JPH11185037A - Device and method for processing defect information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the omission of a defect by amplifying signal change corresponding even to a slight defect of an object to be inspected. SOLUTION: Two-dimensional image of an object to be inspected is fetched, and the two-dimensional image is converted into multilevel digital data in an AD converter 2, while a vertical filter 4 calculates a difference value in a vertical direction for respective picture elements in a prescribed range for multilevel digital image data and a horizontal filter 5 calculates a difference value in a horizontal direction for the difference value calculated by the vertical filter 4. A binarization circuit 7 converts the difference value calculated by the horizontal filter 5 into binary data based on prescribed reference value, and the defective part of the object to be inspected is detected based on the multilevel digital data and the binary data as the difference value calculated in the horizontal filter 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect information processing apparatus and method for capturing an image of a cylindrical surface of a photosensitive drum or the like and detecting defect information relating to smoothness and homogeneity from the image data.

[0002]

2. Description of the Related Art In recent years, the resolution of an image formed by a copying machine or the like or an image displayed on a liquid crystal display or the like has become higher and higher than the resolution of the human eye. With the progress of this technology, a slight defect existing on the surface of a photosensitive drum, a developing sleeve, a liquid crystal display substrate, or a dedicated photographic paper used in a copying machine, a printer, or the like greatly affects the improvement in resolution. Further, the defect itself to be detected has become extremely fine due to the sophistication of the resolution.

FIG. 7 is a circuit block diagram of a conventional defect information processing apparatus.

As shown in FIG. 7, a conventional defect information processing apparatus scans an inspection surface of an inspection object by irradiating a laser beam, and inputs an analog image of the inspection surface by a light receiving device. An unnecessary frequency component is removed from the analog image by an analog filter 91, and the analog image is converted from an analog signal to a digital signal through an AD conversion circuit 92. The image data is stored in the image memory 95 as multi-valued digital data, converted into a binary signal through the binarization circuit 93, and stored in the binary image memory 96.

The CPU 94 calculates defect data corresponding to a defect on the inspection surface from the multi-valued digital image data stored in the image memory 95 and the binary signal stored in the binary image memory 96, and calculates the defect data. The maximum value, the minimum value, the number of pixels, the coordinates, and the like are stored in the feature data memory 97 as feature data. Then, the CPU 94 sets the feature data memory 9
The quality of the object to be inspected is determined from the characteristic data stored in 7, and the determination result is displayed on a CRT or the like via the output interface 98.

[0006]

However, the filter used in the conventional apparatus only works on an image of the laser beam in the main scanning direction due to its time-sequential operation, so that the detection signal does not change in the main scanning direction. In such a case, defect data cannot be obtained through an analog filter. For example, when a photosensitive drum made of aluminum is manufactured by drawing, and a defect such as a flaw that is elongated in the longitudinal direction of the cylindrical surface of the drum occurs, even if the photosensitive drum is scanned in the longitudinal direction (main scanning direction) of the drum, the detection signal is not detected. There is a problem that the defect is hard to be detected because there is almost no change.

When a two-dimensional spatial filter is used to solve the above problem, if the filter size in the sub-scanning direction of the laser beam is set to 5 lines or more,
There is a problem that the hardware configuration becomes large-scale and the cost becomes high.

Further, when converting the binarized signal into the coordinate value of the change point, since the minimum detection size is one pixel, it is necessary to store the coordinate value of the change point in the memory at the time of one pixel clock. However, in order to realize this, there is a problem that a high-speed memory is used, which leads to an increase in cost.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and a first object of the present invention is to amplify a signal change corresponding to even a small defect in the longitudinal direction of an object to be inspected so that the defect is not overlooked. It is an object of the present invention to provide a defect information processing apparatus and method which can perform the defect processing.

A second object of the present invention is to provide a defect information processing apparatus capable of improving a defect detection capability and reducing an apparatus cost by satisfying necessary functions while miniaturizing a filter circuit in a main scanning direction of a laser beam. Is to provide a way.

A third object of the present invention is to increase the interval at which a change point of the binarized data appears and to secure a sufficient time for writing the binarized data to the memory, so that a memory having a low processing speed can be used. An object of the present invention is to provide a defect information processing apparatus and method which can be used and can reduce component costs.

