JPH0613417A - Semiconductor chip recognition device - Google Patents

Abstract

PURPOSE:To improve the accuracy of a die bonding process, and to enhance the quality of a manufacturing process by computing the inclination of an edge line more precisely even when there are a flaw, etc., at the edge of a semiconductor chip. CONSTITUTION:A first processing means 14, a chip picture signal obtained by image-picking up a semiconductor chip 10 with a TV camera 12 is processed and converted into a digital signal displaying density, a second processing means 14, in which the chip picture signal acquired by the means is processed and coordinate data on a chip picture display screen are detected regarding a plurality of points at edge sections at every side of the chip, and a third processing means 14, in which the correlation coefficients of the coordinate data detected by the second processing means are compared regarding each side of the chip and the degrees of the irregularities of the edge sections are detected and one side, in which the degrees of the irregularities are minimized, is decided, are provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor manufacturing apparatus, and more particularly to a semiconductor chip recognition apparatus used for recognizing the position of a semiconductor chip in a die bonding process.

[0002]

2. Description of the Related Art FIG. 4 is a flow chart showing the flow of processing by a conventional semiconductor chip recognition method. Figure 5
FIG. 5A and FIG. 5B show two examples of the image of the edge of the semiconductor chip. A conventional chip recognition operation will be briefly described with reference to FIGS. 4, 5A, and 5B.

A chip image signal obtained by picking up an image of a semiconductor chip with a television camera under illumination light is binarized to be converted into a digital signal representing light and shade. Then, as a result of the image processing, as shown in FIG. 5A, several points (eg, A,
The coordinate data on the chip image screen is detected for points B and C).

Thereafter, the inclination θ of the edge line of the chip image 50 with respect to the coordinate axis is calculated using the above-mentioned data of several points, and the inclination of the chip is corrected if necessary based on the calculation result, and die bonding is performed. It is carried out.

By the way, when there is a scratch on one edge to be detected by the chip, the reflection of the illumination light becomes weak, and when the chip image signal is binarized, as shown in FIG. 5 (b). , The edge portion 51 of the chip image 50 has irregularities, and the detection points of the edge portion 51 (for example, D, E, F
The data of points) vary.

In such a case, the calculation of the inclination of the edge line (correction angle of the inclination of the chip) θ is hindered, the correction within the allowable accuracy becomes impossible in the subsequent die bonding process, and a defective product occurs. However, there is a problem in that it causes a problem and hinders the quality of the production process.

[0007]

As described above, the conventional method for recognizing a semiconductor chip has a problem that the inclination .theta. Of the edge line is hindered when the edge of the chip has a flaw or the like.

The present invention has been made to solve the above-mentioned problems, and the inclination θ of the edge line can be calculated more accurately even when the edge of the semiconductor chip has a scratch, and the accuracy of the die bonding process can be improved. An object of the present invention is to provide a semiconductor chip recognition device capable of improving the quality of the production process.

[0009]

A semiconductor chip recognizing device of the present invention is a first process for processing a chip image signal obtained by picking up an image of a semiconductor chip by a television camera and converting the chip image signal into a digital signal representing light and shade. Means and second processing means for performing processing on the chip image signal obtained by the first processing means and detecting coordinate data on a chip image display screen for a plurality of points of an edge portion for each side of the chip. A third side for judging the degree of unevenness of the edge portion by comparing the correlation coefficient of the coordinate data detected by the second processing means with respect to each side of the chip and determining the side with the smallest degree of unevenness. And a processing means.

[0010]

Since the degree of unevenness of the edge portion is detected for each side of the chip and the side with the smallest degree of unevenness is determined, the edge of the side with the smallest degree of unevenness is detected even if the edge of the chip has scratches. It becomes possible to calculate the inclination θ of the edge line of the chip with respect to the coordinate axis using the coordinate data detected for the portion. Therefore, even if the edge of the chip has a scratch or the like, the inclination θ of the edge line can be calculated more accurately, and the accuracy of the subsequent die bonding process and the quality of the production process can be improved.

[0011]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram schematically showing a semiconductor chip recognition device according to an embodiment of the present invention.

In FIG. 1, 10 is a semiconductor chip to be recognized in the die bonding process, 11 is a light source for illuminating the chip 10, and 12 is the chip 1.
0 is an industrial television camera (for example, a CCD camera) for picking up images of 0 and its peripheral portion, and 13 is the camera 1
2, a camera control device for controlling 2, an image processing device 14 for processing an image signal obtained by the camera 12 and an image processing device 15 for obtaining an image obtained by the camera 12. It is an image monitor device for monitoring.

