JP2007078457A - Defect inspection method on substrate for display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a defect inspection method for detecting defective pixels on a substrate for a display device with high accuracy, as to a display device such as a liquid crystal display. <P>SOLUTION: Electric signals are detected from respective pixels 12 on a liquid crystal panel to prepare a histogram Ha thereof. Based on the histogram Ha, thresholds V1 and V2 are set corresponding to variations in the electric signals. Based on the thresholds V1 and V2, the defective pixels are detected in the plurality of pixels 12. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a method for inspecting a defective pixel of a substrate for a display device, and more particularly to a method for setting a threshold value used for inspecting a defective pixel in a substrate having a driving circuit for a liquid crystal display device.

  As a method for detecting defective pixels in a liquid crystal panel, a voltage is applied to each pixel when a driving circuit such as a TFT is formed to create an operating state, and a probe is applied to the signal line drawn from each pixel in that state. In addition, a contact-type inspection method that directly detects pixel voltages to detect defects, and also detects defects by indirectly measuring pixel voltages with an electro-optic element substrate that creates an operating state and is placed opposite the liquid crystal drive substrate. There is a non-contact type inspection method.

  In addition, when the liquid crystal panel is completed, the liquid crystal panel is turned on, the display state is acquired as an electric signal by a CCD camera or the like, and an image generated based on the acquired electric signal is subjected to image processing to detect a defect. There is also a method.

  In any method, in order to detect a defective pixel, a method for separating a non-defective pixel and a defective pixel based on an electric signal obtained from each pixel is necessary.

  A defective pixel is expected to have a signal value that is significantly different from that of a non-defective pixel. For example, when a histogram is created for voltage signals measured on the entire panel or in a small area, it is considered that defective pixels exist outside the non-defective pixel distribution as shown in FIG. That is, if a threshold value is provided between a non-defective pixel and a defective pixel based on this histogram, a defective pixel can be detected.

As a method of determining the threshold value, a histogram is created for the pixel voltage measured on the electro-optic element substrate, and a representative value V0 such as a mode value or an average value of the pixel voltage is obtained based on the histogram, and the representative value V0 is constant. A method has been proposed in which upper and lower thresholds V1 and V2 are determined by adding and subtracting the voltage δ (see, for example, Patent Document 1).
Japanese Patent Laid-Open No. 11-174597

  The signal to be measured varies depending on the measurement state. For example, in the case of a contact method using a probe, the contact resistance becomes a noise source. Further, in the case of a method for acquiring an image with a CCD camera, the light amount of the entire panel input to the CCD camera, noise of the amplifier circuit, and the like cause luminance variations. Furthermore, individual differences may occur due to variations in the manufacturing process of the liquid crystal panel to be inspected.

  That is, if defect detection is performed on a plurality of liquid crystal panels with the fixed threshold values V1 and V2 as in the prior art, problems such as missing a defective pixel or conversely detecting a non-defective pixel as a defective pixel occur. In FIG. 6B, only defective pixels are correctly detected by the set thresholds V1 and V2 in panel B, but not only defective pixels but also non-defective pixels are treated as defects in panel A, and panel C Then, the defective pixel is overlooked.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a defect inspection method capable of detecting a defective pixel of a substrate for a display device with high accuracy in a display device such as a liquid crystal display. There is to do.

  In order to solve the above-described problems, the display device substrate defect inspection method of the present invention is configured as follows.

  That is, an electrical signal is detected from a plurality of pixels on the display device substrate, a histogram thereof is created, a threshold value corresponding to the variation of the electrical signal is set based on the histogram, and the threshold value is set based on the threshold value. A defect inspection method for a substrate for a display device, wherein a defective pixel is detected from a plurality of pixels.

  ADVANTAGE OF THE INVENTION According to this invention, the precision which detects the defective pixel of the board | substrate for display apparatuses can be improved.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  First, the first embodiment will be described with reference to FIGS.

  FIG. 1 is a schematic view showing a liquid crystal driving substrate and an array tester according to the first embodiment of the present invention.

  In FIG. 1, reference numeral 10 denotes a liquid crystal driving substrate of a liquid crystal panel which is a defect inspection target of the present invention, and reference numeral 20 denotes an array tester used for inspection of defects of the liquid crystal panel.

  The liquid crystal driving substrate 10 is a substrate before being bonded to a so-called liquid crystal counter substrate (not shown), and a plurality of pixel regions 11 in the present embodiment are provided on the surface thereof. These pixel regions 11 are regions that become a display screen of a liquid crystal panel by division after bonding, and a plurality of pixels 12 are arranged in a matrix in each pixel region 11.

  FIG. 2 is an enlarged schematic view showing the liquid crystal panel according to the embodiment.

