JP3871333B2 - Foreign object detection method, foreign object detection program, and foreign object detection device - Google Patents

Description

  The present invention relates to a foreign object detection method, a foreign object detection program, and a foreign object detection apparatus that detect the presence or absence of a mixed foreign object by processing an X-ray image of an inspection object.

  In the production line for foods and various industrial products, foreign matter detection using an X-ray image has been conventionally performed in order to eliminate foreign matters. For example, Patent Literature 1 and Patent Literature 2 describe conventional techniques in detail.

  FIG. 2 is a flowchart for explaining a typical conventional foreign object detection method. In step S30, a foreign object detection algorithm using an image filter that emphasizes the likelihood of a foreign object is applied to the X-ray image input in step S10, and a threshold value process is applied to the foreign object detection result in step S40 to detect the presence or absence of mixed foreign objects. It is carried out.

  Various image filters can be applied to the image filter that emphasizes the appearance of a foreign object. However, a differential image filter using a small kernel is often used for detecting a fine line foreign object. For example, in the X-ray image example of FIG. 4, assuming a kernel area of 5 × 5 pixels centering on each pixel, the difference between the kernel center density and the average density in the kernel area is evaluated, A difference image is generated with the pixel density of the pixel corresponding to the pixel at the center of the kernel indicating how much higher the density is than the average density.

  Usually, the threshold image is immediately applied to the difference image to detect the presence or absence of mixed foreign matter. However, when detecting a thin line foreign matter having a low contrast, the threshold value processing using labeling is performed by lowering the threshold value than usual. Do. At this time, the threshold processing is applied after averaging the density in the kernel region of 3 × 3 pixels centering on each pixel of the difference image so that the thin line blob is not unnecessarily divided by labeling. ing. FIG. 5 shows the averaged difference image, and FIG. 6 shows the result of threshold processing using labeling in a white display of the X-ray image.

  In the threshold processing using the labeling, a blob having a small pixel area is erased as noise. However, in FIG. 6, many of the edges of the inspection object are detected together with the thin line foreign matter. Therefore, the threshold value must be increased. Accordingly, in consideration of the fact that the difference image emphasizes the likelihood of a foreign object, the blob pixel area is changed to delete a blob having a small pixel volume instead of the blob pixel area. The result shown in FIG. Is obtained. However, in FIG. 7, a large inspection object edge is detected together with the thin line foreign matter, and the threshold value must be increased in order to prevent erroneous determination.

WO98 / 11456 gazette Japanese Patent Laid-Open No. 2001-307069

  In the conventional foreign matter detection method, it is difficult to distinguish a low-contrast foreign matter from the edge of the inspection object, and the detection sensitivity of the fine line foreign matter is low. An object of the present invention is to provide a foreign matter detection method, a foreign matter detection program, and a foreign matter detection device that have high detection sensitivity for fine line foreign matter.

In order to solve the above-mentioned problem, the foreign matter detection method according to claim 1 is a foreign matter detection method for processing an X-ray image transmitted through an inspection object to detect the presence or absence of mixed foreign matter in the inspection object. A step (S2) of generating a shaded edge enhanced image by enhancing the shaded edge of the X-ray image, and a mask pattern for masking a linear image generated at the edge portion of the inspection object from the X-ray image. A step (S3) of generating, a foreign object candidate is extracted from the grayscale edge enhanced image using a first threshold value set for extracting the foreign object candidate, and the position of the foreign object candidate is not masked by the mask pattern adding the concentration of foreign matter candidate for, for a pixel being masked obtains a value obtained by subtracting the concentration of foreign matter candidate evaluation volume for determining the foreign matter resistance, the evaluation volume and the foreign matter-size It compares the second threshold value set for use is constructed from a step (S4) for judging the foreign matter resistance, is performed whether detection of contaminating foreign matter in the object to be inspected.

The foreign matter detection program according to claim 2 is a foreign matter detection program that causes a computer to process an X-ray image transmitted through an object to be detected to detect the presence or absence of mixed foreign matter in the object. A gray-scale edge enhanced image is generated from a line image, a mask pattern for masking a linear image generated at an edge portion of the inspection object is generated from the X-ray image, and foreign object candidates are extracted from the gray-scale edge enhanced image. A foreign substance candidate is extracted using the set first threshold, the density of the foreign substance candidate is added to a pixel whose position is not masked by the mask pattern, and the masked pixel is a value obtained by subtracting the concentration of foreign matter candidates allowed determined as an evaluation volume for determining the foreign matter resistance, a second threshold value set for determining the evaluation volume and foreign matter To perform the determination of the foreign matter resistance by comparison.

