JP2508662Y2 - Flaw detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention relates to a flaw detection device that detects flaws without erroneous detection even if there is uneven density in the target area.

[Prior Art] FIG. 4 shows a configuration example of a conventional flaw detection device. In FIG. 4, (1) is an input unit for inputting an image, (2) is a low-pass filter for cutting high-frequency components of image data,
(3) is a difference calculation unit that calculates the difference between the image data of the input unit of (1) and the image data that has passed through the low-pass filter of (2), and (7) is the difference data obtained in (3) A defect detection unit that detects a defect by comparing with (6) is a result output unit that outputs a result.

Next, FIG. 5 shows an example of processing by the conventional flaw detection device.
In FIG. 5, the solid line in (a) shows the signal waveform of the input image input through the input section in (1). The broken line in the figure is (2)
3 shows a signal waveform after passing through the low pass filter of FIG. FIG. 5 (b) shows difference data obtained by subtracting the result of the low-pass filter from the input data in the difference calculation unit. Further, the flaw detection unit compares the individual difference data with the flaw judgment value Th, and marks the portion where the difference value exceeds the judgment value as a flaw. As described above, in the conventional method, when there is a gradual density fluctuation in the background, the defect signal is extracted by estimating the base metal part with a low-pass filter and taking the difference from the input signal to detect the defect.

[Problems to be Solved by the Invention] FIG. 6 shows another example of processing by a conventional device. Similar to FIGS. 6 (a) and 6 (b), the solid line shows the input data and the broken line shows the output of the low-pass filter, but unlike FIG. 5, there is no flaw, but the signal level near the center drops greatly. . This is due to differences in surface properties and the like.
If the conventional processing is performed on such a signal as it is, a signal at a non-defective portion as shown in FIG.

As described above, in the conventional method, when the concentration of the base metal part suddenly drops, the normal part may be erroneously detected as a flaw. The present invention was made to solve such a problem. The purpose of this is to prevent the occurrence of erroneous detection by determining whether or not there is a true defect from the density distribution around the defect candidate after detecting the defect candidate.

[Means for Solving the Problems] A flaw detection apparatus according to the present invention is configured to determine the amount of change in the density of a fixed pixel on both sides from the apex of the difference signal of the flaw candidate in order to determine which of the flaw candidates is not a flaw. And a determination unit for determining whether or not there is a defect based on the result.

In addition, the flaw detection device according to the present invention compares the drop in the density in the vicinity of the flaw candidate with the density average at a slight distance in order to determine the non-defect among the flaw candidates, and based on this result, It is provided with a determination unit for determining whether or not.

[Operation] Since the flaw detection device according to the present invention determines whether or not the flaw candidate is a true flaw based on the density distribution pattern around the flaw candidate, there is a concentration drop that induces an erroneous detection in the inspection area. Even if there is, a gradual change in density in the background can be removed and only a flaw can be detected.

[Embodiment] An embodiment of the device according to the present invention will be shown below.

FIG. 1 shows an example of the configuration of the device according to the present invention. In FIG. 1, (1) is an input unit for inputting an image, (2) is a low-pass filter for cutting high-frequency components of image data,
(3) is a difference calculation unit that calculates the difference between the image data of the input unit (1) and the image data that has passed through the low-pass filter of (2), and (4) is the difference data obtained in (3) A defect candidate detection unit that detects a defect candidate by comparing with
(5) is a defect candidate detection unit that determines whether or not the defect candidate is an actual defect from the density information around the defect candidate obtained in (4), and (6) is a result output unit that outputs the result. .

Next, the operation will be described. First, from (1) image input to (3) difference calculation is the same as the conventional method, as shown in FIG. The processing content of the defect candidate detection of (4) is similar to that of the conventional defect detection, and is also as shown in FIG. The processing of the flaw determination unit (5) will be described with reference to FIG.

In FIG. 2, (a) schematically shows an actual flaw, and (b) shows an example of a density distribution in which a change portion having a large density is detected as a flaw candidate. In the figure, P is the peak position of the difference value when detected as a defect candidate. Also A
L and AR are density average values for each of the left and right i pixels from the point P. Further, DL and DR are the differences between the density value at point P and AL and AR.
As shown in the figure, generally, flaws have a concentration distribution as shown in (a), and DL and DR are almost equal. Further, in (b), the shape of the concentration distribution is greatly different on the left and right of the point P, and there is a difference between DL and DR. Therefore, If it satisfies the condition, it is judged as a defect, and otherwise it is rejected.

Next, FIG. 3 shows a flaw determination method based on another embodiment. Fig. 3 is an example of the case where uneven density occurs in a patchy manner, and since there is a drop in density at two locations, the density distribution is such that the normal part between them is like a flaw. In Fig. 3, P is the peak position of the difference value when it is detected as a defect candidate, BL and BR are the minimum density values in the left and right i pixels from the P point, and AL and AR are the density values of the outer j pixels. It is an average value. Furthermore, DL and DR are the differences between AL and BL, and AR and BR, respectively. As shown in Fig. 2, if there is a flaw, the density is high only at the flaw and the surrounding area is uniform. Therefore, DL ≤ Dset DR ≤ Dset Dset is a flaw only when both of the set values are satisfied, and Reject other than.

As described above, according to the present invention, the defect candidates are first picked up from the difference signal level, and then the defect candidates other than the defect are rejected according to the density distribution shape of the periphery, so that false detection is reduced. You can

[Advantage of Device] As described above, according to this device, even if there is a large concentration drop in the inspection area, a defect can be detected without erroneous detection.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of a flaw detection device according to the present invention, FIG. 2 is an explanatory view of operation of the embodiment according to the present invention, FIG. 3 is an explanatory view of operation of an embodiment of the present invention, and FIG. Is a block diagram of a conventional device, and FIGS. 5 and 6 are explanatory views of the operation of the conventional device. In the figure, (1) is an input unit, (2) is a low-pass filter, (3) is a difference calculation unit, (4) is a defect candidate detection unit, (5) is a defect determination unit, (6) is a result output unit, (7)
Is a flaw detection unit. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (2)

(57) [Scope of utility model registration request] 1. An input unit for inputting an image, a low-pass filter for filtering input image data in a scanning line direction, a difference calculation unit for calculating a difference between input data and an output of the low-pass filter, and a result of the difference. A defect candidate detection unit that detects a defect candidate, and a defect determination unit that determines whether or not there is a defect based on the difference in the density change rates of certain pixels on both sides from the apex of the difference signal of the defect candidate obtained by this defect candidate detection unit. And a defect output device for outputting the above determination result. 2. A utility model registration claim (1) characterized in that, in the defect determination unit, it is determined whether or not there is a defect by comparing the average density of the vicinity of the defect candidate and the density average of a place slightly apart.
The flaw detection device according to the item.