  As shown in FIG. 2, each pixel 12 is referred to as a pixel electrode 13 for driving a liquid crystal material (not shown) and a thin film transistor 14 (hereinafter referred to as “TFT”) for charging / discharging the pixel electrode 13. ).

  Each TFT 14 has a source electrode 14a, a drain electrode 14b, and a gate electrode 14c. The signal line 15 is connected to the source electrode 14a, the pixel electrode 13 and the auxiliary capacitor 19 are connected in parallel to the drain electrode 14b, and the gate line 16 is connected to the gate electrode 14c.

  The signal line 15 and the gate line 16 are formed in a lattice shape along the boundary line between the adjacent pixel 12 and the pixel 12, and an electric signal is applied to the signal line 15 and the gate line 16 from the input unit. The pixel electrode 13 and the auxiliary capacitor 19 connected to the drain electrode 14b can be charged and discharged.

  Each signal line 15 is electrically connected to one of a plurality of pads 18 formed in a line at the surface end of the liquid crystal drive substrate 10 via a switching circuit 17 formed on the liquid crystal drive substrate 10. .

  The switching circuit 17 bundles a plurality of signal lines 15, selects one signal line 15 from these and connects it to the pad 18. Therefore, the number of pads 18 is smaller than that of the pixel electrodes 13 formed on the liquid crystal driving substrate 10.

  As shown in FIG. 1, the array tester 20 is a so-called probe contact type tester, and includes a plurality of testers 21. A probe 23 is connected to each tester 21 via a measurement cable 22. One pixel selected by the switching circuit 17 is obtained by bringing the tip of the probe 23 into contact with the pad 18 of the liquid crystal driving substrate 10. The voltage of the electrode 13, that is, the pixel voltage can be detected.

  As a method of applying a voltage to the pixel electrode 13, there are a method of applying a driving voltage from the array tester 20 to the input unit of the liquid crystal driving substrate 10, a method of applying a driving voltage to the input unit of the liquid crystal driving substrate 10 using another device, and the like. .

  In the present embodiment, the pixel voltage of the pixel 12 is detected. However, the present invention is not limited to this. For example, the charge amount of the auxiliary capacitor 19 may be detected. Alternatively, an electro-optical element substrate may be disposed opposite to the liquid crystal driving substrate 10, and the amount of light of each pixel 12 may be converted into a voltage and detected by the electro-optical element substrate.

  Next, a defect inspection process for detecting defective pixels from the liquid crystal panel will be described with reference to FIG.

  When a defective pixel is detected from the liquid crystal panel, first, a drive voltage is applied from the array tester 20 to the input portion of the liquid crystal drive substrate 10 to bring each pixel 12 into an operating state. Then, the probe 23 of the array tester 20 is brought into contact with each pad 18 of the liquid crystal driving substrate 10 to detect the pixel voltage of each pixel 12.

  Next, a histogram Ha is created for the measured pixel voltage of each pixel 12, and a threshold value for separating the pixel 12 into a non-defective pixel and a defective pixel is set based on the histogram Ha.

  FIG. 3 is an explanatory diagram for explaining a threshold setting method according to the embodiment.

  As shown in FIG. 3, when the threshold value is set, first, in the histogram Ha, a region a (hereinafter referred to as a “first region”) that is the largest among the regions in which data is accumulated exists. A pixel distribution, a region b (hereinafter referred to as a “second region”) scattered around the first region a is regarded as a defective pixel distribution.

  This is based on an empirical rule that, among the pixels 12 of the liquid crystal panel, the pixel voltage of the non-defective pixel is almost the same, and the pixel voltage of the defective pixel is significantly different from the pixel voltage of the non-defective pixel. .

  When the histogram Ha is approximately normally distributed as in the present embodiment, the first region a has a width obtained by multiplying the standard deviation σa by a constant α around the mode V0 of the histogram Ha. It is thought that it will fit almost inside.

  Therefore, in the present embodiment, the value α × σa obtained by multiplying the standard deviation σa by the constant α is added to or subtracted from the mode V0 as the upper and lower threshold values V1 and V2 for separating non-defective pixels and defective pixels. To do.

  Note that the constant α is obtained by measuring pixel voltages for a plurality of liquid crystal panels in advance so that the threshold values V1 and V2 can separate non-defective pixels and defective pixels, that is, the threshold values V1 and V2 are the first region a and the second region. It is determined to fall between b.

The standard deviation σa is shown in the following [Equation 1], and the threshold values V1 and V2 are shown in the following [Equation 2].

na is the total number of pixels, and va is the pixel voltage of each pixel 12.

  It should be noted in calculating the mode value V0 and the standard deviation σa that the standard deviation σa is targeted for non-defective pixels included in the first region a. Therefore, if the mode value V0 and the standard deviation σa are calculated based on the generated histogram Ha as it is, it is not desirable because the data of defective pixels is included.

