JP2003068813A - Method and device for measuring probe mark - Google Patents

Abstract

PROBLEM TO BE SOLVED: To classify the picture of the needle mark of a probe in the conduction inspection of a pad formed in a semiconductor chip and to determine right or wrong of the conduction inspection at high speed. SOLUTION: A CCD camera 10, a lamp 11, a stage 12, a driving part 13 and a control part 14 are installed in a probe mark measuring device 1. The driving part 13 drives the stage 12 where a wafer 91 is fixed based on an instruction from the control part 14 and the wafer 91 is transferred to a prescribed position with respect to the CCD camera 10. The CCD camera 10 photographs the picture, and the lamp 11 generates illumination light in prescribed light quantity at the time of photographing. The CPU board of the control part 14 classifies the photographed pictures and stores them in a storage part 142 based on the position and the depth of the needle mark of the probe. Only the stored picture which is minutely analyzed is selectively extracted and analyzed based on a classification result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for measuring contact marks of a probe formed on an electrode of a semiconductor chip in a continuity test.

[0002]

As is well known, a semiconductor chip is manufactured by forming a predetermined circuit pattern on a wafer. And
In a normal semiconductor chip manufacturing process, a continuity test is performed after the formation of a circuit pattern to check its electrical characteristics. The continuity test is carried out by bringing a needle-shaped test probe into contact with the electrodes of the semiconductor chip. The probe contacts the electrodes with a predetermined resistance value or less by applying a constant pressure to each chip unit or the entire wafer at once.

In order to normally judge the quality of the electrical characteristics of the semiconductor chip, these probes must contact the electrodes in the following good conditions.

(1) Contact within a predetermined area: This is because a pass / fail judgment cannot be made unless the probe needle is in contact with the electrode to be measured.

(2) Needle tip is inserted to a normal depth: Generally, the material of the electrode of the semiconductor device is aluminum, and it is easily oxidized and the surface of the electrode is made of insulating aluminum oxide (Al ₂ O _3). ). In order to perform a proper continuity test, it is necessary for the probe to break through the aluminum oxide film and come into contact with the metal part of the electrode during the test.

(3) Low contact resistance value: Although it depends on the type of probe needle, it is usually desirable to be about 0.1 to several Ω. For example, if the needle tip is crushed or dust is clogged, normal pass / fail judgment cannot be performed. In this case, it can be judged by periodically checking the needle.

Conventionally, in order to judge whether the contact state of the probe with the electrode is good or bad, in order to determine whether or not the state is (1) or (2), the contact mark of the probe formed by the continuity test is checked. Therefore, a visual inspection using a microscope is performed by a skilled person.

Further, as a device for measuring the depth of the contact trace of such a probe, there is a micro-roughness measuring device using a laser microscope or interference. These devices can move the height position of the observation sample (semiconductor chip) and measure the depth of the contact mark of the probe based on the optical information obtained at different height positions. Is.

[0009]

However, when the number of contact traces of the probe formed on the wafer is large, tens of thousands to hundreds of thousands or more on the entire wafer, visual inspection must be performed for all of them. Had the problem that it took a very long time.

In addition, each of the above-mentioned conventional devices has a problem that it takes a long time for the measurement because it is necessary to change the height position of the sample a plurality of times for one measurement of the depth of the contact mark. .

The present invention has been made in view of the above problems, and an object thereof is to make it possible to measure the state of contact marks of a probe easily and in a short time.

[0012]

According to a first aspect of the present invention, there is provided a probe mark measuring method for measuring a contact mark of a probe formed by a continuity test of a semiconductor chip. An image acquisition step of acquiring a captured image, a storage step of classifying and storing the images based on image information about the contact mark of the image acquired by the image acquisition step, and a storage step of the storage step And an analysis step of selectively analyzing the generated image based on the classification result.

According to a second aspect of the present invention, there is provided a probe mark measuring device for measuring a contact mark of a probe formed in a continuity inspection of a semiconductor chip, wherein an image obtained by photographing the contact mark on the semiconductor chip is acquired. Image acquisition means for
A storage unit that classifies and stores the image based on image information about the contact mark of the image acquired by the image acquisition unit.

