JP4956077B2 - Defect inspection apparatus and defect inspection method - Google Patents

Description

  The present invention relates to a defect inspection apparatus, and more particularly, to a defect inspection apparatus and a defect inspection method capable of easily determining setting of appropriate detection conditions.

  2. Description of the Related Art Conventionally, there is known a defect inspection apparatus that detects various types of defects such as foreign matter, dirt, or wrinkles on an inspection target and determines the quality of the target. For example, when the inspection object is in the form of a sheet such as a film or a metal plate, the inspection object is run in the longitudinal direction, the surface of the traveling inspection object is scanned by the line sensor, and the line sensor outputs Defect detection is performed by processing the signal. A technique is also used in which the detected defects are classified based on various feature quantities and the defect type is determined.

A defect inspection apparatus as such an example is disclosed in Patent Document 1. The defect inspection apparatus according to Patent Document 1 detects a surface of a sheet-like object as a detection image, extracts a plurality of feature amounts from the detection image, and determines a defect type based on a defect type definition for each defect type. It is. The “defect” in Patent Document 1 corresponds to a “defect”.
JP 2004-109069 A

However, in the conventional defect inspection apparatus shown in Patent Document 1, it is difficult to determine a defect detection condition for properly detecting a defect and determining the detected defect as an appropriate defect type. There is.
For example, the defect inspection apparatus inspects based on the set defect detection conditions and extracts defects, but there may be no problem with some extracted defects. This may depend on, for example, a difference or deviation between a case where a person or a person who performs an inspected inspection performs an inspection and an inspection standard in a defect inspection apparatus.

In addition, when an inspection is performed for the first time, a predetermined detection condition may not be appropriate. Conventionally, in such a case, several inspections are performed, and appropriate detection conditions are determined by the inspections. However, in order to perform several inspections, it is necessary to run the inspection object in the longitudinal direction for each inspection. If the same inspection object is inspected many times, the inspection will cause the inspection, There was a problem that defects such as scratches occurred on the object itself.
For example, when the inspection object is a prototype that is produced only once and the inspection conditions for inspecting the prototype are determined, the problem that it is difficult to determine the detection conditions as described above becomes significant. .

  In addition, in the conventional defect inspection apparatus disclosed in Patent Document 1, when a plurality of defect images are included in the detected image, it is not specified which defect image is to be measured. There was a problem that it was difficult to determine properly.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a defect inspection apparatus and a defect inspection method capable of easily determining the setting of appropriate detection conditions.

The present invention has been made to solve the problems described above, an imaging means for generating image data by imaging the inspection object, a threshold value of the generated image data and the first detection condition of said image pickup means The first defect detection means for executing the first defect detection process based on the first defect detection means and detecting the center position of the defect detected by the first defect detection process, and the detection of the first defect detection means An image extraction means for extracting a defect inclusion area image, which is an image of a predetermined area centered on the center position of the defect, from the image data generated by the imaging means, and a defect inclusion area image extracted by the image extraction means A defect image area that is an area in the defect inclusion area image in the image and that is surrounded by the defect center of the defect detected by the first defect detection means according to the Ferre diameter of the defect. 2 Have a second defect detecting means performs the detection process of the second defect based on a threshold as a detection condition, the second threshold value as a detection condition, the threshold as the first detection condition Is a defect inspection apparatus characterized in that it is a value far from the value of the normal part .

This invention is the first defect detection means detects a plurality of the defect image region, said second defect detecting means, independently on the second for each of a plurality of the defect image region a defect inspection apparatus and the client performs detection processing of the second defect based on detection condition.

This invention includes an imaging means for generating image data by imaging the inspection object, the detection processing of the first defect based on the threshold value as a generated image data and the first detection condition of said image pickup means A first defect detection unit that executes and detects a center position of the defect detected by the first defect detection process; and a predetermined center around the center position of the defect detected by the first defect detection unit. An image extraction unit that extracts a defect-included region image that is an image of the selected region from image data generated by the imaging unit, a second detection condition setting unit that sets a second detection condition, and the image extraction unit A region in the defect inclusion region image in the defect inclusion region image that is surrounded by the center of the defect detected by the first defect detection means according to the ferret diameter of the defect. Defective image area A second defect detecting means performs the detection process of the second defect based on a threshold as a second detection condition set by the second detection condition setting means Te, by the second defect detecting means possess an output means for outputting a result of the detection processing of the second defect, the detection condition reflecting means for setting the second detection condition to the first detection condition in the first defect detecting means, the ,
The defect inspection apparatus is characterized in that the threshold value as the second detection condition is a value farther from the value of the normal part than the threshold value as the first detection condition .

