JP2007101320A - Processor and method for processing ultrasonic flaw detection image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a processor and a method for processing an ultrasonic flaw detection image, capable of obtaining an ultrasonic flaw detection inspection result of stable quality. <P>SOLUTION: The processor 1 transmits an ultrasonic wave toward an inspected object, and receives an echo image corresponding an echo reflected by a reflection source inside the inspected object to detect a defect, the processor has an image preparation part 3 for converting the received echo image into an image data, a continuous image display part 5 for displaying continuously a plurality of images generated in the image preparation part 3, an image area information processing part 6 for limiting a pixel resolution and a display area of the image and for preparing a limited area image, an image-to-image motion tracking processing part 7 for tracking the moving of a pixel serving as a defect candidate, using the successive front and rear side images in the images, and a defect position determination part 8 for extracting the defect from the continuity of the plurality of defect candidates to determine a defect position thereof. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an ultrasonic flaw detection image processing apparatus and an ultrasonic flaw detection image processing method for detecting a defect from an image acquired by an ultrasonic flaw detection apparatus.

  Ultrasonic flaw detection (UT) has been widely performed as a means for inspecting welds of structures in plants and the like. It is an inspector who evaluates the ultrasonic flaw detection data obtained by the ultrasonic flaw detection inspection to determine whether or not there is a defect in the welded portion of the structure, and evaluates it from expertise and experience. Therefore, considerable skill is required, and the evaluation results may vary depending on the inspector, and further objectivity of evaluation has been demanded.

  Therefore, an ultrasonic flaw detection data evaluation apparatus that evaluates defects using a neural network based on an image obtained by ultrasonic flaw inspection has been proposed (see Patent Document 1).

In addition, an image processing method for an ultrasonic reflection source is proposed in which the position of the reflection source in the inspection object is calculated from the elapsed time from the transmission to reception of the ultrasonic wave using the longitudinal and transverse sound velocities. (See Patent Document 2).
Japanese Patent Laid-Open No. 9-210970 JP-A-9-292371

  In the method using the neural network described in Patent Document 1, there is a problem that the inspection quality may not be uniform because each ultrasonic flaw detection data evaluation apparatus has characteristics depending on the learning order and data.

  Further, there is a problem that it is a very difficult task to evaluate a large amount of inspection data with an ultrasonic flaw detection data evaluation apparatus.

  In the image processing method of the ultrasonic reflection source using the elapsed time from the transmission to the reception of the ultrasonic wave described in Patent Document 2, there is a problem that only the reflection source at the same position or a close position can be imaged. It was.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an ultrasonic flaw detection image processing method and an ultrasonic flaw detection image processing apparatus that can obtain ultrasonic flaw detection test results with stable quality.

  In order to solve the above-described problem, in the ultrasonic flaw detection image processing apparatus according to the present invention, as described in claim 1, ultrasonic waves are transmitted to an inspection object and reflected by a reflection source in the inspection object. An ultrasonic flaw detection image processing apparatus that receives an echo image corresponding to the received echo and detects a defect, an image creation unit that A / D converts the received echo image into image data, and a plurality of images created by the image creation unit A continuous image display unit that continuously displays the image, an image region information processing unit that creates a limited region image by limiting the pixel resolution and display region of the image, and images before and after the image The image processing apparatus includes an inter-image movement tracking processing unit that tracks the movement of a pixel that is a defect candidate, and a defect position determination unit that extracts a defect from the continuity of a plurality of defect candidates and determines the defect position.

  According to a fourth aspect of the present invention, an ultrasonic wave is transmitted to the inspection object, an echo image corresponding to an echo reflected by a reflection source in the inspection object is received, and the received echo image is converted into image data. A / D conversion is performed for continuous display, a limited area image is created by limiting the pixel resolution and display area of the image, and pixels that are defect candidates are moved using images before and after the image is continuous. , The defect is extracted from the continuity of a plurality of defect candidates, and the defect position is determined.

  Furthermore, as described in claim 6, ultrasonic flaw detection is performed by transmitting an ultrasonic wave to an inspection object and receiving an echo image corresponding to an echo reflected by a reflection source in the inspection object to detect a defect. The image processing apparatus is characterized in that a defect extraction unit is provided for obtaining an illuminance distribution of the reflected wave of the ultrasonic wave from the received echo image and determining a defect based on the illuminance distribution.

  Then, as described in claim 8, an ultrasonic wave is transmitted to the inspection object, an echo image corresponding to an echo reflected by a reflection source in the inspection object is received, and an ultrasonic wave is received from the received echo image. An illuminance distribution of the reflected wave is obtained, and a defect is determined based on the illuminance distribution.