[0012]

In order to solve the above problems and achieve the object, a defect information processing apparatus according to the present invention comprises the following means. That is, image input means for capturing a two-dimensional image of the inspection object, digital conversion means for converting the two-dimensional image into multi-value digital data, and vertical filter means for vertically filtering the multi-value digital image data, Horizontal filter means for filtering the multi-value digital image data filtered by the vertical filter means in the horizontal direction; and converting the multi-value digital data filtered by the horizontal filter means into binary data based on a predetermined reference value. And a detecting means for detecting a defective portion of the inspection object based on the multi-valued digital data and the binarized data.

The defect information processing method of the present invention includes the following steps. That is, an image input step of capturing a two-dimensional image of the inspection object, a digital conversion step of converting the two-dimensional image into multi-valued digital data, and a vertical filtering step of vertically filtering the multi-valued digital image data, A horizontal filtering step of filtering the multi-valued digital image data filtered in the vertical filtering step in a horizontal direction, and converting the multi-valued digital data filtered in the horizontal filtering step into binary data based on a predetermined reference value And a detecting step of detecting a defective portion of the inspection object based on the multi-valued digital data and the binarized data.

[0014]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. [First Embodiment] A defect information processing apparatus according to a first embodiment will be described with reference to FIG.

FIG. 1 is a circuit block diagram of a defect information processing apparatus according to a first embodiment of the present invention.

As shown in FIG. 1, 1 is an input amplifier, and 2 is an AD that converts an analog signal into an 8-bit digital signal.
Converter 3, digital comparator 4, vertical 1
Dimensional filter, 5 is a one-dimensional filter in the horizontal direction, 6 is 8
An image memory 7 for storing multi-valued digital image data of bits, binarizing the multi-valued digital image data by comparing it with a predetermined reference value, and having a horizontal one-dimensional filter function;
A binarizing circuit, 8 is a run memory for converting the binarized signal into a run code and storing the run code; 9 is a run from the multi-level digital image data stored in the image memory 6 and the binarized signal stored in the run memory 8 A control computer that divides a code into one continuous area, calculates a maximum value, a minimum value, an area, coordinates, and the like as respective feature data, and determines a defect when the feature data deviates from any predetermined standard value. Reference numeral 10 denotes a timing generation circuit which receives a horizontal synchronizing signal, a vertical synchronizing signal, and a start signal and generates operation timing of each circuit.

Next, a mechanical configuration of the defect information processing apparatus according to the first embodiment will be described with reference to FIG.

FIG. 2 is a diagram showing a mechanical configuration of the defect information processing apparatus according to the first embodiment of the present invention.

As shown in FIG. 2, 11 has a diameter of about 100 μm.
a laser beam scanner for one-dimensionally scanning a laser beam spot of m; 12 a beam detector for detecting a laser beam; 13 a developing sleeve of a copying machine cartridge for the object to be inspected; 1 is a signal processing unit shown in FIG. 1, 16 is a sequencer that controls the entire apparatus, and 17 is a sub-scanning direction by rotating the developing sleeve 13 of the inspection object at a constant speed. Motor and a clamp mechanism for scanning.

When the entire surface of the object to be inspected is scanned by the laser scanner 11, the signal detected by the light receiver 14 is represented as a time-series two-dimensional signal as shown in FIG.

Since the laser light is scanned along the generatrix of the developing sleeve, the main scanning direction is the longitudinal direction of the sleeve. When the developing sleeve 13 is rotated slowly with respect to the scanning cycle of the laser beam in the main scanning direction, the sub-scanning can be performed, and the rotation direction of the developing sleeve 13 becomes the sub-scanning direction.

In the entire scanning area, 10M in the horizontal direction
Sampling is performed with a Hz clock, the number of pixels is about 1000 pixels, the length of the inspection object is about 300 mm, and the resolution is about 30 μm. The inspection object is about 12 mm in diameter.
By scanning 2,000 lines in the main scanning direction, the surface of the object to be inspected is rotated so as to make one round, and the resolution per line becomes about 20 μm.

At the time of scanning with a laser beam, a detection signal from the photodetector 14, a horizontal synchronization signal from the beam detector, and a vertical synchronization signal from a rotation detector (not shown) for detecting rotation of the inspection object are signals. Output to the processing unit 15.

