JP3181543B2 - Deterioration / corrosion detection method for surface-treated steel - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting and judging the deterioration or corrosion of a surface-treated steel material for detecting and judging a corrosion state or a deterioration state of an external surface of a member constituting a large structure such as a steel tower. It is.

[0002]

2. Description of the Related Art In a steel tower or the like, if the outer surface of a steel material subjected to hot-dip galvanizing or zinc spraying is used for a long period of time, the outer surface of the steel material may be deteriorated due to the surrounding environment, weather conditions such as ozone and rainwater, or aging. Is eroded, and corrosion progresses as the steel material surface is exposed. At present, the deterioration state and the corrosion state of steel materials are periodically inspected by humans, and the progress of corrosion is visually inspected.

[0003]

However, in the conventional manual inspection, the evaluation results vary depending on the inspector, making it difficult to make a quantitative judgment. It was difficult to manage the plan.
The labor required for these inspections is also considerable.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to automatically detect a corrosion state of an external surface of a member constituting a structure and to externally detect the external surface. Utilizing the characteristics of the rust color, it is possible to reliably determine the degree of corrosion or the deterioration level of the corroded part, and to easily know the progress of the corroded part from the color of these corroded parts. Deterioration / corrosion detection / determination method .

[0005]

In order to achieve the above object, the present inventors have filed Japanese Patent Application No. Hei 8-4100.
The technology used in No. 5 “Method of detecting internal corrosion of pipes and its internal corrosion detection system device” was further developed, and a new technology was developed. In the method of detecting internal corrosion of pipes, a video detector (a lamp, a CCD camera, a video image scope method using an electric signal cable, or a light source using an optical fiber) is provided by providing an illumination unit and an imaging unit on the head of a long image transmission unit. Guide, image guide, etc.) is inserted into the body to be detected, and the image captured by the imaging means is directly processed by a computer having an image processing function to detect and determine a corroded portion, or The corroded part is recorded on a recording medium (video tape or magneto-optical disk, etc.) while detecting and determining the corroded part using a computer. Regarding the image processing, the inner surface of the pipe is imaged by a color camera or the like, and the color image is subjected to image processing. That is, the computer detects a corroded portion using the difference in color between the corroded portion and the other portions in the detected image. Further, the corroded portion is determined using three values of lightness, saturation, and hue of the color. In addition, regarding the detection and determination of a corroded portion, an image of lightness, saturation, or hue is divided into blocks, and a representative value (mode value) of each block is obtained. This representative value is within a predetermined set value range. Block is the corroded part.

On the other hand, in the present invention, since the corrosion state on the outer surface of a member constituting a structure such as a steel tower is targeted, an image detection unit for photographing the inner surface of the pipe (a lighting unit is provided on the head). Elongated video transmission means is not necessary, and the equipment is simple just by directly installing a camera, and various photographing methods can be used. Although the above-described image processing can be used for the image processing, special processing needs to be performed on the outer surface of the component because the corrosion state is different.

According to a first aspect of the present invention, there is provided a deterioration / corrosion detection / judgment method, wherein an image pickup means (a video camera, a photo CD, etc.) capable of picking up an image of an external surface of a member constituting a structure is provided along a member of the structure. It is installed on a moving robot or flying machine (such as a helicopter) or installed at a distance, and the whole or part of the structure is photographed with this imaging means.
A color image is input to a computer having an image processing function, and the image processing is used to extract an inspection object and to obtain three values of lightness, saturation, and hue obtained from a color original image.
At least the brightness, saturation and hue of the inspection object from the image of
And obtained by image processing from corrosion test pieces of each deterioration level.
Deterioration degree standard consisting of at least brightness, saturation and hue
Values, at least lightness, saturation, and hue of the inspection object.
Comparing, you and judging the degree of corrosion or degradation level of the test object exterior surface.

[0008] A color image photographed by the image pickup means.
Can be processed as follows. That is,
Color video on a recording medium (video tape or magneto-optical
Disc, etc.) and play this recorded video
Input to a computer with image processing function. Profit
Computer with image processing function for color images
To process directly. Image processing machine
Input to a computer equipped with
Extraction of the inspection target, and decay of this inspection target
Detects the state of corrosion and determines whether the external surface of the inspection
The recording medium (video tape or
Recorded on a magneto-optical disk.