According to a third aspect of the present invention, in the probe mark measuring device according to the second aspect, the storage unit classifies and stores the images based on the position of the contact mark.

The invention according to claim 4 is the probe mark measuring device according to claim 3, wherein the image is divided into small areas and the contact marks are present in a predetermined small area. Classification is performed based on the position of the contact mark.

According to a fifth aspect of the present invention, in the probe mark measuring device according to any of the second to fourth aspects, the storage unit classifies the images based on the depth of the contact mark. And save.

According to a sixth aspect of the present invention, in the probe mark measuring apparatus according to the fifth aspect, the storage unit sets the depth of the contact mark to the density of pixels forming the contact mark in the image. Classify based on value.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

<1. First Preferred Embodiment> FIG. 1 is a diagram showing the structure of a wafer 91. A plurality of semiconductor chips (more accurately, prototypes of semiconductor chips) 92 are provided on the wafer 91.
Are formed, and pads 93 that will serve as electrodes are formed on the semiconductor chip 92. Whether the electrical characteristics of the semiconductor chip 92 are good or bad is determined by a continuity test, and the continuity test is performed by bringing a probe needle into contact with the pad 93. The material of the pad 93 is generally aluminum, which is easily oxidized to form a film of insulating aluminum oxide (Al ₂ O ₃ ) on the surface. Therefore, in order to perform an appropriate continuity test, it is necessary for the probe to break through the aluminum oxide film and contact the aluminum portion of the pad 93 during the test.

FIG. 2 is a block diagram showing the configuration of the probe mark measuring apparatus 1 according to the embodiment of the present invention. The probe mark measuring apparatus 1 includes a CCD camera 10, a lamp 11 that illuminates when the CCD camera 10 takes an image, and a wafer 91.
It is composed of a stage 12 that fixes and conveys the sheet, a drive unit 13 that drives the stage 12, and a control unit 14 that controls them.

The CCD camera 10 captures a contact trace of the probe on the semiconductor chip as a two-dimensional image by a camera trigger from the control unit 14. Further, the drive unit 13 drives the stage 12 based on an instruction from the control unit 14 so that the wafer 91 is transported to a predetermined position with respect to the imaging range of the CCD camera 10, and also the stage 12
Each time the driving of the
And a control unit 14 to transfer the data. It should be noted that the CCD camera 10 may be moved to take an image of each pad.

The control unit 14 mainly includes a CPU board 140 for executing programs and various arithmetic processes, and the CCD camera 1.
It has an image processing board 141 that performs correction processing and the like on an image captured by 0, and a storage unit 142 that stores programs and images as data.

FIG. 3 is a block diagram showing the configuration of the CPU board 140. The CPU board 140 includes a CPU 143 that performs various arithmetic processes and an RO that stores an operation program.
A RAM 144 for storing M145 and various data is provided, and the ROM 145 and the RAM 144 are the CPU 14
3 is connected.

FIG. 4 is a diagram showing a functional configuration realized by the CPU 143 operating according to a program in the ROM 145, together with other configurations. In the configuration shown in FIG. 4, the trimming unit 200, the position determination unit 201, the depth determination unit 202, and the data storage unit 203 are the CPU 14
The function realized by 3 is shown.

The trimming unit 200 is the image processing board 1
From the shooting area of the first image data 100 transferred from 41, a portion (trimming area) required for determination is cut out,
The second image data 101 is created.

The position determination unit 201 uses the second image data 10
The second image data 1 among the pixels forming the needle trace in FIG.
The coordinates of the four points closest to the outer edge of 01 (details will be described later) are obtained, coordinate data 102, which is data indicating the coordinates of the obtained four points, is created, and the needle mark The position is determined and the classification data 103 is created.

The depth determination unit 202 further cuts out a small area including the needle trace from the second image data 101 based on the coordinate data 102, determines the depth of the needle trace based on the density value of the pixel, and classifies data. Information on the depth of the needle trace is added to 103.

The data storage unit 203 stores the first image data 1
00 or the second image data 101 is associated with the classification data 103 and stored in the storage unit 142.

Note that some of these functions are performed by the CPU 143.
The image processing board 141 may be used instead of the above. For example, the function of the trimming unit 200 is the image processing board 141.
By doing so, the processing can be performed at a higher speed than when the CPU 143 directly processes.