In the signal processing unit 15, the detection signal is amplitude-converted by the input amplifier 1 to the input range of the AD converter 2, and A
The digital signal is converted by the D converter 2 into an 8-bit digital signal. The digital signal is compared with a predetermined value by the comparator 3. If the digital signal exceeds the predetermined value, the digital signal is output and it is known that the digital signal is normal. It can be seen that a failure has occurred in the device. The output signal from the comparator 3 is output as a light amount detection signal to a sequencer, which is a higher-level controller.

On the other hand, the digital signal is subjected to one-dimensional filter processing in the vertical direction by the vertical filter 4, and subsequently to one-dimensional filter processing in the horizontal direction by the horizontal filter 5. The contents of the filter processing can be arbitrarily changed.
When the difference between the difference value of 27 lines and the difference value of the scanning lines from 50 lines to 53 lines is calculated, the base line of the signal corresponding to the length of about 2 mm is calculated, and the length of about 80 μm is calculated. Calculating the size of the signal corresponding to this, and calculating the difference between them, a gradual signal change having a period of 2 mm or more such as undulation of the surface of the inspection object is canceled, and a defect signal of about 80 μm is canceled. Can be emphasized.

Similarly, the content of the filtering process by the horizontal filter 5 can be set arbitrarily to emphasize the waveform of the signal band to be detected as a defect signal from the filtered signal.

Further, by arranging the vertical filter 4 before the horizontal filter 4, it is possible to emphasize a defect signal such as an elongated flaw along the main scanning direction, which cannot be achieved only by the analog signal filtering. Therefore, the defect on the surface of the inspection object is not overlooked, and the detection accuracy is improved.

FIG. 4 is a detailed diagram of the vertical filter, the horizontal filter, and the binarization circuit shown in FIG.

As shown in FIG. 4, both the vertical filter 4 and the horizontal filter 5 are composed of non-recursive digital filters. The vertical filter 4 has a delay line variable up to 128H.

In FIG. 4, the vertical filter 4 is provided with delay lines 18 to 20 from the nearest line to three lines before. In the vertical difference processing of the vertical filter 4, the difference value between the pixel value in the range of the nearest line (input signal) and the line three lines before (delay line 18), and the difference value between the line two lines before (delay line 19) and one line The difference (or sum) between the pixel value and the difference value in the range of the previous line (delay line 20) is calculated. That is, in the delay line 23,
The pixel value of the most recent input signal is added to the difference value one line before (21), and the pixel value is subtracted from the pixel value of the line three lines before (22). Similarly, in the delay line 26, the pixel value of the preceding line is added to the difference value of the preceding line (24), and the pixel value is subtracted from the pixel value of the preceding line (25). Then, the difference value on the delay line 23 and the difference value on the delay line 26 are further subtracted and output to the horizontal filter 5. In the vertical filter 4, by continuing to perform the above-described difference calculation on the delay lines 23 and 26, it is possible to always hold the sum of the pixel values in the range from the nearest line to a predetermined line before (three lines before). Further, although not shown, by selecting and extracting pixel values in a predetermined range from the calculation result, a calculation of the sum / 2 和 n (representing 2 to the power of n) can be performed. By calculating the difference between the pixel values in the two ranges, it is possible to realize a calculation that removes the undulation of the input signal and emphasizes the defect signal.

If the vertical filter 4 is arbitrarily added with a function, the circuit scale becomes large and the cost is increased, which is not practical.
For this reason, the configuration is such that the filter function is limited to only the filter constant of the difference calculation suitable for enhancing the defect signal. As described above, even if the filter function is suppressed, the function is sufficient and the circuit scale can be reduced.

The horizontal filter 5 has a simple structure,
A 28-tap filter IC 28 made into one chip can be used, and one having an arbitrary response characteristic can be selected. The horizontal filter 5 subtracts the difference value adjacent in the horizontal direction from the difference value for one line output from the vertical filter 4 and outputs the result to the image memory 6 and the binarization circuit 7.