This invention is a defect inspection apparatus you characterized by having repeating unit, repeating the process with the second detection condition setting means and said second defect detecting means and the output means.

This invention generates image data by imaging the inspection object, and performs the detection process of the first defect on the basis of the threshold value as the image data and the first detection condition that the product, said first Detecting the center position of the defect detected by the defect detection process, and extracting a defect inclusion region image that is an image of a predetermined region centered on the detected defect center position from the generated image data; A defect image area that is an area in the extracted defect inclusion area image and that is surrounded by the defect diameter of the defect centered on the center position of the defect detected by the first defect detection process . The second defect detection process is executed based on the threshold value as the second detection condition , and the threshold value as the second detection condition is farther from the normal part value than the threshold value as the first detection condition it is a value A defect inspection method characterized by the features.

  According to the present invention, after performing the defect detection process on the inspection object, it is possible to execute the defect detection again under different conditions, thereby making it possible to easily set an appropriate detection condition. Play.

  Further, according to the present invention, the second defect detection means measures a defect even if a plurality of defect images exist in the detected image because there is a defect to be measured at the center position of the defect inclusion region image. There is an effect that it is possible to extract the feature amount of only the defective defect.

  In addition, the second defect detection means performs a defect detection process only on the defect image area as the center position of the defect detected by the first defect detection means, so that a plurality of defect images are included in the detected image. Even if it exists, the feature amount of only the defect to be measured can be measured by referring to the minimum pixel.

  Further, the second defect inspection means picks up the inspection object again by changing the repetitive conditions and reviewing the measurement value based on the defect inclusion area image detected and stored by the first defect detection means. The effect that an appropriate detection condition can be set without doing is produced.

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration of a defect inspection apparatus according to an embodiment of the present invention.
In the present embodiment, a sheet-like inspection object that is running is imaged, and defect detection processing is executed based on the captured image. Therefore, as shown in FIG. 1, the defect inspection apparatus according to the present embodiment captures an image of an inspection object, generates image data, and a defect under a first detection condition based on the image data. A defect centered on the first defect detection means 2 that executes the detection process and the center position of the detected defect (for example, information on the position specified by the horizontal axis value cx and the vertical axis value cy). First, a defect detection process is executed under a second detection condition different from the first defect detection means 2 based on the image extraction means 3 for extracting the inclusion area image and the image extracted by the image extraction means 3. 2 defect detection means 4.

The imaging means 1 used in the present embodiment has a line illumination device and a line sensor, images a traveling inspection object, and generates image data of the inspection object. Further, the generated image data is output to the first defect detection means 2. The captured image is, for example, an image in which the long direction of the inspection object is the vertical axis (y axis) and the direction orthogonal to the long direction is the horizontal axis (x axis).
The line illumination device irradiates the inspection object with light from a predetermined direction, and the line sensor images the inspection object illuminated by the line illumination device with a predetermined number of elements to generate an image. . Moreover, the imaging means 1 has a frame memory for storing a running image. For example, if a 128 MB frame memory is used, a 64K line scan image can be temporarily stored in the case of a line sensor with 2048 elements. When the number of scanned images exceeds the number of scans due to the capacity of the frame memory, the old scanned images are sequentially deleted, and new scanned images are stored in the frame memory in order. The line illumination device and the line sensor are attached so that the longitudinal direction thereof is parallel to the direction orthogonal to the longitudinal direction of the inspection object.

  For example, as the linear illumination device in the image pickup means 1, quartz rod illumination with an illumination length of 500 to 2200 mm can be used, and as the line sensor, a line sensor having 2048 elements, 5000 elements, or 7450 elements can be used. . Each model and number of units may be arbitrarily selected depending on the width of the inspection target, the traveling speed, and the resolution.