The image processing apparatus 14 includes a computer system and has processing means as described below.
That is, a first processing means for processing a chip image signal obtained by picking up an image of a semiconductor chip by a television camera and converting it into a digital signal representing light and shade, and a chip image signal obtained by the first processing means. Is processed by the second processing means for detecting coordinate data on the chip image display screen for a plurality of points of the edge portion for each side of the chip, and the second processing means for each side of the chip. The third processing means for detecting the degree of unevenness of the edge portion by comparing the correlation coefficients of the coordinate data and determining one side with the smallest unevenness, and the chip judged by the third processing means. Using the coordinate data detected by the second processing means for a plurality of points on the edge portion of one side, the inclination θ of the edge line of the one side of the chip with respect to the coordinate axis is calculated. It has and a fourth processing means. FIG. 2 is a flowchart showing the flow of chip recognition processing in the semiconductor chip recognition device of FIG. Figure 3
2 shows an example of an image of the edge of the chip in the process of the chip recognition processing of FIG. Next, the chip recognition operation in the semiconductor chip recognition device of FIG. 1 will be described with reference to FIGS.

In the first step, the chip image signal obtained by picking up the image of the semiconductor chip 10 by the CCD camera 12 under illumination light is binarized, for example, to be converted into a digital signal representing light and shade.

In the second step, processing is performed on the chip image signal, and as shown in FIG. 3, a plurality of points (for example, a1 to a5, b) of the edge portion 31 are provided for each side of the chip image 30.
The coordinate data on the chip image display screen for 1 to b5, c1 to c5, d1 to d5) is detected.

In the third step, for each side of the chip image 30, the correlation coefficient of the coordinate data detected in the second step is obtained to detect the degree of unevenness of the edge portion 31, and the correlation coefficient of each side. And the side having the largest correlation coefficient is determined to be the side having the smallest degree of unevenness. In this case, the correlation coefficient P is calculated by the following equation (1).

[0017]

[Equation 1] It should be noted that the data may be prioritized for each side of the chip image 30 and the degree of unevenness at the edge portion may be detected.

After that, with respect to one side of the chip image 30 determined in the third step, using the coordinate data of a plurality of points of the edge portion 31 detected in the second step, the inclination θ of the edge line with respect to the coordinate axis and Calculate the intercept. In this case, the slope θ (= 1) of the edge line and the intercept b are obtained by the calculation using the least square method as shown in the following equations (2) and (3).

[0019]

[Equation 2]

According to the semiconductor chip recognition apparatus of the above embodiment, the degree of unevenness of the edge portion is detected for each side of the chip and one side having the smallest degree of unevenness is determined, so that the edge of the chip has a scratch or the like. Even in such a case, the inclination θ of the edge line of the chip with respect to the coordinate axis can be calculated using the coordinate data detected for the edge portion of one side having the smallest degree of unevenness.

Therefore, even if the edge of the chip has a scratch or the like, the inclination θ of the edge line can be calculated more accurately,
It is possible to improve the accuracy of the subsequent die bonding process and the quality of the production process.

[0022]

As described above, according to the present invention, the inclination θ of the edge line can be calculated more accurately even when the edge of the semiconductor chip has a scratch, and the accuracy of the die bonding process can be improved and the production process can be improved. It is possible to realize a semiconductor chip recognition device that can improve quality.

[Brief description of drawings]

FIG. 1 is a block diagram schematically showing a semiconductor chip recognition device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a flow of chip recognition processing in the semiconductor chip recognition device of FIG.

FIG. 3 is a diagram showing an example of an image of a chip edge in the process of the chip recognition process of FIG.

FIG. 4 is a flowchart showing a flow of processing by a conventional semiconductor chip recognition method.

FIG. 5 is a diagram showing an image example of an edge portion of a semiconductor chip.

[Explanation of symbols]

10 ... Semiconductor chip, 11 ... Light source, 12 ... Television camera, 13 ... Camera control device, 14 ... Image processing device,
30 ... Chip image, 31 ... Edge part 31.

Claims (2)

[Claims] 1. A first processing means for processing a chip image signal obtained by picking up an image of a semiconductor chip by a television camera and converting the chip image signal into a digital signal representing light and shade, and the first processing means. Second processing means for performing processing on the chip image signal and detecting coordinate data on the chip image display screen for a plurality of points of the edge portion for each side of the chip; and the second processing means for each side of the chip. And a third processing means for detecting the degree of unevenness of the edge portion by comparing the correlation coefficient of the coordinate data detected by, and determining one side having the smallest degree of unevenness. Recognition device. 2. The semiconductor chip recognizing apparatus according to claim 1, further, the coordinates of a plurality of points of an edge portion detected by the second processing means with respect to one side of the chip determined by the third processing means. A semiconductor chip recognizing device, comprising: a fourth processing means for calculating an inclination of an edge line with respect to a coordinate axis using data.