When a material is sprayed on the surface of the developing sleeve as an object to be inspected, a gentle undulation is generated in the coating, and even if the undulation on the surface is a small defect, the undulation adversely affects the image quality. Therefore, the vertical filter 4 first detects multi-value digital data for each pixel corresponding to the surface shape in the vertical direction, and then detects multi-value digital data for each pixel corresponding to the surface shape in the horizontal direction. The undulations in the vertical and horizontal directions can be reliably detected.

As shown in FIG. 4, the binarizing circuit 7 sets two kinds of reference values to the multi-valued digital data subjected to the difference operation processing by the vertical filter 4 and the horizontal filter 5. If the output multi-valued digital data is within the range of each reference value of the latches 29 and 30, it is converted to a binary signal as "0", and if not, "1".

The binarizing circuit 7 calculates the logical sum of the first reference value of the latch 29 and the multi-value digital data and the logical sum of the second reference value of the latch 30 and the multi-value digital data. When the value is larger than the reference value (the output of the circuit 31 is 1) or smaller than the second reference value (the output of the circuit 32 is 1), the output of the OR circuit 33 is 1, and otherwise, the output is 0. The binarized output value of the OR circuit 33 horizontally extends the pixel value of which the value is 1 in the one-shot circuit 35 having the latches 34 of a predetermined expansion number.
The output value of the one-shot circuit 35 is
6 and the pixel value is inverted via an inverter 37, and then output to a one-shot circuit 39 having a latch 38 with a predetermined number of contractions, and again inverted by an OR circuit 40 and an inverter 41 to form a binary signal. Output to the run memory 8. By inverting the output value of the one-shot 39 again by the OR circuit 40 and the inverter 41, the pixel value of 0 can be extended horizontally.

The one-shot circuits 35, 3 of the binarization circuit 7
9 by cascade connection, the isolated point removal function,
A function of connecting defective binarized signals (run codes) to specify a defective portion can be realized.

FIG. 5 is a detailed diagram of the image memory and the run memory shown in FIG.

As shown in FIG. 5, the image signal of the multivalued digital data output from the horizontal filter 5 has a frequency of 10 MHz.
Since the data is 8-bit data, the bus width conversion circuit 42 collectively writes the data in units of 4 bytes as 32-bit data, and sequentially writes the data in the 32-bit memory 44 at the address specified by the address counter 45 at a clock frequency 43 of 1/4. It is.

The binarized signal (run code) output from the binarizing circuit 7 stores the value of the counter 46, which is the horizontal coordinate X, and the value of the counter 47, which is the vertical coordinate Y, in a 32-bit 32-bit memory. 52 are sequentially written. As the write timing, the X coordinate of the point where the run code changes from 0 to 1 and the X coordinate of the point where the run code changes from 1 to 0 are collectively written to the memory 52 for each run. The Y coordinate is sequentially written to the memory 52 together with a flag for distinguishing between the X coordinate and the Y coordinate only when a run exists on the current line.

By the above-described circuit operation, the run codes according to the binarization determination of the entire predetermined inspection area and the multi-value digital data can be stored in the memories 6 and 7, and these memory data are stored in the control computer. 9 can be freely accessed.

The memory data is processed by software by the control computer 9.

First, the control computer 9 calculates the feature value of the binary signal in units of run codes. That is, the X-coordinate is read from the beginning of the run memory 8, the X-coordinate is read, and the X-coordinate is read until a flag indicating the Y-coordinate is found and stored in the main memory (not shown). Since all the X coordinates on the main memory are on the line of the Y coordinate found here, two-dimensional coordinates on the inspection area can be determined. Then, the multi-valued digital data at a location corresponding to the range of the run is read from the image memory 6, the maximum value, the minimum value, and the number of pixels are calculated, and stored together with the coordinates in the array of the main memory as the run attribute value. This is executed for all the run codes stored in the run memory 8.

Next, the control computer 9 refers to the run codes and their attribute values stored in the array of the main memory,
The run codes adjacent to each other in the horizontal direction and the vertical direction of the inspection area are labeled as one integrated defect portion. The run code neighborhood determination is not so-called general 4 neighborhood or 8 neighborhood, but it is determined that anything that is close to the horizontal and vertical directions within a predetermined interval or less is close to each other. If a defect such as lint is generated on the surface during the coating of the developing sleeve, the signal becomes a binary signal that is interrupted due to the thin and long shape, but this can be determined to be one defect. In addition, it is possible to obtain an inspection result in which an actual defect state is accurately determined.