FIG. 5 is a flow chart showing the processing operation in the probe mark measuring apparatus 1. The operation of the probe mark measuring device 1 will be described with reference to FIGS. 4 and 5.

First, based on an instruction from the controller 14, the drive unit 13 causes the stage 12 so that the pad to be photographed is photographed in the center of the photographing range of the CCD camera 10.
Is driven (step S11). After the driving is completed, the driving unit 13 sends a synchronization signal to the lamp 11 and the image processing board 141 of the control unit 14, and the image processing board 141 sends a camera trigger to the CCD camera 10 (step S1).
2).

The CCD camera 10 captures an image based on a camera trigger, and the image processing board 141 captures an image (step S13). At that time, since the lamp 11 illuminates based on the synchronization signal, shooting is performed under an appropriate light amount. Further, the image is subjected to various corrections and processing by the image processing board 141, and the RAM 1 on the CPU board 140 is
It is transferred to 44 as the first image data 100.

When the first image data 100 is created, the trimming section 200 trims the first image data 100 to create the second image data 101 (step S14). FIG. 6 is a diagram showing the relationship between the photographing area R of the CCD camera 10 and the trimming area (area of the second image data 101) S cut out by the trimming section 200.

The needle of the probe contacts the pad 93 with the center position P as the target. Therefore, in order to check the state of the needle mark 94, it is not necessary to process the entire imaging region R, and it is sufficient to process only the trimming region S in which the needle mark 94 is considered to be present. Is.
For example, if the size of the image captured by the CCD camera 10 is SVGA (760 × 680, 8 bits for each RGB), the size of the trimming area S is 200 ×.
If the value is 200, the size of the image will change from 1.5 MB to 120
Since it can be reduced to about 1/10 of KB, and the amount of data and the amount of processing can be reduced, classification processing can be performed at high speed. Hereinafter, the trimming area S
The size will be described as 200 × 200.

Further, by performing such trimming, there is no influence of a portion which is considered to be irrelevant in examining the state of the needle mark 94 (a portion other than the pad of the semiconductor chip, etc.), so that classification is performed. The processing can be performed accurately.

As shown in FIG. 6, the CCD camera 1
In the case of 0, since the photographing is performed so that the center position P of the pad 93 is the center of the photographing region R, the position and the size of the pad 93 on the photographing region R are substantially constant. Therefore, the pad 9
The process of cutting out the trimming region S, which is a region slightly smaller than 3, from the photographing region R can be performed at high speed without performing complicated image processing such as edge detection.

When the trimming unit 200 creates the second image data 101, the probe mark measuring apparatus 1 performs an image classification process on the acquired image (step S1).
5).

Before describing the detailed operation of the image classification process, a method of classifying the needle marks in the probe mark measuring apparatus 1 according to the state of the image will be described.

FIG. 7 is a diagram showing the relationship of four points which serve as a reference for position determination in the trimming area S. Points Z1 to Z4 shown in FIG. 7 are points closest to the edge of the trimming area S (hereinafter, referred to as “trimming frame”) among the points on the needle trace 94. That is, if the point on the trimming area S is represented by the X and Y coordinates shown in FIG. 7 and the coordinate of the lower left pixel is set to (0, 0), the point where Y becomes the maximum value among the points on the needle mark 94. Is Z1 (X1, Y1), the point where X is the maximum value is Z2 (X2, Y2), and the point where X is the minimum value is Z3.
(X3, Y3), the point where Y becomes the minimum value is Z4 (X4, Y3)
4).

FIG. 8 is a diagram conceptually showing a method of obtaining the points Z1 to Z4. As shown in FIG. 8, scanning is performed from the upper left point (0,200) in the direction of the arrow, and the point having a density value of a predetermined value or more is regarded as a pixel forming the needle mark 94. Then, by comparing the size of the coordinates, the point Z
1 to the point Z4 can be determined.

FIG. 9 is a diagram showing an example in which the trimming area S is divided into small areas and classified according to their positions. For each small area, the closest small area is classified A (corresponding to the hatched areas: A1 to A20 in FIG. 9) according to the distance from the trimming frame, and the small area next to the trimming frame is classified B (B1). Through B12), and the farthest small area is classified as C (C1 through C4 is applicable).

