the could, FIG. 3 e is a graph C ₁₂ and the feature quantity values alpha _C12 of the luminance histogram of the histogram region 9 of Figure 3 d, the third diagram f Figure 3 graph D ₁₂ and the feature quantity values alpha _D12 of the luminance histogram of d histogram region 10, FIG. 4 a set value beta _A1 and graph alpha _A1 feature of Figure 3 b, the fourth Figure b Figure 3 graph of the characteristic amounts of c alpha _B1 and settings beta _B1, Figure 4 c is a set value beta _C1 and feature quantity graph alpha _C1 in Figure 3 e, Figure 4 d is graph of the characteristic of Figure 3 f FIG. 7 is a diagram of α _D1 and a set value β _D1 .

The component inspection apparatus of this embodiment inspects the presence or absence of the component 2 mounted on the circuit board 1 of FIG. 5, and uses the component inspection method of the present invention as it is. TV to capture component 2 to be inspected, lighting devices 3 and 5, and brightness histogram
The positional relationship with the camera 7 is as shown in FIG. 5, and the lighting devices 3 and 5 for inspecting other parts and the TV camera 7 are connected to the circuit board 1.
The robot translates above it.

The operation of the component inspection apparatus according to the embodiment will be described with reference to FIGS. In step 1 (preparation for inspection), the illuminating device 3 emits light to each of the components 2 of the plurality of sample substrates to generate shadows 4 of the components 2 and histogram regions 9 and 10 (FIG. 3).
, The luminance histograms A ₁₂ and B ₁₂ are acquired. Similarly, for other parts (n = 1, 2,... L, n: part number, l: number of parts), luminance histograms A _n2 , B _n2 , (n = 1, 2,
... L (n: part number, l: number of parts), and the luminance histogram A
_nk , B _nk (n = 1, 2,... l, k = 1, 2,... q, where n: part number, l: number of parts, k: board number, q: number of boards). In step 2, the illumination 5 is emitted in exactly the same manner as in step 1,
Each board (k = 1,2, ... q, where k: board number, q: number of boards)
, Where n = 1, 2,... L, where n is the part number and l is the number of parts, and the shadow areas 9 and 10 (the third
In the figure, the luminance histograms C _nk and D _nk are taken. In step 3, each brightness histogram A _nk , B _nk , C _nk , D _nk for each component of each board obtained in steps 1 and 2 (eg, FIG. 3 and A ₁₂ ,
B ₁₂ , C ₁₂ , D ₁₂ ) are subjected to statistical processing (Table 1) to obtain feature quantities α _Ank ,
α _Bnk , α _Cnk , α _Dnk (eg, α _A12 , α _B12 , α _C12 ,
α _D12 ).

In step 4, the feature amounts α _Ank , α _Bnk , α _Cnk , and α _Dnk of each component obtained in step 3 are statistically processed for each component (Table 1) to determine the presence or absence of a shadow of the component. Set values β _An , β _Bn , β _Cn , β _Dn (eg, β _A1 , β _B1 , β _C1 , β _C in FIG. 4)
_D1 ).

In step 5, in the actual inspection, the illumination device 3 emits light for each component (n = 1, 2,... L) of the inspection board to generate a shadow 4 of the component. ₁ , G ₁ is taken. Similarly, luminance histograms E _n and F _n are fetched for other components. In step 6, the illuminating device 5 emits light in the actual inspection in exactly the same manner as in step 5, causing shadows 6 of the components on the inspection board.
In histogram areas 9 and 10, luminance histograms G _n and H _n
Take in. In Step 7, the luminance histograms obtained in Steps 5 and 6 are statistically processed (Table 1), and the feature amounts γ _En , γ _Fn ,
γ _Gn and γ _Hn are obtained. In step 8, each luminance histogram E _n , F _n ,
The feature amount γ _En , obtained by statistically processing G _n and H _n (Table 1)
γ _Fn , γ _Gn , γ _Hn and the set values β _An , β _Bn ,
β _Cn and β _Dn are compared, and the presence or absence of a shadow is determined in step 9. (Eg β _Cn -β _An γ _Gn -γ _En and β _Bn -β _Dn γ
_{When Fn} − _γHn is established, it is determined that a shadow is formed, and when either one is not established, it is determined that a shadow is not formed. In step 10, the presence or absence of a component is determined based on the presence or absence of a shadow obtained in step 9.

In addition, it is more effective to mount two illuminating devices at a position perpendicular to the line connecting the illuminating devices 8 and 9 to illuminate the components obliquely from above and to inspect the components by four illuminations using the component inspection device (FIG. 5). is there.

As described above, according to the present invention, a set value obtained by statistically processing two types of luminance histograms obtained in advance for each component of a plurality of sample substrates and two types of luminance histograms for each component of an inspection substrate are obtained. To compare with the statistically processed values,
The influence of the variation for each part will also be considered,
In addition, it is easy to obtain a set value for judging the presence / absence of a component, and since the set value has numerical support, a highly reliable component presence / absence inspection can be performed.

[Brief description of the drawings]

1 and 2 are flowcharts of a component inspection method according to an embodiment of the present invention. FIG. 3a is a diagram showing a histogram area when a shadow of a component is formed by a lighting device. FIG. FIG. 3C is a graph showing the relationship between the graph of the luminance histogram and the value of the characteristic amount, FIG. 3C is a graph showing the relationship between the graph of the luminance histogram in the histogram area of FIG. 3A and the value of the characteristic amount, and FIG. FIG. 3E is a diagram showing a relationship between a graph of a luminance histogram of the histogram region of FIG. 3D and a value of a feature amount, and FIG. 3F is a diagram of a histogram region of FIG. 3D. FIG. 4A is a graph showing the relationship between the graph of the luminance histogram and the value of the characteristic amount, FIG. 4A is a graph showing the relationship between the graph of the characteristic amount of FIG. 3B and the set value, and FIG. Fig. 4c shows the relationship with the set values, FIG. 4d is a graph showing the relationship between the graph of the characteristic amount and the set value, FIG. 4d is a graph showing the relationship between the graph of the characteristic amount of FIG. 3f and the set value, and FIGS. 5a and 5b are side and plan views of the component inspection apparatus. It is. 1 ... circuit board, 2 ... parts, 3,4 ... lighting equipment, 5,6 ...
... Shadow, 7 ... TV camera, 8 ... Image recognition device, 9,10 ...
Histogram area.

 ──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Toshio Morimoto 1006 Kazuma Kadoma, Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (56) References JP-A-64-86090 (JP, A) JP-A-63- 106508 (JP, A) JP-A-62-175606 (JP, A)

Claims (1)

(57) [Claims] 1. A luminance in a range where a shadow of a component can be formed by a lighting device installed obliquely above a component to be inspected for each component of a plurality of inspection boards as a standard inspection object for setting a set value serving as an inspection standard. A first step of obtaining a histogram and a luminance histogram in a range not forming a shadow; a second step of performing a statistical process on each luminance histogram of each component of each board obtained in the first step to obtain a feature amount; A third step in which a value obtained by statistically processing the feature amount of each component of each board for each component is set as a set value for determining the presence or absence of a shadow of the component;
A fourth step of obtaining a luminance histogram in a range where shadows of each component of the inspection board can be formed and a luminance histogram in a range where no shadow is formed; and a fourth step of statistically processing each luminance histogram of each component obtained in the fourth step to obtain a feature amount. A fifth step, a sixth step of comparing the feature value of the actual inspection part obtained in the fifth step with the set value of each part obtained in the third step for each part to determine the presence or absence of a shadow; A component inspection method including a seventh step of determining the presence or absence of a component.