Further, the foreign object detection apparatus according to claim 3 receives the X-ray irradiation means (2) for irradiating the inspection object (3) with X-rays and the X-rays transmitted through the inspection object to form a digital image. X-ray detector (4), image input means (5) for taking a digital image output from the X-ray detector by logarithmically converting it into an X-ray image, and an X-ray image taken by the image input means In the foreign matter detection apparatus having image processing means (6) for detecting the presence or absence of foreign matter mixed in the object to be inspected by the image processing, the image processing means emphasizes the shading edge of the X-ray image to Gray level edge enhancement means (61) for generating an edge enhancement image; and mask pattern generation means (65) for generating a mask pattern for masking a linear image generated at the edge portion of the inspection object from the X-ray image. , From the shade-enhanced image First threshold setting means (62) in which a first threshold value that is a predetermined value for extracting an object candidate is set, and a second threshold value in which a second threshold value that is a predetermined value for foreign object determination is set A foreign substance candidate is extracted from the gray level edge-enhanced image using the first threshold value with a setting means (63), and the density of the foreign substance candidate is set for a pixel whose position of the foreign substance candidate is not masked by the mask pattern. For the masked pixel, a value obtained by subtracting the density of the foreign substance candidate is obtained as an evaluation volume for determining the foreign substance property, and the evaluation volume is compared with the second threshold value. And foreign matter determination means (64) for determining the presence or absence of foreign matter.

  According to the present invention, the low-contrast foreign material and the inspection object edge can be distinguished, and the detection sensitivity of the fine line-shaped foreign material is remarkably improved.

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

  FIG. 1 is a flowchart for explaining the foreign object detection method of the present invention. With respect to the X-ray image input in step S1, the shading edge of the X-ray image is emphasized in step S2, a mask pattern for preventing erroneous determination is generated from the X-ray image in step S3, and the first threshold value is determined in step S4. The evaluation volume of the foreign substance candidate extracted by the process is obtained by addition / subtraction with reference to the mask pattern, and the foreign substance property is evaluated by the second threshold process to determine whether the foreign substance is a foreign object.

  For example, when the example of the X-ray image of FIG. 4 is input in step S1, the shaded edge emphasized image is as shown in FIG. 5, the mask pattern is as shown in FIG. 8, and the determination result is as shown in FIG. .

In the following explanation, moving average, maximum value, and minimum value filters using a kernel of N × N pixels are respectively Ave (N), Max (N), and Min (N).
The image filter that takes the difference from the moving average Ave (N) of the kernel is expressed as ADBIN (N), and the frequency of the density histogram in the moving kernel for the high density part [B, 255] HADBIN is an image filter that takes the difference in the density on the high density side where it is 1 / (A + 2) or less of the peak
Indicated as (N, A, B). Also, binarization by threshold C is expressed as BIN (C), density inversion of the image is expressed as INV (image), and the logical product of the two images is expressed as AND (image 1, image 2). A series of image processing is indicated by “→ Process 1 → Process 2 → ...”.

  The method for emphasizing the light and dark edges of the X-ray image in step S2 in FIG. 1 is based on the X-ray image → ADBIN (5) → Ave (3). The grayscale edge enhanced image generated here is subjected to threshold processing in step S4. Although the steps of setting the first threshold value (= 2) and the second threshold value (= 30) to be set in advance are not shown in FIG. 1, the average density of the grayscale edge enhanced image is not shown for the first threshold value. It is also possible to set automatically using.

The mask pattern in step S3 in FIG. 1 is generated as follows. First, the density ibottom of the bottom between the first peak and the second peak is obtained from the density histogram of the image obtained by X-ray image → Ave (3), and X-ray image → BIN (ibottom) → Min (27) → Pattern 1 is a binary image generated by Max (21). Next, a binary image generated by AND (pattern 1, X-ray image) → HADBIN ( N, A, ibottom) → BIN (1) is set as pattern 2. Next, a binary image generated by AND (X-ray image, INV (pattern 2)) → ADBIN (17) → BIN (5) → Min (3) → Max (3) is set as pattern 3. Finally, AND (AND (Pattern 1, INV (Pattern 2)), INV (Pattern 3))
The binary image generated by the above is used as a mask pattern. In this mask pattern, as shown in FIG. 8, a black mask is generated at the edge portion such as the outer edge of the object to be measured.