Then, based on the coordinates of the points Z1 to Z4, it is determined in which small area each point exists.
If it exists in a small area classified as classification A, the image of the needle mark is classified as “attention”, and if it is on the trimming frame, the needle of the probe sticks out of the pad 93 and contacts in the continuity test. Therefore, it is classified as “abnormal” because it may have occurred.

In this embodiment, the trimming area S
Although the example in which the small area in the above is classified into three areas A, B, and C is shown, it may be further classified. Also, the area may be divided into smaller areas closer to the trimming frame. The above is the method of classifying the images of the needle marks based on the positions of the needle marks.

Next, a classification method based on the depth of the needle marks will be described. When the needle mark is photographed, the density value of the pixel of the needle mark increases with the depth. Further, by photographing the needle marks under a predetermined light amount, it is possible to determine the depth based on the same reference for the needle marks photographed separately. That is,
If the density values of the pixels forming the needle marks in the image are obtained and the images are classified based on the density values, the images can be classified based on the depth of the needle marks.

FIGS. 10 (a) to 10 (c) are views showing an example of a quadrangle in which the needle marks are circumscribed. Quadrangle 30 shown
1 to 303 are respective needle marks 94a to 94c
Can be obtained from the points Z1 to Z4 at.
For example, explaining the example of FIG.
Since 4a is included in the region of the quadrangle 301, the quadrangle 30
Pixels included in the area 1 having a density value of a predetermined value or more are regarded as pixels forming the needle trace 94a, and the average value of the density values of those pixels is calculated.

As described above, by processing only the portion included in the quadrilateral area circumscribing the needle mark, the amount of data to be processed can be reduced as compared with the case where the entire trimming area S is processed. Processing can be performed at high speed.

Table 1 shows an example of classification by the average value of the density values thus obtained. In the example shown in Table 1, the images are classified into classification I or classification VI based on the average value of the density values. Here, those classified as “class I” are classified as “caution” because there is a possibility that the needle mark is too deep and breaks the bottom of the pad. Also, those classified as "Class V" may be classified as "Caution" because the needle marks may be too shallow to break the aluminum oxide film. These are classified into the same "caution", but different processing is performed later because of different reasons. Although the average value of the density values in Table 1 is shown by subtracting the predetermined value, the average value of the density values may be used for classification. The above is the method of classifying the images of the needle marks based on the depth of the needle marks.

Returning to FIG. 5, the specific operation of the image classification process (step S15) in the probe mark measuring apparatus 1 will be described. 11 to 13 are flowcharts showing details of the image classification process.

First, the position determining unit 201 determines the maximum value Y1 of Y1.
0, the minimum Y value Y4 is 200, the maximum X value X2 is 0,
The minimum value X3 of X is 200, the flag F indicating whether or not the density value of the previous pixel is equal to or more than a predetermined value is 0, and Y is 200.
Are set respectively (step S101), and X is set to 0 (step S102). The flag F
Is 1 indicates that the density value of the immediately previous pixel is equal to or greater than the predetermined value.

Next, it is judged whether or not the density value of the pixel at the coordinates (X, Y) is not less than a predetermined value (step S103),
If the density value is greater than or equal to the predetermined value, the flag D is set to 1 (step S104), and if the density value is not greater than or equal to the predetermined value, the flag D is set to 0 (step S105). Accordingly, the flag D indicates whether or not the density value of the pixel at the coordinates (X, Y) is not less than the predetermined value.

Since the points Z1 to Z4 exist on the edge of the needle mark (so-called edge portion), when the conditions of the flag F being 0 and the flag D being 1 or the flags F being 1 and the flag D being 0 are satisfied. The point (X, Y) corresponds to the edge portion.
Therefore, the points (X,
It is determined whether Y) is a point on the edge (step S10).
6), if it is not a point on the edge, the coordinates are not compared,
The flag D is set to the flag F for the next determination (step S107), and the process is skipped until step S205.

When X is 0 or 200 and flag D is 1 (corresponding to the case where the point (X, Y) is on the trimming frame and there is a needle mark), regardless of the value of flag F. It is determined to be an edge. In addition, X is 200 and flag D
When is 1, the point (X, Y) is regarded as the point Z2, and X2 becomes X.
And Y to Y2.

