JP3365824B2 - Solder height detection method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting the height of solder at a lead position by applying image processing technology. 2. Description of the Related Art As shown in FIG. 1, when the height of a solder fillet 2 is determined at the position of a lead 1 by applying an image processing technique, illuminations L1 to L4 having different irradiation angles of at most about four stages.
L4 is sequentially turned on, its surface is imaged by the camera 3, and the data is input to the image processing device 4. In addition, in image processing, the gradient of the point that illuminates best at each pixel on the solder fillet 2 is determined. This is, for example, in the case of a concave fillet as shown in FIG. 1, the point that reflects the light Ln (n = 1 to 4) best is x
_{If n} and the gradient are g _n , points x ₁ , x ₂ , x ₃ , and x ₄ are formed from the outside of the land 5 to the inside. In practice, the illumination number that shines best even near the point _xn is n. Therefore, as shown in FIG. 2, the area illuminated by the illumination L1, the area illuminated by the illumination L2, the area illuminated by the illumination L3, and the area illuminated by the illumination L4 are arranged on the solder fillet 2 from the outside to the inside of the land 5 as shown in FIG. Next, by integrating the gradient vectors of the respective pixels from the outside to the inside of the land 5, the shape of the solder fillet 2 is restored as shown in FIG. This is called a gradient vector integration method. Incidentally, the gradient g _n is obtained as FIG. Disadvantages of the Prior Art Although the surface shape of the solder fillet 2 is a curve, it can be expressed by a piecewise straight line. Therefore, when the number of illuminations is small, the angular resolution is low, and the accumulated error of each pixel height increases. [0007] The illumination angle φ _n of the illumination is at most 0 <φ _n
<(Π / 2), | g _n | <1. Therefore, the shape of the fillet surface having a gradient of 1 or more cannot be restored. Therefore, the height of the solder fillet 2 in contact with the lead 1 is not known. Accordingly, an object of the present invention is to make it possible to measure the solder height with high accuracy regardless of the shape of the fillet surface. [0009] A spline curve is well known as a smooth curve. Now, consider restoring the fillet surface with a quadratic spline curve as shown in FIG. Note that the fillet surface is represented by a point (x, y). Interval [x _{n + 1 ,} x
_n ] is interpolated by the quadratic curve of the equation (1),
Since the derivative of the quadratic curve equation is equal to the gradient g, the derivative y 'of the quadratic curve in the section is expressed by equation (2). The interval [x _{n +} 1, x _n] endpoint of _{(x n, y n),} (x n + 1,
y _{n + 1} ), the differential coefficients y ' _n, y' _{n + 1,} that is, the gradients g _n , g
_{n + 1} is obtained by the equations (3) and (4). [0010] ## EQU2 ## ## EQU3 ## [0013] (3) (4) A coefficient a _{n + 1} is obtained from both equations. [0015] Then, from equation (2), equation (6) becomes
Equations (7) and (8) are derived from the equations, and Equation (9) is derived from the equations (7) and (8). [Equation 6] [Equation 7] [Equation 8] [Equation 9] Here, if y ₀ = 0 and y _{n + 1} are sequentially calculated from n = 0 to m, the differential coefficient y ′ becomes x _n (n
= 0,1,2, ..., because the continuous m), the connection point of the quadratic curve (x _n, connection y _n) in the interval curve becomes smooth. The curve in the section [x _{n + 1} , x _n ] is given by (9)
It can be calculated from the equation (10). [Equation 10] [0024] In the interval [0, x _m], y '= 2a _m x +
Since it is b _m , the equation of the curve is (11). [Equation 11] If x = 0, y = h (solder height), and the following equation (12) is obtained. [Equation 12] The gradient at the contact point can be obtained from y '= bm by the following equation (13). (Equation 13) The points x _n (n = 0, 1, 1) obtained by the prior art
2,..., M: m = the number of illumination steps −1), the cross section of the solder fillet 2 is restored by a quadratic curve.
In addition, the fillet section is assumed to be smoothly continuous. Incidentally, as shown in FIG. 6, it is introduced just above illumination to determine the Kurobe end point x ₀ of the solder fillet 2. Step 1 In a certain measurement line passing through the center of the lead 1 in the image of the illumination L0 (this is an angle almost right above), the end point of the lead 1 is set to the origin 0, and the black end point of the solder fillet 2 is set. determine the x _0. Since the solder portion appears black, x ₀ is the end point of the solder fillet 2. From the image of the illumination Ln from n = 1 to m, the midpoint _xn of the white part (reflecting part that shines strongly) is determined. Step 2 The gradient g _n from n = 0 to m is calculated from equation (14) (φ _n is the illumination angle of the illumination n). (Equation 14) _Let y ₀ = 0 (y _n corresponds to the height of the solder fillet 2 at the x _n coordinate value). From n = 0 to m-1, Equation (15) is calculated from Equation (5). (Equation 15) Equation (16) is calculated from equation (6). [Mathematical formula-see original document] The equation (17) is calculated from the equation (9). [Mathematical formula-see original document] Step 3 From the formulas (12) and (13), the height h and the contact angle θ of the solder fillet 2 at the position of the lead 1 are obtained from the formulas (18) and (19). (Equation 18) (Equation 19) In the present invention , each of the solder fillet surfaces
The section of the point that reflects the strongest for each lighting is
Interpolate with quadratic spline curve with derivative equal to gradient
To from the surface of the solder fillet smooth 2 Tsugisu
It can be restored with a pline curve . For this reason, the height detection accuracy of each pixel is high, and the height can be calculated by curve interpolation even in the portion of the solder fillet surface where the gradient exceeds 1.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an explanatory diagram of illumination of a solder fillet and an imaging state. FIG. 2 is an explanatory diagram of a region that shines on the surface of a fillet. FIG. 3 is an explanatory diagram of a relationship between a shining point on a fillet surface and a gradient. FIG. 4 is an explanatory diagram for obtaining a gradient from an illumination angle. FIG. 5 is an explanatory diagram of a relationship between a solder fillet surface and a curve. FIG. 6 is an explanatory diagram of a planar image of a solder fillet. [Description of Signs] 1 lead 2 solder fillet 3 camera 4 image processing device 5 land

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-218407 (JP, A) JP-A-5-60531 (JP, A) JP-A-3-280446 (JP, A) JP-A-4- 296980 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01B 11/24 B23K 1/00 G01N 21/88

Claims (1)

(57) [Claims] 1. The surface of a solder fillet is sequentially illuminated by illuminations having different illumination angles, and the position and gradient of the point on the solder fillet surface that is most strongly reflected for each illumination are obtained. , The interval between these points, the derivative at these points equal to the gradient
Quadratic spline interpolating a curve, the detection method of the solder height of the contact point between the quadratic spline curves and lead and obtains the height of the solder fillet at the read position with.