The control computer compares the inspection result with a predetermined standard value to determine OK / NG, and outputs the result to a higher-level sequencer. [Second Embodiment] A defect information processing apparatus according to a second embodiment will be described with reference to FIG.

FIG. 6A is a diagram showing a configuration of a defect information processing apparatus according to a second embodiment of the present invention. FIG.
FIG. 6B is a side view of FIG.

As shown in FIG. 6, reference numeral 61 denotes an illuminating device for illuminating the entire inspection object, and reference numerals 62 to 67 denote line sensor cameras, which receive reflected light from the inspection object in the visual field range of each camera. Reference numeral 15 denotes a signal processing unit that inputs an analog signal detected by each of the cameras 62 to 67 for each camera.

As described above, the resolution of the inspection area can be increased by increasing the number of line sensors. Further, by receiving the specularly reflected light and the scattered light of the inspection object by different line sensors, the types of detectable defects can be increased, and the quality can be more reliably guaranteed.

In the second embodiment, inside the signal processing unit 15, circuits corresponding to one channel shown in FIG. 1 are mounted for the number of line sensors (six systems in this example).

The configuration and operation of the signal processing unit 15 are as follows.
As in the first embodiment, the maximum value for each defect data,
The minimum value, area, coordinates, etc. can be obtained.

As described above, according to the present embodiment, the vertical filter can apply one-dimensional filter processing such as smoothing and differentiation to the multi-valued digital data in the sub-scanning direction to emphasize the defect signal. . In addition, the horizontal filter can perform one-dimensional filter processing such as smoothing and differentiation in the main scanning direction on the multi-valued digital data to emphasize a defect signal.

The binarization circuit can reduce the spatial frequency of the binarized signal by removing one isolated pixel or connecting adjacent but separated pixels.

The vertical filter calculates the difference value of the multivalued digital data in a predetermined line range, and reduces the random noise component and the high frequency component by reducing the averaged line range to the defect size to be detected. S / N of signal
The ratio can be increased.

The vertical filter calculates a difference value of the multi-value digital data in another line range, and averages a range similar to a large swell component of a detection signal due to mechanical shake or warpage of the inspection object. This makes it possible to extract only the swell component.

The vertical filter calculates the difference between these two difference values and subtracts the swell component from the detection signal having the improved S / N ratio to emphasize and extract only the signal of the defect to be detected. Can be.

Further, the binarization circuit performs an expansion process in which a predetermined pixel value adjacent to a pixel having a pixel value of 1 is also 1;
The pixel values between adjacent but separated pixel values 1 are set to 1 to fill and connect the pixels.

Further, the binarization circuit performs a contraction process for setting a predetermined pixel adjacent to a pixel having a pixel value of 0 to 0, thereby removing an isolated pixel having a pixel value of 1. Further, by performing the contraction processing on the data subjected to the expansion processing, a set of adjacent pixel values 1 remains without being removed, and pixels having the pixel value 1 that is isolated at intervals of a predetermined pixel or more are removed.

Therefore, a large detection signal can be obtained by providing a digital filter having a large number of taps in the horizontal and vertical directions even if the defect has a small change in the main scanning direction.

Further, by satisfying the required filter function while keeping the circuit scale of the filter processing in the main scanning direction small, the defect detection capability can be improved and the apparatus cost can be reduced.

Further, by increasing the interval at which the transition point of the binarized signal appears, and by ensuring sufficient time for writing the coordinate values of the run code into the memory, a low-speed memory can be used and the apparatus cost can be reduced. it can.

[0060]

Another embodiment of the present invention provides a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and a computer (or CPU or MPU) of the system or apparatus. ) May read out and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided on a function expansion board inserted into the computer or a function expansion unit connected to the computer, based on the instructions of the program code, It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

When the present invention is applied to the storage medium, the storage medium stores program codes corresponding to the flowcharts described above.

[0066]

As described above, according to the present invention, even a small defect in the longitudinal direction of the inspection object can amplify a signal change corresponding to the defect and eliminate oversight of the defect.