  Therefore, in the present embodiment, it is assumed that there are at most several defective pixel data, and from the obtained data, a predetermined number of data from a higher pixel voltage and a predetermined number from a lower pixel voltage are obtained. Based on the histogram Ha excluding data, the mode value V0 and the standard deviation σa are calculated.

  Next, when the threshold values V1 and V2 are set, the pixel voltage of each pixel 12 is again measured by the array tester 20, and the plurality of pixels 12 are divided into two by comparing the measured value with the threshold values V1 and V2. Then, the pixel 12 having a pixel voltage outside the range between the threshold values V1 and V2 is regarded as a defective pixel, and position information of the pixel 12 is created.

  This completes the defect inspection process for detecting defective pixels from the liquid crystal panel. The liquid crystal panel that has completed the defect inspection process is carried out to the next repair process, and defective pixels are repaired based on the obtained position information.

  In the present embodiment, the mode V0 and the standard deviation σa are used when setting the thresholds V1 and V2, but the present invention is not limited to this. That is, if the center of pixel voltage variation can be expressed, an average value, median value, or the like may be used instead of the mode value V0. If the degree of pixel voltage variation can be expressed, the standard value may be used. Other parameters may be used instead of the deviation σa.

  According to the above-described defect inspection method for a liquid crystal panel, the mode value V0 and the standard deviation σa are calculated based on the pixel voltage histogram Ha created for each liquid crystal panel, and the mode value V0 and the standard deviation σa are used. The threshold values V1 and V2 are set. That is, threshold values V1 and V2 corresponding to pixel voltage variations are set for each liquid crystal panel.

  Therefore, even if the variation in pixel voltage of each pixel 12 is different for each liquid crystal panel, defective pixels can be detected with high accuracy.

  Next, a second embodiment of the present invention will be described with reference to FIGS. In addition, about the same structure as the said embodiment, the same code | symbol is attached | subjected and the description is abbreviate | omitted. Also in the present embodiment, in the histogram Ha, the first region a that is the largest among the regions in which data is present is scattered in the distribution of non-defective pixels, and the first region a that is scattered around the first region a. The region b of 2 is regarded as a defective pixel distribution.

  FIG. 4 is a flowchart showing a defect detection process according to the second embodiment of the present invention, and FIG. 5 is an explanatory diagram for explaining a threshold setting method according to the embodiment.

  As shown in FIG. 4, when detecting a defective pixel from the liquid crystal panel, first, a drive voltage is applied from the array tester 20 to the input portion of the liquid crystal drive substrate 10 to put each pixel 12 into an operating state. Then, the probe 23 of the array tester 20 is brought into contact with each pad 18 of the liquid crystal drive substrate 10, and the pixel voltage of each pixel 12 is detected (step S1).

  Next, a histogram Ha is created for the measured pixel voltage of each pixel 12 (step S2), and a threshold value for separating the pixel 12 into a non-defective pixel and a defective pixel, which is a point of the present invention, is set based on the histogram Ha. I do.

  When setting the threshold, first, the count number of data, that is, the frequency of the pixel 12 is counted in the direction in which the pixel voltage increases with reference to the mode value V0 of the histogram Ha (step S3). It is determined whether there is a portion where the frequency of 0 is zero, that is, whether the histogram Ha is interrupted (step S4).

  If the histogram Ha is interrupted (YES in step S4), it is determined whether or not the width of the region where the histogram Ha is interrupted exceeds a voltage section Δ specified in advance (step S5). On the other hand, when the frequency of the histogram Ha is not interrupted (NO in step S4), the frequency of the pixel 12 is counted again in the direction in which the pixel voltage increases with reference to the mode value V0 of the histogram Ha again. (Step S3).

  If the width of the region where the histogram Ha is interrupted exceeds the voltage interval Δ (YES in step S5), the pixel voltage Vmax immediately before the histogram Ha is interrupted is set as the upper limit threshold value V1 of the non-defective pixels (step S6). . On the other hand, when the width of the region where the histogram Ha is interrupted does not exceed the voltage interval Δ (NO in step S5), the pixel voltage rises again on the basis of the mode value V0 of the histogram Ha. The frequency of the pixel 12 is counted (step S3).

  When the upper limit threshold value V1 of the non-defective pixels is set, the count number of data, that is, the frequency of the pixel 12 is counted in the direction in which the pixel voltage decreases with the mode value V0 of the histogram Ha as a reference ( Step S7), it is determined whether or not the histogram Ha is interrupted (step S8).

  If the histogram Ha is interrupted (YES in step S8), it is determined whether or not the width of the region where the histogram Ha is interrupted exceeds the voltage interval Δ (step S9). On the other hand, when the histogram Ha is not interrupted (NO in step S8), the frequency of the pixel 12 is counted again in the direction in which the pixel voltage decreases with reference to the mode value V0 of the histogram Ha again (step S8). S7).

