JPH0552763A - Inspecting apparatus of appearance of semiconductor device - Google Patents

Abstract

PURPOSE:To obtain an appearance inspecting apparatus for a semiconductor device to control the quality with the improved inspecting accuracy and inspecting speed. CONSTITUTION:This apparatus is provided with an illuminating means for illuminating the surface of a semiconductor device, a photoelectric converting device 1a which photographs the surface of the semiconductor device illuminated by the illuminating means and converts and outputs a video signal, and a lightness converting means 1b for converting the video signal output from the photoelectric converting device 1a to digital signal of a predetermined gradation with a suitable sampling frequency. Moreover, this apparatus is provided with a binarizing means 1c for binarizing the digital signal output from the lightness converting means 1b based on the binarization threshold value, a binarized data memory means 1c for storing the binarized data of the binarizing means 1c, and a CPU 1e which determines the binarization threshold value based on the digital signal from the photoelectric converting device 1a of the binarized data and evaluates whether or not a defect of a predetermined size or larger size is present.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a visual inspection apparatus for semiconductor devices such as integrated circuits, and more particularly to a visual inspection apparatus for inspecting external defects such as voids and chips on the package surface.

[0002]

2. Description of the Related Art The appearance of defects such as voids and chips on the package surface of a semiconductor device impairs quality.
The voids sometimes affected the operation of the semiconductor device. Therefore, conventionally, the package surface has been visually inspected for defects.

[0003]

The visual inspection has a problem in quality control because there are individual differences in inspection accuracy and inspection speed. The present invention has been made in view of such circumstances,
A configuration in which a video signal obtained by picking up an image of a semiconductor device is binarized by a threshold value set based on the video signal, and whether or not there is a defect is automatically determined using the binarized data. By doing so, it is an object of the present invention to provide a semiconductor device appearance inspection device capable of improving inspection accuracy and inspection speed.

[0004]

SUMMARY OF THE INVENTION A semiconductor device appearance inspection apparatus according to the present invention is an apparatus for inspecting a package appearance defect of a semiconductor device, and an illumination unit for illuminating the surface of the package, and an illumination unit for illuminating the package surface. A photoelectric conversion device that images the surface of the package, converts it into a video signal, and outputs the image signal, and a brightness that converts the video signal output by the photoelectric conversion device into a digital signal with a predetermined gradation at an appropriate sampling frequency. Conversion means, and means for determining a binarization threshold value based on the digital signal output by the brightness conversion means,
The binarizing means for binarizing the digital signal output by the lightness converting means based on the threshold value, and the binarizing means
Binarized data storage means for storing binarized data, and 2
It is provided with means for judging the presence or absence of data corresponding to the defect in the binarized data, and means for judging the size of the defect related to the data when the data exists.

[0005]

When the package surface of the semiconductor device is illuminated and imaged, the defect appears darker than the other parts. In the present invention, the binarization threshold value is determined from the video signal obtained by imaging the surface of the semiconductor device. This binarization threshold value is determined so that the defect is "0" and the other parts are "1". If there is "0" in the binarized data, the size of the defect is checked using the binarized data, and if one defect is a predetermined size or more, it is determined to be a defective product. Therefore, unlike the visual inspection, the inspection accuracy and the inspection speed can be kept constant.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be specifically described below with reference to the drawings showing the embodiments thereof. FIG. 1 is a schematic structural diagram of a semiconductor device appearance inspection apparatus according to the present invention. Reference numeral 2 in the figure denotes a rectangular parallelepiped semiconductor device, which are arranged side by side. Illuminating means 1g is arranged diagonally above the semiconductor device 2, and a photoelectric conversion device 1a such as a television camera is arranged immediately above. In this device, the illumination means 1g illuminates the semiconductor device 2, the photoelectric conversion device 1a takes an image of the surface of the illuminated semiconductor device 2, and the video signal of the image is output to the defect detection device 1f. . When illuminated by the illumination means 1g,
The defects on the surface of the semiconductor device 2 appear darker than the other parts.

FIG. 2 is a block diagram of the defect detecting device 1f of the present invention. The photoelectric conversion device 1a is a brightness conversion means 1b and a binarization means 1c in a defect detection device 1f for a captured image signal of one screen.
Output to. The lightness conversion means 1b converts the input video signal into a digital signal of 256 gradations at a sampling frequency of the same frequency as the pixel-corresponding clock of the photoelectric conversion device 1a, and converts the converted digital signal to the CPU 1e and binarization means. Output to 1c. The CPU 1e takes in the digital signal of 256 gradations converted by the brightness conversion means 1b, totals the number of 256 brightness data, and creates and stores a brightness histogram.

This brightness histogram is 0 to 25 on the horizontal axis.
The lightness is taken from 5 (dark to light), and the vertical axis shows the number of data as frequency. The CPU 1e determines a binarization threshold value from this brightness histogram and outputs this to the binarization means 1c. In creating the histogram, the CPU 1e removes the data at the background portion, the lead portion, and the edge of the package in the captured image based on the information of the inspection target given in advance.

FIG. 3 is a flowchart for determining the binarization threshold value. A brightness histogram is created as described above (step S1), and the binarization threshold value is set to the darkest 0 (step S2). Next, read the number of data of the binarization threshold value from the brightness histogram (step S3)
.. Then, it is determined whether or not the number of data of the binarization threshold value is α or more (step S4).

