KR100282133B1 - A system to inspect VTR head shapes by using image process techniques and the inspection method using the same - Google Patents

Abstract

The present invention relates to a shape inspection apparatus for a video head using an image processing technique and an inspection method using the same, in particular, automatically obtains a gap line of a video head in which foreign matter exists, and automatically calculates four measurement items that classify geometric shapes. An image processing system that can measure and an algorithm thereof.

A video has two or more heads mounted on a rotating drum to record or reproduce a video signal. However, in order to obtain consistent quality control in mass production of video, the geometrical and electrical characteristics of the head mounted on the drum must be the same.

Accordingly, the present invention provides a method for inspecting a shape of a video head, which can automatically measure four inspection items that roughly distinguish a geometric shape of a video head using an image processing algorithm using an image processing technique.

Description

A system to inspect VTR head shapes by using image process techniques and the inspection method using the same

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for inspecting a shape of a video head using an image processing technique and an inspection method using the same, and in particular, automatically obtains a gap line of a video head in which foreign substances exist, and roughly identifies geometric shapes. The present invention relates to an image processing system and algorithms capable of automatically measuring four measurement items.

Magnetic recording and reproducing techniques are used for storing and reproducing massive video signal data. Among these, the emergence of V.T.R (Video Tape Recoder) (hereinafter referred to as VTR) still plays a large role in transmitting and storing a large amount of information. Since the quality of the VTR is evaluated by the storage of the video signal and the image quality of the reproduced video, the quality of the head recorded and reproduced on the magnetic tape is an important component that determines the quality of the VTR.

Therefore, with reference to the drawings will be described in detail the characteristics of the video head.

1 is a block diagram of a head mounted on a video drum. As shown in the figure, two or more heads are mounted on a rotating drum to record or reproduce a video signal.

FIG. 2 is a magnetic recording pattern diagram of the magnetic recording tape by the head of FIG. 1, and shows a magnetic recording form in which the tape is recorded on the magnetic recording tape by the head during constant speed running on a drum rotating at a constant speed. In order to realize a constant recording interval during constant drum rotation and constant speed of magnetic tape, the height of the head mounted on the drum must be constant.

3 is a front view of the head of FIG. 1, in which a semiconductor chip made of two ferrites and glass is mounted on a brass base. The boundary line between the two ferrites is a gap line where the image signal is substantially recorded on the magnetic tape, and the magnetic recording interval is determined by the gap line positions between the heads fixed to the drum. Therefore, in order to achieve consistent quality control in mass production of videos, the same geometrical characteristics as above on the head mounted on the drum should be the same. The geometrical characteristics of the head are measured by shape inspection, as shown in FIG.

FIG. 4 is a front view illustrating the shape inspection measurement item for the head of FIG. 1. FIG. In order to record at regular intervals on a constant-speed magnetic tape against a video drum rotating at constant speed, four items of the head must be measured in the field, as shown in the drawing. The first is CW, which represents the chip width, the second is the distance H1 from the top interface of the chip to the gap line, the third is the thickness T _W of the gap line, and the fourth is determined from the pre-calibration specimen before measurement. This is the distance H ₀ from the hypothetical base line to the gap line. Of these four items, H ₀ is the most important measurement item. This is because the chip is not uniformly manufactured at a constant height due to a process error during the assembly of the head.

That is, Figure 5 is a cross-sectional view for explaining the assembly error for the head of Figure 1, as shown in the figure, to bond the chip to the base of the brass to fix the head, the adhesive state of the adhesive and the base surface Depending on the degree of processing, the chips may be glued in an inclined state rather than being horizontally bonded to the baseline. If a large number of such heads are assembled in the drum, the magnetic recording interval by each head will be irregular, and high quality image quality will not be obtained when the video signal is reproduced. Thus, the value obtained by measuring the position of the gap line for which virtual absolute reference position of the H _0. Therefore, H ₀ is the most important measurement item among the four measurement items described above, and in the process of assembling the head on the drum, if the same H ₀ group is classified and the heads of the same group are assembled, the head of the same height can be manufactured, and so on. Magnetic recording of intervals can be obtained.