When it is determined in step S106 that the point (X, Y) is a point on the edge, the flag D is set to the flag F for the next determination (step S10).
8) Further, it is determined whether the flag D is 0 (step S
109), if the flag D is 0, the point (X, Y) is not the pixel forming the needle mark, but the point (X-1, Y) is the edge portion of the needle mark and is a candidate for the point Z2. If X-1 is larger than X2, X-1 is set to X2 and Y is set to Y2 (steps S112 and S113). afterwards,
The process is skipped until step S205.

On the other hand, when the flag D is 1 (step S1
No. in 09. ) Is point Z1 at point (X, Y), point Z3
Or it may be point Z4. Therefore, when X is smaller than X3, point (X, Y) is a candidate for point Z3, so X is set to X3 and Y is set to Y3 (steps S110 and S111). If Y is smaller than Y4, the point (X, Y) is a candidate for the point Z4.
To X and Y4 to Y (step S2
01, S202). Further, when Y is larger than Y1, the point (X, Y) is a candidate for the point Z1, so X is set to X1 and Y is set to Y1 (step S203,
S204).

From the point (0,200) to the point (2
(00, 0) (steps S205 to S2)
07), the coordinates of the points Z1 to Z4 can be obtained.

When the coordinates of the points Z1 to Z4 are obtained,
The position determination unit 201 determines the coordinate data 102 based on each coordinate.
And the points Z1 to Z4 are shown in FIG.
It is determined which of the classification A to the classification C shown in 0 corresponds, and the classification data 103 is created based on the result (step S209).

As a result, the trimming area S can be divided into small areas, and classification can be performed based on the position of the contact mark depending on whether or not the needle mark is present in a predetermined small area. It is possible to quickly and automatically determine whether or not a person is in contact. Further, the image can be classified and stored based on the position of the needle mark, and the image can be selectively analyzed later based on the classification result.

Next, the depth determination unit 202 causes the coordinate data 10
Based on 2, the quadrilateral region in which the needle mark is circumscribing is obtained (step S301), and pixels having a predetermined density value or more are extracted in the obtained quadrilateral region (step S302).

The average value of the density values of the extracted pixels is obtained (step S303), the classification is performed according to the average value as shown in Table 1 (step S304), and the classification information based on the depth of the needle trace is stored in the classification data 103. Add.

[0060]

[Table 1]

Thus, the depth of the needle trace can be classified based on the density value of the pixel forming the needle trace in the image, and it can be classified whether or not the probe needle is in contact with the normal depth. It can be done automatically. Further, the image can be classified and stored based on the depth of the needle mark, and the image can be selectively analyzed later based on the classification result.

Further, the data storage unit 203 adds the classification data 103 to the first image data 100 and saves the needle marks which are “abnormal” and “caution” in the classification by position, and are classified as “normal”. Regarding the needle mark, the classification data 103 is added to the second image data 101 and stored (step S305). This is because the images classified as “abnormal” and “caution” in the classification by position may have the needle marks protruding from the trimming area S. Therefore, when the detailed determination is performed thereafter, the data of the imaging area R ( This is because the first image data 100) is necessary, whereas the image classified as “normal” has a needle mark near the center of the trimming area S, and the trimming is performed when the detailed determination is performed thereafter. This is because the data of the area S (second image data 101) is sufficient.

The data included in the classification data 103 includes the serial numbers of needle marks, semiconductor chip numbers, pad numbers, needle edge classifications (A3, B12, B3, C4, etc.) and density classifications (class I, Class II, etc., and image data (first image data 100 or second image data 10)
1).

As a result, the images can be classified and stored based on the image information about the needle marks among the images acquired by photographing, and the classified and stored images can be selected later based on the classification result. Can be analyzed.
In addition, the amount of data to be stored can be reduced as compared with the case where the first image data 100 is stored for all needle marks.

Returning to FIG. 5, the image classification process (step S
15) is completed, all pads 9 on the wafer 91 are
The processing of steps S11 to S15 is repeated until the photographing of step 3 is completed (step S16).