  Next, the first defect detection unit 2 preprocesses the image data input from the imaging unit 1 based on a predetermined first detection condition, and binarizes the image data with a preset threshold value. After data compression, connectivity processing is performed to detect defects, and the defect image area specified by the detected defect center position (hereinafter referred to as defect image center position information), defect area, and defect ferret diameter (Hereinafter referred to as defect image area information). The first defect detection means 2 executes a classification process for the detected defect based on the defect classification condition which is one of the first detection conditions from the measured value of the detected defect.

Note that the defect image area specified by the defect ferret diameter will be described later in detail with reference to FIG.
The center position of the detected defect is, for example, the center position of the defect image area specified by the defect diameter of the defect, and is designated by a value cx on the horizontal axis and a value cy on the vertical axis. This is location information.

The predetermined first detection conditions include preprocessing conditions, detection conditions, and defect classification conditions. For example, the pre-processing conditions for the first detection condition determined in advance include threshold information when image binarization is performed, and the detection conditions include area, width from the binarized image. , Information such as a threshold for measuring a measured value such as length.
For example, when the area of an image that is a candidate for a certain defect is larger than a predetermined area A, it is detected as a defect, and when it is smaller, it is not detected as a defect. The detection condition is information on a value as such a predetermined threshold value.
Also, the defect classification condition of the first detection condition determined in advance is, for example, when classification is performed, when the width of the defect is larger than the first value and smaller than the second value. The defect classification is executed as A or the like. In this case, the information is a condition for classification such as the first value and the second value of the defect width.

In the present embodiment, for example, in the binarization processing of an image, a threshold for binarization is based on a ratio (%) with respect to the average of the pixel level in the reference area with the image input as a reference. Set. Next, the binarization processing of the image is executed by detecting whether or not an arbitrary pixel is larger than the threshold value. Note that the method of determining the threshold value is not limited to this, and is arbitrary.
Next, the first defect detection unit 2 outputs the detected defect image center position information and defect image region information to the image extraction unit 3.

The image extracting unit 3 captures defect inclusion region image information having a predetermined number of pixel sizes in the horizontal and vertical directions around the position of the defect image center position information from the first defect detecting unit 2. 1 is extracted from the image data picked up by 1 and stored in the frame memory.
For example, as for the size of the number of pixels in the horizontal direction and the vertical direction of the defect inclusion region image information, the number of pixels in the horizontal direction is 256 pixels, and the number of pixels in the vertical direction is 256 pixels. Note that the number of pixels in the horizontal direction and the vertical direction of the defect inclusion region image information is not limited to the number of 256 pixels, and may be an arbitrary number of pixels.

The image extraction unit 3 includes an extracted image storage unit 301. The defect inclusion region image information extracted by the image extraction unit 3, the defect image center position information from the first defect detection unit 2, and the defect image region. In association with the information, the image extraction unit 3 records the defect image in the extracted image storage unit 301 for each defect image.
The image extracting unit 3 is described as having the extracted image storage unit 301. However, the present invention is not limited to this, and the image extracting unit 3 and the extracted image storage unit 301 may be configured as different devices.
In the present embodiment, a control PC (personal computer) may be used as the image extraction means 3.

The second defect detection means 4 measures the area specified by the defect image area information of the defect inclusion area image information extracted by the image extraction means 3 based on the second detection condition, and the area, width, length Measure the measured value. Further, the second defect detecting means 4 executes defect classification from the measured value according to the defect classification condition based on the second detection condition.
The second detection conditions are preprocessing conditions, detection conditions, and defect classification conditions, and are the same conditions as the first detection conditions. Normally, the value of the second detection condition is different from that of the first detection condition.
The second defect detection unit 4 may directly acquire the defect inclusion region image information, the defect image center position information, and the defect image region information from the image extraction unit 3, or the extracted image of the image extraction unit 3. You may acquire from the memory | storage means 301. FIG.

The second defect detection unit 4 includes a second detection condition setting unit 401 for setting a second detection condition. The user can set the second detection condition using the second detection condition setting unit 401.
Further, the second defect detection means 4 has a detection confirmation means 402 for displaying or outputting the measurement result of the defect detected under the second detection condition and the result of classification of the detected defect. The user can confirm the measurement result of the defect detected under the second detection condition set by the second defect detection means 4 and the result of classification of the detected defect using the detection confirmation means 402.