For this reason, the shape measurement of the VTR head has a great influence on the shape of the magnetic recording, and a measurement automation system is required to maintain the consistency of the measurement when producing a large VTR head. However, such heads cause irregular noise images due to cracks, cracks, and adhesion of foreign substances in the air generated during machining, such as cutting and polishing operations, and thus, it is difficult to measure gap lines. Therefore, in order to measure the gap line effectively, it is required to develop a high speed image processing algorithm in which the geometry of the head, the ferrite and glass, which are the constituent materials of the head, and the boundary between the two ferrites are considered considering the gap line and the reflection intensity of the background.

The present invention is to solve the above problems, it is an object to easily check the items that can distinguish the geometric characteristics by automatically measuring the gap line of the head.

A shape inspection apparatus for a video head using an image processing technique for achieving the above object includes a head fixing jig for measuring the shape of a video head at a predetermined position, and an external lighting device for illuminating the head for image acquisition of the head. And an optical system including a high magnification lens and a camera mounted at a predetermined distance from the jig, and connected to the camera to perform image processing. It comprises a computer consisting of an image processing board equipped with a chip, and an analog monitor connected to the image processing board and outputting a shape measurement result of the head.

In addition, the shape inspection method of the video head using the image processing method, in order to measure the geometric shape of the video head, using the image represented by the reflection intensity of the head to extract the gap line between the two ferrites of the chip constituting the head, A first step of estimating an area where a gap line is expected to remove noise due to foreign matters on the surface, a second step of obtaining an outline of a gap line by using a contour extraction operator in the estimated area; A third step of calculating an approximation straight line passing through the gap line from the contour data of the gap line by using a least square method, and then repeatedly removing the contour data spaced a predetermined distance from the immediately before the approximation to remove noise; In the image with the reflection intensity of the noise-free area using a straight line, two ferrites And a fourth step of extracting a contacting gap line, and measuring each item representing a geometric shape of a video head using the extracted gap line.

1 is a block diagram of a head mounted on a video drum.

2 is a magnetic recording pattern diagram of a magnetic recording tape by the head of FIG.

3 is a head front view of FIG.

4 is a front view for explaining the shape inspection measurement item for the head of FIG.

5 is a cross-sectional view illustrating the assembly error with respect to the head of FIG.

6 is a configuration diagram of a shape inspection apparatus for a video head using an image processing technique according to the present invention.

FIG. 7 is a configuration diagram of an external lighting device illustrated to explain a lighting device for image acquisition of a head in the apparatus of FIG. 6.

8 is a cross-sectional view for explaining the various shapes of the video head.

9 is an image and histogram diagram of a video head as seen through the apparatus of FIG.

FIG. 10 is an image diagram of a noisy video head shown through the apparatus of FIG. 6. FIG.

11 is an image diagram of a video head shown for explaining a gap line position estimation method.

12 is an image diagram of a video head shown for explaining a method for obtaining a contour of a gap line.

FIG. 13 is an image diagram of a video head shown for explaining a method of calculating a linear equation passing through a gap line. FIG.

14 is an image diagram of a video head shown for explaining a method of obtaining a gap line.

15A to 15C are image views of the heads sequentially shown to explain the shape inspection method of the video head using the image processing technique according to the present invention.

FIG. 16 is an image diagram of a head shown to explain a test result of a head in which noise exists. FIG.

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

6 is a configuration diagram of a shape inspection apparatus for a video head using an image processing technique according to the present invention. This device is a high magnification lens equipped with a black and white camera with a field of view of 170 μm × 200 μm, which is 20 mm away from the head fixing jig for measuring the VTR head at a fixed position. It consists of a configured optical system. The monochrome camera is connected to a computer comprised of an image processing board equipped with a DSP chip for image processing. Measurement results are also output to an analog monitor connected to the image processing board for easy viewing by the operator.