In step S4 in FIG. 1, foreign matter determination is performed as follows. First, a high density blob is extracted from the grayscale edge enhanced image generated in step S2 with a first threshold value to be a foreign substance candidate. Next, the evaluation volume of each foreign matter candidate is calculated while referring to the mask pattern generated in step S3 at the same position as the foreign matter candidate. That is, the evaluation volume is calculated for each foreign substance candidate by adding the density of the foreign substance candidate for a pixel whose reference mask pattern is not black and subtracting the density of the foreign substance candidate for a pixel whose reference mask pattern is black. Finally, a foreign object candidate whose evaluation volume is larger than the second threshold is determined as a foreign object. FIG. 9 shows the determination result as a white display of the X-ray image.

  As described above, in the foreign matter detection method according to the present invention, the foreign matter property of the high density blob in the grayscale edge enhanced image is evaluated by the evaluation volume based on the mask pattern. Since this low-contrast foreign object can be distinguished from the inspection object edge by referring to the mask pattern, erroneous determination can be prevented without increasing the threshold as in the prior art. As a result, it is possible to remarkably improve the detection sensitivity of the thin line foreign matter compared to the conventional case. Note that the gray edge enhancement method and the mask pattern generation method described here are specific but only examples.

  FIG. 3 is a block diagram for explaining a foreign matter detection apparatus using X-rays, which is an embodiment of the foreign matter detection apparatus of the present invention. That is, a conveying means 1 that conveys the object to be measured 3, an X-ray source 2 that irradiates the object to be measured 3 with X-rays, an X-ray detector 4 that digitalizes an X-ray transmission image of the object to be measured 3, This is a foreign object detection device including an image input means 5 that takes a logarithmic conversion of this into an X-ray image, and an image processing means 6 that processes the X-ray image and determines the presence or absence of mixed foreign substances. Image display means 7 may be added as necessary.

  The transport means 1 is realized by, for example, a belt conveyor that transmits X-rays well, and transports the object to be measured 3 between the X-ray source 2 and the X-ray detector 4 that are arranged to face each other. The X-rays irradiated from the X-ray source 2 are absorbed by the object to be measured 3 and slightly absorbed by the belt conveyor, pass through them, and then reach the X-ray detector 4.

  The X-ray detector 4 is realized by an X-ray line sensor, for example, and converts the X-ray transmission image of the inspection object 3 into a digital image. This digital image is sampled in the transport direction at a sampling pitch substantially equal to the sampling pitch on one line by the X-ray line sensor, and is taken into the memory as an X-ray transmission image.

  Assuming that the X-ray absorption rate of the substance is α and the thickness of the substance is L, the intensity S ′ after the X-ray of the intensity S is transmitted through the substance is theoretically S ′ = S · exp (−α · L ). If transformation is performed by taking the logarithm of both sides, it can be written as α · L = log (S) −log (S ′). The X-ray transmission image corresponds to a two-dimensional distribution of S ′, and can be converted into an X-ray absorption image showing a two-dimensional distribution of the absorption amount α · L due to a substance by taking a logarithm as described above. Both the X-ray transmission image and the X-ray absorption image include physical property information of the object to be inspected, that is, the amount of absorption α · L due to the substance. This is advantageous for emphasizing and detecting foreign matter having a high X-ray absorption rate. In this case, for example, in an X-ray absorption image of an object to be inspected, it is possible to evaluate the likelihood of a foreign object by treating a part having a high density locally or a sharp edge part as a foreign substance candidate point.

  The image input means 5 logarithmically converts the X-ray transmission image into an X-ray absorption image to form an X-ray image and outputs the X-ray image to the image processing means 6. In this logarithmic conversion, correction depending on the physical properties of the inspection object and the X-ray wavelength may be added.

  The image processing means 6 is a CPU or the like having a parameter setting function, and is implemented with the foreign matter detection method of the present invention described with reference to FIG. 1, and receives an X-ray image output from the image input means 5 in advance. Presence / absence of mixed foreign matter is detected by image processing with parameters set. That is, with respect to the X-ray image input in step S1 of FIG. 1, gray level edge enhancement means 61 for generating an averaged difference image from the X-ray image in step S2 and first threshold value setting means 62 for foreign substance candidate extraction. The second threshold value setting means 63 for foreign substance determination, the mask pattern generation means 65 for generating a mask pattern for preventing erroneous determination in step S3, and the evaluation volume of the foreign substance candidate extracted by the first threshold process as a mask. It consists of the foreign matter determination means 64 in step S4 which is obtained by addition / subtraction with reference to the pattern and evaluates the foreign matter by the second threshold processing. When the image display unit 7 is added, the X-ray image input in step S1 is output to the image display unit 7 together with the foreign substance determination result in step S4.