The user who uses this apparatus uses the second detection condition setting means 401 and the detection confirmation means 402 of the second defect detection means 4 so that a specific defect becomes a desired measurement result or classification result. The second detection condition appropriate for detecting the defect can be determined by setting again.
Further, in order to use the second detection condition setting unit 401 and the detection confirmation unit 402 in order to determine an appropriate second detection condition, that is, to execute detection condition determination using the measured image, There is no need to take an image of the inspection object again, or to run the inspection object again for imaging.

The second defect detection means 4 can be executed in units of defect images, but can also be in units of inspection. In the case of the inspection unit, a plurality of images measured in the inspection unit can be automatically measured.
Note that the term “defect image unit” here means that detection processing is executed on one selected defect image based on the second detection condition, and the inspection unit is stored in the extracted image storage unit 301. For all the defect images, the detection process is executed based on the second detection condition.
In the present embodiment, an operation PC may be used as the second defect detection means 4.

FIG. 2 is a diagram for explaining a defect image as an example. In FIG. 2, the defect inclusion region image D is an image in which the horizontal pixels on the horizontal axis (x axis) are 256 pixels and the vertical pixels on the vertical axis (y axis) are 256 pixels.
In the center of the defect inclusion area image D, the defect image A1 detected by the first defect detection means 2 is displayed. The center position of the defect image A1 (that is, defect image center position information) is specified by a value cx on the horizontal axis of the center position and a value cy on the vertical axis of the center position.
The ferret diameter is the length of the covered area on the horizontal axis and the vertical axis of the rectangular area covering the defect image A1, as shown for the defect image A1 in FIG. The length of the ferret diameter on the horizontal axis is fx, and the length of the ferret diameter on the vertical axis is fy. The centers of the ferret diameters fx and fy are the positions (cx and cy) of the defect image center position information.

The defect image area F is an area surrounded by the ferret diameter fx and the ferret diameter fy. The defect image area information is information for specifying the defect image area F, and is information on values of the ferret diameter fx and the ferret diameter fy.
In the extracted image storage unit 301, the defect inclusion area image D information, defect image center position information (cx, cy), and defect image area information (fx, fy) as described in FIG. Each image is stored in association with each other.

  The defect image area F is not limited to the area based on the defect image area information (fx, fy) centered on the defect image center position information (cx, cy), for example, as shown in FIG. As described above, the position information fx1 and fx2 at both ends of the ferret diameter fx of the defect image region F and the position information fy1 and fy2 at both ends of the ferret diameter fy may be specified.

  Further, the defect inclusion region image D may include another defect image B different from the defect image A1, but because it is outside the range of the region by the defect image region information about the defect image A1, It is easy to distinguish from what should be detected by the second defect detection means. In addition, since the second defect detection unit 4 executes the detection process only within the range of the area based on the defect image area information, the detection process of the second defect detection unit 4 is not affected by the defect image B.

As shown in FIG. 2, for example, the defect image detected by the second defect detection unit 4 with respect to the defect image A1 detected by the first defect detection unit 2 is the defect image A2.
Although the defect is detected for the same image, the first detection condition in the first defect detection unit 2 and the second detection condition in the second defect detection unit 4 are different. It is different like A1 and defect image A2.

  In general, the second detection condition in the second defect detection unit 4 is made sweeter than the first detection condition in the first defect detection unit 2. That is, as many defects as possible are extracted under the first detection condition. Therefore, for example, as a threshold value for executing binarization as a defect detection condition, a ratio with respect to a normal part (normal part is assumed to be 100%) is set. Next, in the second detection condition, a value farther from the normal part than the first detection condition is set. By doing in this way, a defect detected under the first detection condition is detected as a defect and a certain defect is detected as not a defect under the second detection condition. The same applies to detection conditions other than the conditions for performing binarization. Due to the relationship between the second detection condition and the first detection condition as described above, the defect image A2 is usually smaller than the defect image A1, and the defect image A2 is included in the area of the defect image A1.