  According to the ultrasonic flaw detection image processing apparatus and the ultrasonic flaw detection image processing method according to the present invention, it is possible to make the inspection quality constant without depending on the experience and level of the inspector during the ultrasonic flaw inspection. Become.

  An embodiment of an ultrasonic flaw detection image processing apparatus according to the present invention will be described with reference to the accompanying drawings.

[First Embodiment]
1 to 5 are views showing a first embodiment of an ultrasonic flaw detection image processing apparatus according to the present invention.

  As shown in FIG. 1, the ultrasonic flaw detection image processing apparatus 1 according to the first embodiment receives ultrasonic reflected waves (echoes) that are transmitted to an inspection object and reflected by a reflection source of the inspection object. An ultrasonic flaw detector 2 that performs an image creation unit 3 that digitizes the received echo image and converts it into image data to create an image, an image storage unit 4 that stores the created image, a series of image data, and the like, A continuous image display unit 5 that continuously displays a plurality of images created from the image creation unit 3, an image region information processing unit 6 that creates a limited region image by limiting the pixel resolution and region of the image, and An inter-image movement tracking processing unit 7 that tracks the movement of pixels that are defect candidates using previous and subsequent images, and a defect position determination unit that finally extracts defects from the continuity of a plurality of defect candidates and determines a defect position 8.

  Further, the ultrasonic flaw detection image processing apparatus 1 uses a history management unit 9 that manages flaw detection result history such as defect position data extracted by the defect position determination unit 8 and an image acquired as a digital image as necessary. An automatic report creation unit 10 that automatically creates a report, and an automatic processing unit 11 that continuously processes a plurality of data.

  Furthermore, the ultrasonic flaw detection image processing apparatus 1 includes an evaluation method learning unit 12 that displays an evaluation method as needed and prompts the inspector to evaluate.

  The ultrasonic flaw detector 2 transmits ultrasonic waves to the inspection object and receives reflected waves from the ultrasonic reflection source, and performs nondestructive inspection on the presence / absence, dimensions, etc. of the inspection object such as steel materials, An apparatus mainly composed of an ultrasonic flaw detector and an ultrasonic probe, and is generally commercially available.

  The image creation unit 3 performs A / D conversion on the analog signal raw data received from the ultrasonic flaw detection procedure 2 into digital signal image data, and transmits the image data to the continuous image display unit 5.

  The image storage unit 4 stores data such as image data created by the image creation unit 3 as necessary. The stored data can be retrieved and referenced whenever necessary.

  The continuous image display unit 5 has a function of continuously displaying a plurality of images received from the image creation unit 3, and creates moving image data from still image data.

  When the image area information processing unit 6 creates a moving image using the continuous image display unit 5, the position of the crack is obtained by acquiring the appearance position in the image of the defect (crack) to be detected and the information of the pixel resolution. The pixel resolution is converted to the actual size.

  The inter-image movement tracking processing unit 7 performs processing for calculating the position of the defect such as a crack and the pixel resolution by the image region information processing unit 6, or in parallel, The movement and the movement of the pixel set are traced, and defect candidates are extracted and determined.

  The defect position determination unit 8 finally determines a defect position from a plurality of defect candidates using the image set. As a method for creating a pixel set, a commonly used labeling process is used, but if necessary, the result of comparing the target pixel and its surrounding pixels is treated as a set of vectors by embedding the target pixel in the target pixel. Perform vector image processing. Note that the labeling process is a processing method in which connected pixels are recognized as one block and distinguished from other backgrounds.

  The length of the movement tracking result by the inter-image movement tracking processing unit 7 (continuity of the region of the focused reflection source) is adjusted by moving the probe of the ultrasonic flaw detector 2 manually or at a constant speed. It is possible to adjust the detection accuracy.

  The history management unit 9 stores the defect data extracted by the defect position determination unit 8 as a flaw detection result history, and manages the defect data so that it can be taken out when necessary. By attaching the inspection date and time to the defect data, it is possible to observe and infer the progress of the defect according to the time series. In addition, it is possible to determine the replacement time of the inspection object by observing according to the time series.

  The automatic report creation unit 10 automatically creates report data by acquiring image data stored in the image storage unit 4 and associating it with previously stored document data. Further, the automatic report creation unit 10 stores report data in which image data and document data are associated with each other.

  The automatic processing unit 11 urges the above-described image creation unit 3 to automatic report creation unit 10 to sequentially perform the processes performed by the ultrasonic flaw detector 2 sequentially on the echo images. . Therefore, the automatic processing unit 11 can analyze a large amount of data with a constant inspection quality, and can process a large amount of data without degrading the inspection quality. In addition, the use of the automatic processing unit 11 eliminates the need for human processing, so that continuous processing can be performed for 24 hours.