Α is for each type of semiconductor device to be inspected
It is preset in the CPU 1e. This α is determined as follows. That is, a brightness histogram of a reference sample having a defect is created under the same optical conditions as in the actual inspection, and the frequency of the brightness which is slightly higher than the brightness of the defect in this case is defined as α. In other words, the lightness which is slightly brighter than the defect is used as a threshold to prevent defect detection error.

If the data number of the binarization threshold value is less than α, the process proceeds to step S5. If the data number of the binarization threshold value is α or more, the process proceeds to step S7. When the data number of the binarization threshold is less than α, 1 is added to the binarization threshold (step S5). Then the binarization threshold <25
It is determined whether it is 5 (clearest) (step S6). If the binarization threshold value <255, the process returns to step S3. If the binarization threshold = 255, the process ends. If the number of data of the binarization threshold value is equal to or larger than α in step S4, the binarization threshold value is set as the binarization threshold value for defect detection.
(Step S7).

The binarizing means 1c binarizes the digital signal output from the lightness converting means 1b based on the binarizing threshold value. By this binarization, the defective portion is represented by "0", and the other portions are represented by "1". The CPU 1e writes the binarized data in the binarized data storage means 1d using a video RAM or the like in association with the picked-up image, and at that time, based on the information of the inspection object given in advance, the ejector pin The data of the mark notch part (this becomes "0" like a defect) is set to "1".

The CPU 1e then checks the size of the defect based on this stored data. This is performed by the horizontal projection image and the vertical projection image of the storage contents of the binary data storage means 1d. Horizontal (vertical) projection is performed by accumulating the number of "0" s of stored data for each line in the horizontal (vertical) direction of the captured image. As shown in FIG. 4, when the horizontal and vertical directions are continuous for a predetermined number of lines or more (when the width is equal to or more than a predetermined value), it can be determined that there is a defect of a size to be defective, but as shown in FIG. If there are thin defects in the vertical and horizontal directions that do not need to be determined as defective, it is erroneously determined to be defective.

Therefore, in the present invention, first, a wide defect is detected by vertical projection, a horizontal projection is taken within the width as shown in FIG. 6, and the width of the defect in this projection (vertical width) is examined. It is checked whether this is a predetermined value or more. Then, the product of the horizontal and vertical detection widths is used to calculate the area, and the value is used to judge the quality. As a result, it is possible to prevent the presence of a thin defect as in the case of FIG. 5 from being determined as defective. Of course, vertical and horizontal projection may be performed in reverse order.

FIG. 7 is a flowchart showing the processing procedure of the above principle. This is not judged at once for the entire area of the captured image, but it is divided into a plurality of areas to perform the quality judgment. Therefore, in step S1, the area is first set. Next, a vertical projection is created (step S2), the size of the horizontal width is checked (step S3), and if there is more than a predetermined value, a vertical projection is created for that part (step S4).
, Check the vertical width (step S5). If there is a predetermined value or more, the area is obtained by multiplying the horizontal and vertical widths (step S6). If the area is more than the predetermined value, it is determined as a defective product (step S7). If the detected value or calculated value is smaller than the predetermined value in steps S3, S5, S6, move to the next area (steps S8, S6).
9) After the determination of all areas is completed, if all have small values, it is determined as a good product (step S10).

In the above embodiment, the brightness conversion means
1b describes the case where the video signal is converted into a digital signal having 256 gradations, but the present invention is not limited to this, and it may be converted into a digital signal having an appropriate gradation number.

[0017]

As described above, in the device of the present invention, the image signal obtained by imaging the package surface of the semiconductor device is binarized by the threshold value determined based on the image signal, and this binarized data is converted into binary data. Since it is automatically determined whether or not there is a defect by using it, it is possible to improve inspection accuracy and inspection speed and perform quality control without individual differences.

[Brief description of drawings]

FIG. 1 is a schematic structural diagram of an appearance inspection device of the present invention.

FIG. 2 is a block diagram of a defect detection device.

FIG. 3 is a flowchart for determining a binarization level.

FIG. 4 is an explanatory diagram of determination of defect size.

FIG. 5 is an explanatory diagram of determination of defect size.

FIG. 6 is an explanatory diagram of determination of defect size.

FIG. 7 is a flowchart of quality determination.

[Explanation of symbols]

 1a Photoelectric conversion device 1b Brightness conversion means 1c Binarization means 1d Binarized data storage means 1e CPU 1f Defect detection device 1g Illumination means 2 Semiconductor device

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 7/18 B 8626-5C

Claims (1)

[Claims] 1. An apparatus for inspecting a package appearance defect of a semiconductor device, wherein an illuminating means for illuminating the surface of the package and an image of the surface of the package illuminated by the illuminating means are imaged, converted into a video signal and output. A photoelectric conversion device, a brightness conversion unit that converts a video signal output from the photoelectric conversion device into a digital signal having a brightness of a predetermined gradation at an appropriate sampling frequency, and a binary value based on the digital signal output by the brightness conversion unit. A unit for determining a binarization threshold value, a binarization unit for binarizing the digital signal output from the lightness conversion unit based on the threshold value, and storing data binarized by the binarization unit. A binarized data storage unit, a unit for determining the presence / absence of one of the binarized data corresponding to the defect, and a unit for determining the size of the defect associated with the data when the data exists. An appearance inspection apparatus for a semiconductor device, comprising: a step.