Next, the operation of the embodiment of the present invention will be described using the flowchart of FIG.
First, in the image input process (step S1), an image is captured from the line sensor with a preset resolution to generate an image.
Next, in the binarization 1 process (step S2), the input image is binarized with a preset threshold value. Next, in the feature amount extraction 1 process (step S3), after data compression of the binarized image is performed, connectivity processing is performed, and defect image center position information (cx, cy) that is information on the center position of the defect ), Area, defect image area information (fx, fy) and other feature quantities are measured. Next, in the defect classification 1 process (step S4), defect classification is executed according to preset defect classification conditions based on the measured feature values.

Next, in the image extraction process (step S5), a defect inclusion region image is extracted with the position of the defect image center position information (cx, cy) as the center.
Next, in binarization part 2 (step S6), the image extracted is binarized with a different threshold based on a second detection condition different from binarization 1 (step S2). Execute. Next, in the feature amount extraction process 2 (step S7), after the data compression of the binarized image, the connectivity process is performed, and the center position (cx, cy), area, width, length, etc. of the defect Measure the feature amount. Next, in the defect classification 2 process (step S8), defect classification is executed according to preset defect classification conditions in the same manner as defect classification 1 (step S4) based on the measured feature values.

  Note that the processing from step S2 to step S4 executes processing based on the first detection condition, and the processing from step S6 to step S8 executes processing based on the second detection condition.

As a processing procedure, the following two methods are possible. Method 1 performs processing from image input processing (step S1) to image extraction processing (step S5), and detection of binarization 2 processing (step S6) to defect classification 2 processing (step S2). This is a method for processing one image.
In this case, a process for setting the second detection condition (step S100) is added between step S5 and step S6, and a process for outputting the result (step 101) is added after step S8. By repeating between step 100 and step 101, the user can set an appropriate second detection condition.

Method 2 performs image input processing (step S1) to image extraction processing (step S5), and binarization 2 processing (step S6) to defect classification 2 processing (step S8). This is a method for collectively detecting a plurality of defect images detected in (1).
In method 2, steps S6 to S8 may be collectively processed for a plurality of defect images based on the appropriate second detection condition obtained in method 1.

Next, Example 1 of the embodiment according to the present invention will be described. Example 1 is an example of method 1.
Here, for example, the inspection object has a width of 1600 mm and travels at 30 m / min, and the image is input with a resolution of 0.07 × 0.07 mm / pixel. The line illumination device uses one quartz rod illumination with an illumination length of 2000 mm. As the line sensors, five line sensors having a performance of 5000 elements, a drive frequency of 40 MHz, and a maximum scanning cycle of 0.131 ms are used.

FIG. 4A is a diagram illustrating a defect inclusion region image of a defect detected by the second defect detection unit in the first embodiment. In the first embodiment, a binarization threshold is set as the second detection condition. For example, the threshold value is set based on the ratio with respect to the threshold value setting area H that is a part of the defect inclusion area image D.
FIG. 4A shows the defect image area F (Ferret diameter fx × fy) detected by the first defect detection means and the defect image A2 detected by the second defect detection means.

FIG. 4B shows the measured value K1 of the defect detected by the first defect detection means, the area is 2.04 mm square, the width is 2.00 mm, and the length is 2.80 mm. It is shown that.
FIG. 4C shows the measured value K2 of the defect detected by the second defect detecting means, the area is 1.56 mm square, the width is 1.40 mm, and the length is 2.20 mm. It is shown that.
Here, only the area, width, and length are exemplified as the defect measurement values K1 and K2 as an example, but other feature values may be included.
By inputting the above threshold value, that is, by setting the second detection condition and again detecting and measuring the second defect detection means 4, the measurement result can be obtained as many times as the defect measurement value K2. It is possible to confirm, and it is possible to obtain an appropriate second detection condition.
As described above, the second detection condition in which the measurement value is appropriate can be easily obtained by the method in the first embodiment, for example, the binarization threshold condition.