  The evaluation method learning unit 12 displays evaluation method data indicating the evaluation method stored in advance for the inspector, and prompts the inspector to evaluate. When the frequency of inspection decreases, the level of the inspector can be maintained, and the inspection quality can be maintained.

  As an example of the result of the flaw detection inspection using the ultrasonic flaw detection image processing apparatus 1, still images at three time points among the continuous images in which the defect candidates D to G move in the left direction are shown in FIGS. It was shown to.

  According to FIG. 2, defect candidate D1 that appears to be a defect candidate D in area A, defect candidates E1 and F1 that appear as defect candidates E and F in area B, and defect candidate G1 that appears to be defect candidate G in area C It is displayed. According to FIG. 3, defect candidate D2 that appears to be defect candidate D is displayed in area A, and defect candidates E2 and F2 that are considered to be defect candidate E are displayed in area B, but defect candidates are displayed in area C. Absent. Next, according to FIG. 4, defect candidates E3 and F3 considered to be defect candidates E and F are displayed in area B, and defect candidate G3 considered to be defect candidate G is displayed in area C. No defect candidates are displayed.

  Here, FIG. 5 shows the feature areas of the defect candidates shown in FIGS. 2 to 4 superimposed on one screen. As can be seen from FIG. 5, the defect candidate D3 that appears to be the defect candidate D is not displayed in the area A, and the defect candidate G2 that appears to be the defect candidate G is not displayed in the area C.

  In the ultrasonic flaw detection image processing apparatus 1, the difference between the image obtained from the defect and the image obtained from the noise and other material non-uniform portions was evaluated by the continuity. Since the defect candidates E and F in the area B are continuously displayed in FIGS. 2 to 4, the defect candidates E and F in the area B are finally determined as cracks from the three areas A to C. Extracted.

  According to the ultrasonic flaw detection image processing apparatus 1 of the first embodiment, by evaluating defect candidates with continuous images, there is no difference in evaluation due to different inspectors, and the quality of inspection without depending on the experience and level of the inspector. Can be made constant.

  In addition, according to the ultrasonic flaw detection image processing apparatus 1 of the first embodiment, a large amount of data can be smoothly evaluated.

[Second Embodiment]
Next, a second embodiment according to the present invention will be described with reference to FIGS. In addition, the same code | symbol is attached | subjected to the structure same as 1st Embodiment, and the overlapping description is abbreviate | omitted.

  As shown in FIG. 6, the ultrasonic flaw detection image processing apparatus 1 </ b> A according to the second embodiment includes an ultrasonic flaw detection apparatus 2 that transmits an ultrasonic wave and receives a reflected wave (echo) of the ultrasonic wave. On the other hand, it includes a defect extraction unit 13 that determines a defect by using a property that hits with a minute spread, and an image storage unit 4 that stores an image, a series of data, and the like.

  In addition, the ultrasonic flaw detection image processing apparatus 1A automatically creates a report as necessary using a history management unit 9 that manages flaw detection result histories such as determined defect data and images acquired as digital images. And an automatic processing unit 11 that continuously processes a plurality of data.

  Due to the performance of the ultrasonic flaw detector 2, the ultrasonic wave transmitted by the ultrasonic flaw detector 2 has a property of hitting the defect with a minute spread. When an ultrasonic wave is transmitted to an inspection object and the reflection intensity of the ultrasonic wave hitting the defect perpendicularly is 100, the reflection intensity of the ultrasonic wave not hitting the vertical object becomes non-perfectly elastic as it becomes non-perpendicular. It becomes 90, 80, and so on. In addition, it is considered that the decrease in the reflection intensity includes factors such as influence (diffraction) due to the wraparound of ultrasonic waves.

  7 and 8 are diagrams illustrating the illuminance distribution of the reflected wave of the ultrasonic wave with the ordinate indicating the illuminance and the abscissa indicating the position. When the reflection intensity is converted into luminance in consideration of a decrease in the reflection intensity of the ultrasonic wave, a luminance distribution as shown in FIG. 7 is obtained if there are minute cracks. On the other hand, the luminance distribution as shown in FIG.

  The defect extraction unit 13 uses the fact that the spread of the luminance distribution shown in FIG. 7 is caused by the performance of the ultrasonic flaw detector 1A, and uses a half width of the histogram of the illuminance distribution as a pair of defect candidates. And the positions of the defect opening and defect end in the pipe UT inspection are determined.

  According to the ultrasonic flaw detection image processing apparatus 1A of the second embodiment, by obtaining the illuminance distribution from the ultrasonic reflection intensity, the difference in evaluation due to the difference between the inspectors is eliminated, so that the quality of the inspection can be made constant. .