  When the width of the region where the histogram Ha is interrupted exceeds the voltage interval Δ (YES in step S9), the pixel voltage Vmin immediately before the histogram Ha is interrupted is set as the lower limit threshold V2 of the non-defective pixels (step S10). . On the other hand, when the width of the region where the histogram Ha is interrupted does not exceed the voltage interval Δ (NO in step S10), the pixels are again increased in the direction in which the pixel voltage increases with reference to the mode value V0 of the histogram Ha. The frequency of 12 is inspected (step S7).

  When the upper and lower thresholds V1 and V2 for the non-defective pixels are set, a driving voltage is again applied from the array tester 20 to the input unit of the liquid crystal driving substrate 10 to set each pixel 12 in an operating state. Then, the probe 23 of the array tester 20 is brought into contact with each pad 18 of the liquid crystal driving substrate 10 to detect the pixel voltage of each pixel 12.

  Then, the detected pixel voltage of each pixel 12 is compared with the threshold values V1 and V2 of the non-defective pixels, and the pixel 12 that is outside the thresholds V1 and V2 is detected as a defective pixel (step S11). This completes the detection process for detecting defective pixels from the liquid crystal panel.

  That is, in the present embodiment, as shown in FIG. 5, the voltage interval Δ is designated in advance in the direction in which the voltage rises with reference to the highest pixel voltage Vmax among the pixels 12 included in the first region a. If the pixel 12 is not counted over this time, the above Δ is added to the highest pixel voltage Vmax, and this is set as the upper limit threshold value V1 of the non-defective pixels.

  In addition, when the pixel 12 is not counted over the voltage section Δ specified in advance in the direction in which the voltage decreases with reference to the lowest pixel voltage Vmin among the pixels 12 included in the first region a, the lowest pixel The above Δ is subtracted from the voltage Vmin, and this is set as the lower limit threshold V2 of the non-defective pixels.

  In the voltage interval Δ, the pixel voltages of a plurality of liquid crystal panels are measured in advance so that the non-defective pixels and the defective pixels can be separated by the threshold values V1 and V2, that is, the threshold values V1 and V2 are the first region a and the second region. It is determined to enter between the areas b.

The threshold values V1 and V2 are shown in the following [Equation 3].

  In this way, appropriate threshold values V1 and V2 can be set regardless of the voltage distribution shape of the pixel electrode 13, so that a single inspection apparatus performs defect inspection of a plurality of types of liquid crystal panels having different distribution shapes. be able to. Furthermore, it is not necessary to know in advance the degree of variation in pixel voltage for each type of liquid crystal panel.

  In the present embodiment, the mode V0 is used when setting the thresholds V1 and V2, but the present invention is not limited to this. That is, any value may be used as long as the non-defective pixels are present in the first region a in which the non-defective pixels are distributed.

  The present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

Schematic which shows the liquid-crystal drive substrate and array tester which concern on the 1st Embodiment of this invention. The schematic diagram which expands and shows the liquid crystal panel which concerns on the same embodiment. Explanatory drawing explaining the setting method of the threshold value concerning the embodiment. The flowchart which shows the defect inspection process which concerns on the 2nd Embodiment of this invention. Explanatory drawing explaining the setting method of the threshold value concerning the embodiment. Explanatory drawing explaining the setting method of the conventional threshold value.

Explanation of symbols

  12: Pixel, α: Constant, Ha: Histogram, V1: Threshold, V2: Threshold, V0: Mode, σa: Standard deviation.

Claims (6)

An electrical signal is detected from a plurality of pixels on the display device substrate, and a histogram thereof is created.
Based on the histogram, set a threshold according to the variation of the electrical signal,
A defect inspection method for a substrate for a display device, wherein a defective pixel is detected from the plurality of pixels based on the threshold value.   2. A defect inspection method for a substrate for a display device according to claim 1, wherein a representative value and a standard deviation of the electric signal are calculated based on the histogram, and the threshold value is set based on the representative value and the standard deviation. .   3. The display device substrate defect inspection method according to claim 2, wherein the threshold value is set by adding or subtracting a value obtained by multiplying the standard deviation by a predetermined constant to the representative value.   The display device substrate defect inspection method according to claim 2, wherein the representative value is a mode value or an average value of the electric signal. In the above histogram,
A maximum existence region in which the frequency of the pixel does not become zero continuously over a predetermined interval;
When there is a non-existing region where the frequency of the pixel is zero continuously over a predetermined interval,
The threshold value is set to a value of an electrical signal that is deviated by a predetermined value from the value of the electrical signal located closest to the non-existing region in the existing region in a direction away from the existing region. The defect inspection method for a substrate for a display device according to claim 1.   2. The defect inspection method for a display device substrate according to claim 1, wherein the plurality of pixels are divided into two by the threshold value, and one of them is detected as a defective pixel.

Images