FIG. 7 is a block diagram of an external lighting apparatus illustrated to describe a lighting apparatus for image acquisition of a head in the apparatus of FIG. 6. The external lighting is divided into two types, reflective and transmissive, and the brightness of the image can be adjusted by the lighting angle and the brightness of the lighting itself. FIG. 7 is a vertical illumination method, which is the basis of the fallout illumination method, and vertical illumination of a general metal microscope. It is a device used as a device. This device is used for specular metal specimens and plastic specimens with good reflectivity. Therefore, in order to measure the shape of the head made of ferrite and glass components having high surface roughness and superior reflection characteristics due to the polishing process, the system as shown in FIG. 6 is configured by using the vertical lighting device as shown in FIG. 7 as an external lighting device. do. On the other hand, the camera used was a black and white camera, and in consideration of the chromatic aberration effect of the lens, red L.E.D. (LED) of monochromatic light was used as a light source of the illumination device.

The shape of the video head has various shapes depending on the device to be used, the recording method, and the manufacturing process method of the company. 8 is a cross-sectional view illustrating various shapes of a video head, in which a gap line close to the vertical direction of the center of the head has a common feature of having a narrow area where two ferrites meet. That is, the gap line means an interface where two ferrites come into contact with each other, and the interface between two ferrites may be completely coincident with each other, or may be adhered to each other depending on the degree of processing. In this case, it is not treated as a defect unconditionally, and it is determined by the four measurement items mentioned above.

9 is an image and histogram diagram of a video head seen through the apparatus of FIG. 6. As shown in the figure, the head is largely divided into regions having four kinds of reflection intensities, which are reflected in the order of background, glass gap line and ferrite. The thickness of the gap line is represented by a thickness distribution of 3 pixels by 7 pixels. However, in the shape measurement of the head, noise is generated on the head surface due to dust floating around and various foreign matters generated during processing.

FIG. 10 is an image diagram of a noisy video head shown through the apparatus of FIG. 6. In order to perform robust measurement even in such a noise state, an image processing algorithm that is insensitive to noise and can measure at high speed is required.

Therefore, we present a five-step image processing algorithm for extracting gap lines.

The first step is the estimation of the gap line position. After estimating where the gap line is expected to be, finding the gap line using a more complex image processing algorithm is a way to increase the measurement success rate and the image processing speed. Since the gap line is in the concave position of the ferrite, since the general shape of the head is used, the intensity profile of the X and Y axes is used to locate the position insensitive to noise. That is, when the top of the left side of the image as the origin, assuming that the i-th position, Y-axis in the X-axis of the pixel intensity of the j-th position P (i, j), the profile of the i-th position in the X axis P _X ( i) is as shown in Equation 1 below.

[Equation 1]

The gap line estimated position on the Y axis is taken as the center of gravity of the Y profile, and the gap line estimated position on the X axis is determined as the center of gravity of the distribution after subtracting each profile value from the maximum value of the X axis profile. Using the X- and Y-axis profiles, the position of the gap line can be estimated insensitive to noise.

FIG. 11 is an image diagram of a video head for explaining a method of estimating a gap line position. After estimating the position of a gap line in the same manner as above, the estimation region is inserted into the gap line region to obtain a more detailed position of the gap line. Decide to come. Here, the estimation region was selected to be up to three times larger than the gap line thickness T _W of the head to be measured. In addition, the thickness CW of the chip and the top starting position H-init of the head can be obtained from the distribution of the Y-axis profile.

The second step is to find the contour of the gap line in the estimation region of the first step, using a contour extraction operator. FIG. 12 is an image diagram of a video head shown for explaining a method for obtaining a contour line of a gap line, and represents features of the contour in a general image of the head. That is, A and B shown in the drawing indicate the boundary between the ferrite and the base and the boundary between the ferrite and the glass, and C represents the boundary surface where the two ferrites are in contact in the X-axis direction. In this case, an operator such as Equation 2 is used to obtain the C region from the A and B regions.