Further, the capability of detecting defects can be improved, and the cost of the apparatus can be reduced.

Further, a memory having a low processing speed can be used, and the cost of parts can be reduced.

[0069]

[Brief description of the drawings]

FIG. 1 is a circuit block diagram of a defect information processing apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a system configuration of a defect information processing apparatus according to the first embodiment of the present invention.

FIG. 3 is a diagram illustrating a scanning area defined by a main scanning direction and a sub scanning direction according to the embodiment.

FIG. 4 is a block diagram of a vertical filter, a horizontal filter, and a binarization circuit of FIG. 1;

FIG. 5 is a block diagram of an image memory and a run memory of FIG. 1;

FIG. 6 is a diagram illustrating a system configuration of a defect information processing apparatus according to a second embodiment of the present invention.

FIG. 7 is a circuit block diagram of a conventional defect information processing apparatus.

Claims (35)

[Claims] 1. An image input unit for capturing a two-dimensional image of a test object, a digital conversion unit for converting the two-dimensional image into multi-value digital data, and a vertical filter for filtering the multi-value digital image data in a vertical direction Means, horizontal filter means for horizontally filtering the multi-value digital image data filtered by the vertical filter means, and binarization of the multi-value digital data filtered by the horizontal filter means based on a predetermined reference value A defect information processing apparatus comprising: binarizing means for converting data into data; and detecting means for detecting a defective portion of the inspection object based on the multi-valued digital data and the binarized data. 2. The vertical filter means comprises: first calculation means for calculating a difference value of the multi-value digital data in the first range in the vertical direction; and the multi-value digital data in the second range in the vertical direction. 2. The apparatus according to claim 1, further comprising: second operation means for calculating a difference value of data; and difference operation means for calculating a difference between the difference values calculated by the first and second operation means. The defect information processing apparatus according to the above. 3. The binarizing unit according to claim 1, wherein the multi-valued digital data filtered by the horizontal filter unit exceeds a first reference value or falls below a second reference value. 1 and the first and second
3. The defect information processing apparatus according to claim 1, wherein the pixel value is set to 0 when the pixel value is between the reference values. 4. The binarizing means sets a pixel value of a pixel adjacent to a pixel having a pixel value of 1 to 1
Expansion means, and the binarized data processed by the expansion means,
4. The defect information processing apparatus according to claim 3, further comprising: a shrinking unit that sets a pixel value of a pixel adjacent to a pixel having a pixel value of 0 to 0. 5. The dilation means sets a pixel value of a pixel which is separated from the pixel having the pixel value of 1 but is close in a predetermined area to 1, and connects the pixels. The defect information processing apparatus according to claim 4. 6. The method according to claim 4, wherein the shrinking means sets a pixel having a pixel value of 1 which is isolated from a set of pixels having a pixel value of 1 and separated by a predetermined area to a pixel value of 0. The defect information processing apparatus according to the above. 7. A storage unit for storing the multi-value digital data filtered by the horizontal filter unit, and calculating the characteristic data of the pixel having the pixel value 1 based on the multi-value digital data and the binarized data. The defect information processing apparatus according to claim 3, further comprising a calculating unit. 8. The defect information processing apparatus according to claim 7, wherein the feature data is a maximum value, a minimum value, the number of pixels, and a coordinate value of the multilevel digital data. 9. The defect information processing apparatus according to claim 8, further comprising a determination unit configured to determine whether the inspection object is a non-defective product based on the characteristic data. 10. The defect information processing apparatus according to claim 2, wherein the first range of the vertical filter means is set to correspond to a size of a defect to be detected with respect to the inspection object. . 11. The image sensor according to claim 1, wherein the image input unit is an optical sensor that captures reflected light of light applied to an inspection surface of the inspection object. Defect information processing device. 12. The defect information processing apparatus according to claim 1, wherein the inspection object is a photosensitive drum. 13. The defect information processing apparatus according to claim 1, wherein the inspection object is a developing sleeve. 14. The defect information processing apparatus according to claim 1, wherein the inspection object is a charging roller. 15. The defect information processing apparatus according to claim 1, wherein the inspection object is a photographic paper transport roller. 16. The defect information processing apparatus according to claim 1, wherein the inspection object is a liquid crystal display substrate. 17. The defect information processing apparatus according to claim 1, wherein the object to be inspected is photographic paper. 