  In addition, according to the ultrasonic flaw detection image processing apparatus 1A of the second embodiment, a large amount of data can be smoothly evaluated.

1 is a configuration diagram showing a first embodiment of an ultrasonic flaw detection image processing apparatus according to the present invention. FIG. The figure which shows the image of B scope in piping. The figure which shows the image of B scope in piping. The figure which shows the image of B scope in piping. FIG. 5 is a diagram in which FIGS. 2 to 4 are superimposed and displayed on one screen. The block diagram which shows 2nd Embodiment of the ultrasonic flaw detection image processing apparatus which concerns on this invention. The figure which shows the luminance distribution of the echo which shows a crack. The figure which shows the luminance distribution of the echo which shows the disturbance by a material etc.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 1A Ultrasonic flaw detection image processing apparatus 2 Ultrasonic flaw detection apparatus 3 Image creation part 4 Image storage part 5 Continuous image display part 6 Image area information processing part 7 Inter-image movement tracking processing part 8 Defect position determination part 9 History management part 10 Automatic report creation unit 11 Automatic processing unit 12 Evaluation method learning unit 13 Defect extraction unit D1, D2 Defect candidates E1-E3 Defect candidates F1-F3 Defect candidates G1, G3 Defect candidates

Claims (12)

In the ultrasonic flaw detection image processing apparatus for detecting a defect by transmitting an ultrasonic wave to an inspection object, receiving an echo image corresponding to an echo reflected by a reflection source in the inspection object,
An image creation unit for A / D converting the received echo image into image data;
A continuous image display unit for continuously displaying a plurality of images generated by the image creation unit;
An image area information processing unit for creating a limited area image by limiting the pixel resolution and display area of the image;
An inter-image movement tracking processing unit that tracks the movement of pixels that are defect candidates using images before and after the image; and
An ultrasonic flaw detection image processing apparatus comprising: a defect position determination unit that extracts a defect from the continuity of a plurality of defect candidates and determines the defect position. The ultrasonic flaw detection image processing apparatus according to claim 1, further comprising an image storage unit that stores image data created by the image creation unit. The ultrasonic flaw detection image processing apparatus according to claim 1, further comprising a defect extraction unit that obtains an illuminance distribution of an ultrasonic reflected wave from the received echo image and determines a defect based on the illuminance distribution. An ultrasonic wave is transmitted to the inspection object, an echo image corresponding to the echo reflected by the reflection source in the inspection object is received, and the received echo image is A / D converted into image data and continuously displayed. Creating a limited area image by limiting the pixel resolution and display area of the image,
Ultrasonic flaw detection characterized in that the movement of pixels as defect candidates is tracked using images before and after the image is continuous, defects are extracted from the continuity of a plurality of defect candidates, and the defect positions are determined. Image processing method. The ultrasonic flaw detection image processing method according to claim 4, wherein an illuminance distribution of a reflected wave of ultrasonic waves is obtained from the received echo image, and a defect is determined based on the illuminance distribution. In the ultrasonic flaw detection image processing apparatus for detecting a defect by transmitting an ultrasonic wave to an inspection object, receiving an echo image corresponding to an echo reflected by a reflection source in the inspection object,
An ultrasonic flaw detection image processing apparatus, comprising: a defect extraction unit that obtains an illuminance distribution of a reflected wave of an ultrasonic wave from a received echo image and determines a defect based on the illuminance distribution. The ultrasonic flaw detection image processing apparatus according to claim 6, further comprising an image storage unit that stores data of illuminance distribution obtained by the defect extraction unit. The ultrasonic wave is transmitted to the inspection object, an echo image corresponding to the echo reflected by the reflection source in the inspection object is received, the illuminance distribution of the reflected wave of the ultrasonic wave is obtained from the received echo image, and this An ultrasonic flaw detection image processing method, wherein a defect is determined based on an illuminance distribution. 8. The ultrasonic flaw detection according to claim 1, further comprising a history management unit that stores defect data obtained by ultrasonic flaw detection as a flaw detection result history and manages the data so that it can be taken out when necessary. Image processing device. 10. The automatic report creation unit for automatically creating report data in which the image data created by the image creation unit is associated with document data stored in advance is provided. The ultrasonic flaw detection image processing apparatus according to any one of the above. The ultrasonic flaw detection image processing apparatus according to any one of 1 to 3, 6, 7, 9, and 10 provided with an automatic processing unit that prompts to continuously process a plurality of echo images.   The ultrasonic flaw detection image processing apparatus according to any one of claims 1 to 3, 6, 7, and 9 to 11, further comprising an evaluation method learning unit that displays evaluation method data indicating an evaluation method stored in advance.

Images