[Equation 2]

E (i, j) = 3 × (-P (i-1, j) + P (i + 1, j))

-| -P (i-10, j) + P (i + 10, j) ｜

-| -P (i, j-2) + P (i, j + 2) ｜

= 3 × X-Y-Z

X is a contour extraction (sobel) operator in the X direction, and values corresponding to the contours of B and C are calculated. Y represents the position and density difference of 10 pixels away from the pixels i and j in the X-axis direction. The region B has an arbitrary value, but is negligibly small because the region C is the same in the region where the gap line exists. Z is a contour value in the Y-axis direction and has a value in the A and B areas, but has no value in the C area. Therefore, E (i, j) in Equation 2 above for the A and B regions has a small value because Y and Z have arbitrary values, and a large value in the C region. As a result of the calculation, the contour for the gap line can be obtained.

Since the third step may result in different contours of the gap line in the second step due to noise on the head surface, approximation across the gap line using the least square fit to reduce the effect of this noise. Finding a straight line. In other words, it is repeatedly performed while erasing the contour data away from the approximate straight line, so that the distance between the approximate straight line and the contour data reaches a certain tolerance. FIG. 13 is an image diagram of a video head for explaining a method of calculating a linear equation passing through a gap line, and illustrates an approximation linear calculation process passing through a gap line.

A fourth step is to obtain a gap line, and FIG. 14 is an image diagram of a video head shown to explain a method of obtaining a gap line. As shown in FIG. 14, when two ferrites are displaced and contacted in a head, a gap line is defined as a part to which a ferrite is commonly bonded. Even in this case, contour data is used on the neck of the ferrite to find the gap line. That is, the ferrite on the left side shown in the drawing is called ferrite A, the ferrite on the right is called ferrite B, the point where the upper contour of the ferrite A meets the gap line P _AU , the point where the lower contour of the ferrite A meets the gap line P Similarly for _AD , the point where the upper contour of the ferrite B meets the gap line is P _BU , and the point where the lower contour of the ferrite B meets the gap line is referred to as P _BD . The gap line is from P _AU to P _BD .

Finally, the fifth step is to calculate the shape measurement item of the head. The remaining three items, H ₁ , T _W , and H ₀ , can be obtained from the CW and H-init obtained in the first step and the position of the gap line obtained in the fourth step. However, the most important H ₀ value can be determined by knowing the position of the reference line, which uses the reference specimen head that has already been precisely measured to determine the gap line, and then establish the baseline position from the measured H ₀ value.

15A to 15C are image diagrams of the heads sequentially shown to explain the shape inspection method of the video head using the image processing technique according to the present invention, and show an image processing process by the five-step algorithm described above. . That is, Figure 15a is a process of estimating the position of the gap line through the shape image of the video head, Figure 15b is a process of estimating the position of the gap line, after obtaining the contour of the gap line based on this to obtain the equation of a straight line passing through the gap line It is an image diagram of the process. In the process of FIG. 15C, the gap line is obtained, and the shape measurement item of the head is calculated and output.

FIG. 16 is an image diagram of a head illustrated to explain a test result of a head in which a noise exists, and shows measurement results of various types of heads in which a noise exists. As can be seen from these results, it can be seen that fine gap lines are successfully measured insensitive to noise.

The shape measurement method of the video head according to the present invention using the five-step algorithm as described above can obtain a high reliability compared to the conventional manual measurement results. As a result of inspecting 1,000 video heads by the method according to the present invention, the measured items inspected in 980 heads showed high precision within ± 2 μm. Ten of the remaining 20 showed ± 3 μm, five had ± 4 μm, three had ± 5 μm, and two showed measurement failure. The failed heads were the noise on the face of the head located in the gap line and the failure of cracks in the ferrite. Therefore, according to the method of the present invention, it can be seen that the measurement accuracy of ± 5 μm when no defect such as a large noise or a crack is present in the shape of the head.

As described above, according to the present invention, in the inspection of the geometric shape of the video head, the gap line of the video head in which foreign matters and the like exist can be automatically obtained by using a high speed image processing algorithm, and thus the geometric shape of the video head is roughly determined. Four measurement items can be easily measured.