Next, an image is extracted based on the classification state (step S17). Table 2 shows an example of performing extraction according to the result of classification in the image classification process. Those whose comprehensive determination column is classified as "detailed determination" are selectively extracted in step S17.

[0067]

[Table 2]

That is, when the result of the continuity inspection is “good (the one in the continuity inspection column in Table 2 is“) ”, the image of the needle mark classified as“ class I ”in the classification by depth ( In the example of Table 2, “3 of 2” and “4 of 2” are applicable), and images of needle marks classified as “abnormal” in the classification by position (“5” is applicable in the example of Table 2) are selective. To be extracted.

Even if the contact is made in a shallow depth (“classification VI” or “classification V” in classification by depth), if the result of the continuity test is good, it happens that the aluminum oxide is It can be considered as a good product simply because the film is thin, but when the probe contact position is out of the pad (“abnormal” in classification by position) or when it is too deep and breaks the bottom of the pad (depending on depth). This is because, in the classification, “classification I”) may have a problem even if the inspection of the electrical characteristics is good, and detailed analysis is required.

On the other hand, in the case where the result of the continuity inspection is "No (the continuity inspection column is X in Table 2)",
Only images classified as “classification VI” in classification by depth (“12” corresponds in the example of Table 2) are selectively extracted.

In the classification by depth, this is "classification V
If it is classified as `` I '', it may be judged as "No" in the continuity test because the needle did not contact due to some trouble of the probe, or it may be judged as "No" in the continuity test even though the needle slightly touched. This is because it is necessary to analyze in detail the details.

When the image classified as requiring the detailed determination is extracted, the extracted image is analyzed (step S18). As a method of analyzing the selectively extracted image in detail, the shape, area, maximum density value, integrated density value, etc. of the needle marks in the image are obtained by image processing and comprehensively judged.

As a result, the images classified and stored by the high-speed processing can be selectively analyzed only for those that need to be analyzed in detail based on the classification result, which is faster than the analysis of all contact marks. The contact state of the probe needle can be analyzed.

Thereafter, steps S17 and S18 are repeated until the image corresponding to the extraction condition is completed (step S19), the analysis result is saved (step S20), and the process is terminated.

As described above, the probe mark measuring apparatus 1 can classify and store images at high speed based on the image information of the image of the captured needle mark. Then, by selectively analyzing the stored images based on the classification result,
It is possible to efficiently determine the quality of the contact state of the needle in the continuity test with the probe. In particular, by classifying and storing images based on the position and depth of needle marks,
Pass / fail judgment can be performed easily and at high speed.

As a method of detailed analysis, visual inspection by a person may be performed. In that case, the probe mark measuring apparatus 1 selectively extracts the image of the needle mark based on the classification result and displays it on a display or the like, and a person looks at the image and judges, and inputs the analysis result. May be.

<2. Modifications> The embodiments of the present invention have been described above, but the present invention is not limited to the above embodiments, and various modifications can be made.

For example, in the above embodiment, the point Z1
Or, the average of density values is obtained after obtaining the quadrangle with which the needle mark circumscribes based on the coordinates of the points Z4. However, when obtaining the coordinates of the points Z1 to Z4, the point (X, Y) is set to When it is determined that the density is equal to or higher than the density, the density values may be integrated at the same time to obtain the average value.

In the above embodiment, the coordinates of the points Z1 to Z4 are calculated in order to classify the images of the needle marks according to the positions. However, the coordinates of the points Z1 to Z4 are not calculated. Alternatively, only the maximum value X2 and the minimum value X3 of the X coordinate of the needle mark and the maximum value Y1 and the minimum value Y4 of the Y coordinate may be respectively obtained. FIG. 14 is a diagram showing an example of classification of small areas in that case. In this case, X2, X3, Y
If either 1 or Y4 exists on the classification A, the image of the needle mark is classified as "caution", and if it exists on the trimming frame, it is classified as "abnormal".

The classification result does not necessarily have to be stored on the probe mark measuring device 1. For example, the data may be recorded in a recording medium such as a flexible disk, or the data indicating the classification result may be transmitted to another terminal for detailed analysis via a network. That is, any method may be used as long as the classification result can be used later.