  The image display means 7 generates a foreign substance image from the foreign substance determination result in step S4 of the image processing means 6, and OR-operates the X-ray image input in step S1 of the image processing means 6 to display it on a CRT or the like. .

It is a flowchart for demonstrating the foreign material detection method of this invention. It is a flowchart for demonstrating the conventional foreign material detection method. It is a figure for demonstrating embodiment of the foreign material detection apparatus of this invention. It is an example of an X-ray image. It is an example of a light / dark edge enhancement image generated from an example of an X-ray image. It is the image which showed the result of having threshold-processed the high density blob of the example of a grayscale edge emphasis image with an area. It is the image which showed the result of carrying out the threshold value process of the high density blob of the example of a grayscale edge emphasis image. It is the image which showed the mask pattern produced | generated by the method by this invention from the X-ray image example. It is the image which showed the result of having performed the threshold value process by the evaluation volume by this invention about the high density blob of the example of a grayscale edge emphasis image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Conveyance means, 2 ... X-ray source, 3 ... Object to be measured, 4 ... X-ray detector, 5 ... Image input means, 6 ... Image processing means, 7 ... Image display means, 61 ... Light / dark edge emphasis means, 62 ... 1st threshold value setting means, 63 ... 2nd threshold value setting means, 64 ... Foreign matter determination means, 65 ... Mask pattern generation means

Claims (3)

In the foreign object detection method for processing the X-ray image transmitted through the inspection object and detecting the presence or absence of the mixed foreign material in the inspection object,
Generating a gray edge-enhanced image by emphasizing the gray edges of the X-ray image (S2);
Generating a mask pattern for masking a linear image generated at an edge portion of the inspection object from the X-ray image (S3);
A foreign substance candidate is extracted from the grayscale edge-enhanced image using a first threshold value set for foreign substance candidate extraction, and the density of the foreign substance candidate is set for a pixel whose position of the foreign substance candidate is not masked by the mask pattern. For the masked pixel, a value obtained by subtracting the density of the foreign substance candidate is obtained as an evaluation volume for judging the foreign substance , and the evaluation volume and the second set for the foreign substance judgment are obtained. A foreign matter detection method comprising: a step (S4) of determining foreign matter by comparing with a threshold value. In a foreign object detection program for causing a computer to process an X-ray image transmitted through an inspection object and detect the presence or absence of a foreign object in the inspection object,
Generating a gray edge-enhanced image from the X-ray image;
Generating a mask pattern for masking a linear image generated at an edge portion of the inspection object from the X-ray image;
A foreign substance candidate is extracted from the grayscale edge enhanced image using a first threshold set for foreign substance candidate extraction, and the density of the foreign substance candidate is set for a pixel whose position is not masked by the mask pattern. For the masked pixel, a value obtained by subtracting the density of the foreign substance candidate is obtained as an evaluation volume for determining the foreign substance property , and the evaluation volume and the second set for the foreign substance determination are set. A foreign matter detection program for determining foreign matter by comparing with a threshold value. X-ray irradiation means (2) for irradiating the inspection object (3) with X-rays, an X-ray detector (4) for receiving X-rays transmitted through the inspection object and converting them into digital images, and the X-ray detection Input means (5) for logarithmically converting a digital image output from the instrument into an X-ray image, and foreign matter mixed in the object to be inspected by image processing the X-ray image acquired by the image input means In the foreign object detection device having image processing means (6) for detecting the presence or absence of
Wherein the image processing means,
A light / dark edge emphasizing means (61) for enhancing a light / dark edge of the X-ray image to generate a light / dark edge enhanced image;
Mask pattern generation means (65) for generating a mask pattern for masking a linear image generated at an edge portion of the inspection object from the X-ray image;
A first threshold value setting means (62) in which a first threshold value that is a predetermined value for extracting a foreign substance candidate from the grayscale edge enhanced image is set;
A second threshold value setting means (63) in which a second threshold value that is a predetermined value for foreign object determination is set;
A foreign substance candidate is extracted from the grayscale edge-enhanced image using the first threshold value, and pixels of which the position of the foreign substance candidate is not masked by the mask pattern are added to the density of the foreign substance candidate and masked. For the pixel, a value obtained by subtracting the density of the foreign substance candidate is obtained as an evaluation volume for determining the foreign substance property , and the evaluation volume is compared with the second threshold value to determine the presence or absence of the foreign substance. A foreign matter detection device comprising foreign matter determination means (64).