Next, Example 2 of the embodiment according to the present invention will be described. Example 2 is an example of method 2. In the second embodiment, the defect image information detected by the first defect detection unit and stored in the extracted image storage unit 301 is used as a target, and based on the second detection condition different from the first inspection condition, The defect is detected by the defect detection means 2.
Here, the first detection condition is set as a strict condition, and the defect is set to be excessively detected. Next, as the second detection condition, a value of a condition that is considered appropriate in the first embodiment is used.
Further, all the defect image information stored in the extracted image storage unit 301 may be detected and measured under the second detection condition.
By doing like Example 2, it becomes possible to detect an appropriate defect only by imaging an inspection object once under an appropriate second detection condition. Such method 2 is effective when the vehicle can be run and imaged only once as in a prototype.

Next, Example 3 of the embodiment according to the present invention will be described. Example 3 is an example of a defect that cannot be detected by the first defect detection means, and is an example in which a mark is used.
For example, for a defect to be detected in an inspection target, the user puts a mark in advance near the defect to be detected in the inspection target before the inspection. Thereafter, an inspection is performed, and the mark is detected as a defect by the detection means. At this time, the defect to be detected may not be detected.

Next, since the defect to be detected is imaged together with the mark detected as a defect, the mark is first detected, and the defect to be detected near the detected mark is specified. Next, conditions for detecting a defect to be detected are determined.
The method described in the third embodiment is a method for determining a detection condition that can detect a defect to be detected in advance using a mark as described above.

FIG. 5 is a diagram for explaining a method of detecting a mark by the first defect detection means and setting a target defect detection condition.
In the case of a defect image T that cannot be detected by the first defect detection means, the mark M is detected as a defect and a defect inclusion region image is extracted. The mark M is displayed at the center of the defect inclusion region image D.
Second defect detection means and measurement area selection means for selecting a measurement area in the defect inclusion area image D are provided. Using this measurement area selection means, the defect image T to be detected is surrounded by a measurement area F2 which is a rectangular area.
Next, the measurement area F2 around the defect image T is measured by the second defect detection means.
The user can obtain an appropriate second detection condition for detecting the defect image T by changing the second detection condition and measuring the measurement region F2.

  As described above in detail, according to the defect inspection apparatus and the defect inspection method of the present invention, it is possible to execute defect detection again under different conditions after performing the defect detection process on the inspection object. This makes it easy to set appropriate detection conditions.

In the operation of the inspection apparatus, an appropriate detection condition obtained by the second defect detection means 4 is set as the first detection condition in the first defect detection means 2, and subsequent inspection objects are imaged. Inspection can be performed using the means 1, the first defect detection means 2, and the image extraction means 3.
In order to set the appropriate detection condition (the second detection condition) obtained by the second defect detection means 4 as the first detection condition in the first defect detection means 2, the second defect detection means 4 In addition, a detection condition reflecting unit that sets the second detection condition as the first detection condition of the first defect detection unit 2 may be provided.

In this case, for example, a process for setting the second detection condition (step S100) is added between step S5 and step S6 in FIG. 3, and the result is output after step S8 (step S100). 101), and repeating between step 100 and step 101, the user can set an appropriate second detection condition. However, in the process of outputting the result of step 101, the user When it is confirmed that the detection condition is appropriate by looking at the output result, for example, the processing of the detection condition reflection means may be executed by pressing an operation button that reflects the detection condition.
In order to do this, the detection condition reflecting unit may include an input detecting unit that receives an input from the user and detects that the button is pressed.
The operation button is a button that is displayed and operated on the output means, and may be a button that is operated and pressed with a mouse or a keyboard.

  In the embodiment, the defect image area F is specified by the ferret diameter fx and the ferret diameter fy of the defect image. However, when the image filter condition is changed, the second measurement value is larger. Therefore, it is necessary to set the ferret diameter to the ferret diameter fx + αx and the ferret diameter fy + αy. Here, αx and αy are values larger than 0, respectively.

  Each of the first defect detection means 2, the image extraction means 3, and the second defect detection means 4 may be realized by dedicated hardware, or the first defect detection means 2 Each of the image extraction means 3 and the second defect detection means 4 is constituted by a memory and a CPU (central processing unit), and the first defect detection means 2, the image extraction means 3 and the second defect detection means 4 The function may be realized by loading a program for realizing each function into a memory and executing the program.