[0081]

According to the invention described in claim 1, the images are classified and stored based on the image information about the contact mark area in the acquired image, and the stored images are classified based on the classification result. The selective analysis can be performed at a higher speed than the total contact trace of the probe needle in the continuity test.

According to the second aspect of the present invention, the classified and stored images can be selectively analyzed later by classifying and storing the images based on the image information of the acquired images.

According to the third aspect of the present invention, the images are classified and stored based on the positions of the contact marks, so that those in which the probe needle is not in contact with the normal position are selectively analyzed later. be able to.

According to the fourth aspect of the invention, whether the needle of the probe is in contact with the normal position is determined by performing classification based on the position of the contact mark depending on whether the contact mark exists in a predetermined small area. It is possible to quickly classify whether or not.

According to the fifth aspect of the invention, the images are classified and stored based on the depth of the contact mark, so that the needles of the probe that are not in contact with each other at a sufficient depth can be selectively selected later. Can be analyzed.

According to the sixth aspect of the present invention, the depth of the contact mark is classified based on the density value of the pixel forming the contact mark in the image, so that whether the probe needle contacts at the normal depth. It is possible to quickly classify whether or not.

[Brief description of drawings]

FIG. 1 is a diagram showing a structure of a wafer.

FIG. 2 is a block diagram showing a configuration of a probe mark measuring device according to an embodiment of the present invention.

FIG. 3 is a block diagram showing a configuration of a CPU board.

FIG. 4 is a diagram showing a functional configuration realized by a CPU operating according to a program in a ROM, together with other configurations.

FIG. 5 is a flowchart showing a processing operation in the probe mark measuring device.

FIG. 6 is a diagram showing a relationship between a photographing area of a CCD camera and a trimming area.

FIG. 7 is a diagram showing a relationship of four points serving as a reference for position determination in a trimming area S.

FIG. 8 is a diagram conceptually showing a method of obtaining points Z1 to Z4.

FIG. 9 is a diagram showing an example in which a trimming area S is divided into small areas and classified according to their positions.

FIG. 10 is a diagram showing an example of a quadrangle in which needle marks are circumscribed.

FIG. 11 is a flowchart showing details of image classification processing in the probe mark measuring apparatus.

FIG. 12 is a flowchart showing details of image classification processing in the probe mark measuring apparatus.

FIG. 13 is a flowchart showing details of image classification processing in the probe mark measuring apparatus.

FIG. 14 is a diagram showing an example of classification according to the positions of needle marks.

[Explanation of symbols]

1 Probe mark measuring device 10 CCD camera 11 lamps 12 stages 13 Drive 14 Control unit 140 CPU board 141 image processing board 142 storage 200 trimming section 201 Position determination unit 202 determination unit 203 data storage 301, 302, 303 quadrilateral 91 wafers 92 Semiconductor chips 93 pad 94, 94a, 94b, 94c Needle marks

Claims (6)

[Claims] 1. A probe mark measuring method for measuring a contact mark of a probe formed by a continuity test of a semiconductor chip, comprising an image acquiring step of acquiring an image of the contact mark on the semiconductor chip. A saving step of classifying and saving the image based on image information about the contact mark of the image acquired by the image acquiring step, and selectively analyzing the image saved by the saving step based on a classification result. A method for measuring a probe mark, which comprises: 2. A probe mark measuring device for measuring a contact mark of a probe formed by a continuity test of a semiconductor chip, comprising: an image acquiring unit for acquiring an image of the contact mark on the semiconductor chip. A probe mark measuring device comprising: a storage unit that classifies and stores the image based on image information about the contact trace of the image acquired by the image acquisition unit. 3. The probe mark measuring apparatus according to claim 2, wherein the storage unit classifies and stores the image based on a position of the contact mark. 4. The probe mark measuring device according to claim 3, wherein the image is divided into small regions, and the position of the contact mark is based on whether the contact mark exists in a predetermined small region. A probe mark measuring device characterized by performing classification. 5. The probe mark measuring device according to claim 2, wherein the saving unit classifies and saves the image based on a depth of the contact mark. Probe mark measuring device. 6. The probe mark measuring device according to claim 5, wherein the storage unit classifies the depth of the contact mark based on a density value of a pixel forming the contact mark in the image. A probe mark measuring device characterized by.