In the present invention, the position of the gap line is estimated in advance in the image of the video head indicated by the reflection intensity, and the position of the gap line is held in a straight line by calculating the profile for each pixel by introducing the contour extraction operator and the least square method. An algorithm that can remove the noise present in the system is presented. Therefore, it is possible to solve the difficulty caused by noise that has been a problem in the conventional video head shape inspection. In addition, since the gap line of the video head can be automatically obtained by the above algorithm, the shape measurement items of the head can be easily checked automatically. Therefore, it is easy to classify the head having the same characteristics, there is an excellent effect to improve the productivity and the quality of the VTR in the field to produce a large amount of VTR.

Claims (8)

A head fixing jig for measuring a shape of a video head at a predetermined position, an external lighting device for illuminating the head for acquiring an image of the head, an optical system composed of a high magnification lens and a camera mounted at a predetermined distance from the jig, Connected to the camera, using the image output from the optical system to extract the gap line between the two ferrites of the chip constituting the head, it is expected that there is a gap line in order to remove noise due to foreign matters on the head surface Estimate an area to be used, obtain an outline of a gap line using an outline extraction operator in the estimated area, calculate an approximated straight line passing through the gap line using a least square method from the outline data of the gap line, and then apply this approximated straight line Repeatedly erases the contour data Removes the sound, extracts the gap line between two ferrites from the image represented by the reflection intensity of the noise-free area using the approximate straight line, and uses the extracted gap line to represent the geometric shape of the video head. D.P. to measure them. And a computer comprising a chip-mounted image processing board, and an analog monitor connected to the image processing board and outputting a shape measurement result of the head. The apparatus of claim 1, wherein the external lighting device is illuminated by a vertical illumination method using a light source of monochromatic light. The camera of claim 1, wherein the camera is F.O.V. A shape inspection apparatus for a video head using an image processing technique, characterized by a monochrome camera having a thickness of 170 μm × 200 μm. In order to measure the geometric shape of the video head, the gap line between two ferrites of the chip constituting the head is extracted using the image represented by the reflection intensity of the head, and the gap line is used to remove noise caused by foreign matters on the head surface. A first step of estimating an area expected to be present, a second step of obtaining an outline of a gap line using an outline extraction operator in the estimated area, and a gap line using a least square method from the outline data of the gap line After calculating the approximate straight line passing through, the third step of eliminating the noise by repeatedly eliminating the contour data a certain distance from the approximated straight line, and represented by the reflection intensity of the noise removed region using the approximated straight line Extracting a gap line between two ferrites in the image; and extracting the extracted gap line. And measuring each item representing the geometric shape of the video head by using the image processing method. The method of claim 4, wherein the first step is to obtain the distribution profile of the X- and Y-axis reflection intensity for each pixel from the image represented by the reflection intensity of the head, and there will be a gap line by using the center of gravity for the profile distribution of each axis. And a method of estimating an expected area, wherein the shape inspection method of a video head using an image processing technique. 5. The method of claim 4, wherein the second step calculates reflection intensity profiles for each pixel in the boundary between the materials forming the chip in the estimated area, and finds the boundary between two ferrites, and outlines the contour of the portion. The shape inspection method of a video head using the image processing technique characterized by the above-mentioned. 5. The method of claim 4, wherein the contour extraction operator is calculated by Equation (3). [Equation 3] E (i, j) = 3 × (-P (i-1, j) + P (i + 1, j)) -| -P (i-10, j) + P (i + 10, j) ｜ -| -P (i, j-2) + P (i, j + 2) ｜ = 3 × X-Y-Z Since P (i, j) is the reflection intensity profile at the pixel i, j position in the image of the head represented by the reflection intensity, X is the contour data value in the X-axis direction, and Y is the X-axis direction in the pixels i, j. The difference in density at positions 10 pixels apart, and Z is the contour data value in the Y-axis direction. 5. The method of claim 4, wherein each item representing the geometry of the video head comprises a thickness (CW) of a chip, a distance (H ₁ ) of a gap line branch at a top boundary of the chip, a thickness (T _W ) of a gap line, and a measurement before the measurement. And a distance (H ₀ ) from an imaginary reference line determined from a specimen for correction in advance to a gap line, etc. The shape inspection method of a video head using an image processing technique.