18. An image inputting step of capturing a two-dimensional image of an object to be inspected, a digital conversion step of converting the two-dimensional image into multi-valued digital data, and a vertical filter for filtering the multi-valued digital image data in a vertical direction A horizontal filtering step of filtering the multi-valued digital image data filtered in the vertical filtering step in a horizontal direction; and converting the multi-valued digital data filtered in the horizontal filtering step into two based on a predetermined reference value. A defect information processing method, comprising: a binarization step of converting the data into digitized data; and a detection step of detecting a defect portion of the inspection object based on the multi-valued digital data and the binarized data. 19. The vertical filtering step includes: a first calculating step of calculating a difference value of the multi-value digital data in the first range in the vertical direction; and the multi-value digital in a second range in the vertical direction. 19. The method according to claim 18, further comprising: a second calculation step of calculating a difference value of data; and a difference calculation step of calculating a difference between the difference values calculated in the first calculation step and the second calculation step. The defect information processing method described in the above. 20. In the binarizing step, when the multi-value digital data filtered in the horizontal filtering step exceeds a first reference value or falls below a second reference value, the pixel is converted to pixel data as binary data. 20. The defect information processing method according to claim 18, wherein a value 1 is set, and a pixel value 0 is set when the value is between the first and second reference values. 21. The binarizing step, wherein a pixel value of a pixel adjacent to the pixel having a pixel value of 1 is set to 1
And a contraction step of setting a pixel value of a pixel adjacent to a pixel having a pixel value of 0 to 0 for the binarized data processed in the dilation step. 20. The defect information processing method according to 20. 22. In the expanding step, a pixel value of a pixel which is separated from the pixel having the pixel value of 1 but is close in a predetermined area is set to 1, and the pixels are connected to each other. The defect information processing method according to claim 21. 23. In the shrinking step, a set of pixels having a pixel value of 1 and having a pixel value of 1
3. A pixel having a pixel value of 0 is set to 0.
2. The defect information processing method according to 1. 24. A storing step of storing the multi-value digital data filtered in the horizontal filtering step, and calculating the characteristic data of the pixel having the pixel value 1 based on the multi-value digital data and the binarized data. The defect information processing method according to any one of claims 20 to 23, further comprising a calculating step of: 25. The defect information processing method according to claim 24, wherein the characteristic data is a maximum value, a minimum value, the number of pixels, and a coordinate value of the multilevel digital data. 26. The defect information processing method according to claim 25, further comprising a determining step of determining whether or not the inspection object is a non-defective product based on the characteristic data. 27. The defect information processing method according to claim 19, wherein the first range of the vertical filtering step is set to correspond to a size of a defect to be detected with respect to the inspection object. . 28. The image input step, wherein a two-dimensional image of the inspection object is captured using an optical sensor that captures reflected light of light applied to an inspection surface of the inspection object. 28. The defect information processing method according to any one of 18 to 27. 29. The defect information processing method according to claim 18, wherein the inspection object is a photosensitive drum. 30. The defect information processing method according to claim 18, wherein the inspection object is a developing sleeve. 31. The defect information processing method according to claim 18, wherein the inspection object is a charging roller. 32. The defect information processing method according to claim 18, wherein the inspection object is a photographic paper transport roller. 33. The defect information processing method according to claim 18, wherein the inspection object is a liquid crystal display substrate. 34. The defect information processing method according to claim 18, wherein the inspection object is photographic paper. 35. A computer readable memory storing a program code for processing defect information of an inspection object, an image inputting step of capturing a two-dimensional image of the inspection object,
A digital conversion step of converting the two-dimensional image into multi-valued digital data, a vertical filtering step of filtering the multi-valued digital image data in a vertical direction, and horizontally converting the multi-valued digital image data filtered in the vertical filtering step. The multi-value generated through a horizontal filter step of filtering in the direction and a binarization step of converting the multi-value digital data filtered in the horizontal filter step into binary data based on a predetermined reference value A code for calculating feature data of a pixel whose binary data is 1 based on the digital data and the binary data; and determining whether the inspection object is a non-defective product based on the feature data. A computer readable memory comprising process code.