  The extracted image storage unit 301 includes a nonvolatile memory such as a hard disk device, a magneto-optical disk device, or a flash memory, a volatile memory such as a RAM (Random Access Memory), or a combination thereof. And

  Further, it is assumed that an input device, a display device, and the like (none of them are shown) are connected to the second defect detection means 4 as peripheral devices. Here, the input device refers to an input device such as a keyboard and a mouse. The display device refers to a CRT (Cathode Ray Tube), a liquid crystal display device, or the like.

  In the description of the embodiment, the first detection condition is described as a predetermined value. However, the first detection condition setting in which the first defect detection unit 2 sets the first detection condition. A first detection condition may be set by the first detection condition setting means.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

  The present invention is suitable for use in a defect inspection apparatus.

It is a block diagram which shows the structure of the defect inspection apparatus by one Embodiment of this invention. It is a figure which shows the defect inclusion area | region image and feature-value of the defect detected by the 1st defect detection means. It is a flowchart which shows the process sequence of this invention. It is explanatory drawing which shows the example of a detection as an example detected by the 2nd defect detection means. It is explanatory drawing for demonstrating the method to detect a mark with the 1st defect detection means, and to set the detection conditions of the target defect.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Imaging means 2 1st defect detection means 3 Image extraction means 4 2nd defect detection means D Defect inclusion area | region image A1 Defect image A2 Defect image B Defect image F Defect image area fx Ferre diameter fy Ferre diameter H Threshold setting area K1 Measurement value K2 at the first defect detection means Measurement value M at the second defect detection means M Mark T Defect image F2 Measurement region

Claims (5)

Imaging means for imaging the inspection object and generating image data;
A first defect detection process is executed based on the image data generated by the imaging means and a threshold value as a first detection condition , and a center position of the defect detected by the first defect detection process is detected. First defect detection means;
Image extraction means for extracting a defect inclusion area image, which is an image of a predetermined area centered on the center position of the defect detected by the first defect detection means, from the image data generated by the imaging means;
A defect in the defect inclusion region image extracted by the image extraction unit, the region being centered on the center position of the defect detected by the first defect detection unit and surrounded by the defect diameter of the defect Second defect detection means for executing a second defect detection process based on a threshold value as a second detection condition for the image region;
I have a,
The defect inspection apparatus , wherein the threshold value as the second detection condition is a value farther from the value of the normal part than the threshold value as the first detection condition .   2. The second defect detection means executes the second defect detection process independently for each of a plurality of the defect image areas based on the second detection condition. The defect inspection apparatus described in 1. Imaging means for imaging the inspection object and generating image data;
A first defect detection process is executed based on the image data generated by the imaging means and a threshold value as a first detection condition , and a center position of the defect detected by the first defect detection process is detected. First defect detection means;
Image extraction means for extracting a defect inclusion area image, which is an image of a predetermined area centered on the center position of the defect detected by the first defect detection means, from the image data generated by the imaging means;
Second detection condition setting means for setting a second detection condition;
A defect in the defect inclusion region image extracted by the image extraction unit, the region being centered on the center position of the defect detected by the first defect detection unit and surrounded by the defect diameter of the defect Second defect detection means for executing a second defect detection process based on a threshold value as a second detection condition set by the second detection condition setting means for the image region;
An output means for outputting a result of the detection process of the second defect by the second defect detection means;
Detection condition reflecting means for setting the second detection condition as the first detection condition in the first defect detection means;
I have a,
The defect inspection apparatus , wherein the threshold value as the second detection condition is a value farther from the value of the normal part than the threshold value as the first detection condition . Repetitive means for repeating the processing of the second detection condition setting means, the second defect detection means, and the output means;
The defect inspection apparatus according to claim 3, further comprising: Imaging the inspection object to generate image data,
Performing a first defect detection process based on the generated image data and a threshold value as a first detection condition, detecting a center position of the defect detected by the first defect detection process;
Extracting from the generated image data a defect inclusion region image that is an image of a predetermined region centered on the center position of the detected defect,
A defect image area that is an area in the extracted defect inclusion area image and that is surrounded by the defect diameter of the defect centered on the center position of the defect detected by the first defect detection process . The second defect detection process is executed based on the threshold value as the detection condition of 2 ,
A defect inspection method , wherein the threshold value as the second detection condition is a value farther from the value of the normal part than the threshold value as the first detection condition .

Images