In extracting an inspection object in image processing by a computer, for example, a video obtained by an imaging means is subjected to binarization processing (including noise removal processing and edge extraction), and the binary image is subjected to Hough processing. detecting a straight line by performing the conversion shall be the inspection object areas surrounded by the straight line.

[0010] extracting the inspection object, by utilizing the difference in color in the test object to detect the corrosion state, determine corrosion rate or the deterioration level. That is, using the three colors of lightness, saturation, and hue, to detect corrosion state, determine corrosion rate or the deterioration level. For examples of detection and judgment of corrosion state
For example, the image of lightness, saturation, and hue is divided into each block, and a representative value (mode value) of each block is obtained, and a block whose representative value is within a predetermined set value range is determined as a corroded portion. Also,
Calculate the area or area ratio of red rust area and black rust area and
Determine the level. In addition, corrosion test specimens of each deterioration level
At least lightness, saturation and hue by image processing using
Reference values (brightness, saturation, hue, rust area, rust
Area ratio) is determined in advance, and the inspection target detected
At least lightness, saturation, and hue (of lightness, saturation, hue
Rust area, rust area ratio, etc.)
Comparison of the reference values for the degree of corrosion
Determines the degradation level.

In the above-described configuration, the entire structure is photographed by the photographing means or a part of the structure is photographed by sampling, and the corrosion state is automatically detected by a computer having an image processing function. The color of hot-dip galvanizing etc. on the outer surface of the members constituting the structure is basically achromatic, such as gray, and the corroded portions are rusted, that is, chromatic colors such as light brown to dark brown. The corrosion state can be automatically detected, and the degree of corrosion and the deterioration level can be determined.

In addition, by determining the location of corrosion using the lightness, saturation, and hue of the color, the location of corrosion can be determined reliably without erroneous detection. In other words, it is difficult to identify the corroded part because there is a part with low lightness other than the corroded part only by the lightness, but by using the saturation, the chromatic color of the corroded part and the achromatic color of the steel material surface itself are clearly defined. Can be distinguished. Therefore, for regions with a certain brightness or higher,
By performing the determination of the chromatic color and the determination of the rust color based on the hue, it is possible to more accurately detect the corroded area and determine the corrosion degree and the deterioration level.

Further, the zinc covering the outer surface of the steel material gradually becomes thinner due to erosion with time. For this reason, the color of the outer surface of the steel material changes from gray (a color close to achromatic) peculiar to zinc to brown (chromatic color) peculiar to rust, and partly to black due to partial black rust. I do. The brightness, saturation, hue, area,
By detecting and determining the area ratio and the like, the degree of progress of corrosion and the degree of progress of deterioration can be estimated. In addition, the detected lightness, chroma, hue, area, area ratio, etc. of the outer surface of the steel material, and the deterioration level reference values (brightness, chroma, hue, area, area ratio, etc.) that are strictly investigated and correlated with the corrosion status Can be clearly estimated from the cross-sectional state of the steel material, that is, the remaining amount of the plating film thickness.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below based on an embodiment shown in the drawings. This is an example in which the present invention is applied to a case where a corrosion state of an outer surface of a steel material constituting a steel tower is detected and a deterioration state is determined. FIGS. 1 and 3 (a) show partial images of an object to be inspected using a lifting robot which moves up and down along a main pillar of a steel tower. FIGS. 2 and 3 (b) show airplanes such as helicopters. This shows the case where the entire tower is photographed from a machine or a distant high place.

1 and 2, a deterioration / corrosion detection / judgment system 1 of the present invention includes a camera 2 such as a video camera or a digital camera (including an infrared camera or a special camera used for a space satellite). It comprises a video tape recorder (or magneto-optical disk system) 3 to which a video signal from the camera 2 is input, a color monitor 4 and a computer 5. The computer 5 has an image processing board (video input / output function / image processing function) 6 built therein, performs image processing described in detail later, and detects a corroded portion, and determines the degree of corrosion and the deterioration level. .

In FIG. 1, the camera 2 is mounted on the elevating robot 10, and the inspection object is partially photographed. The lifting robot 10 lifts and lowers by a rack and pinion method using the auxiliary equipment 11 for lifting provided along the main pillar of the steel tower. In FIG. 2, the camera 2 is mounted on a helicopter 12, a small airplane, or the like, or the camera 2 is installed and fixed on a nearby mountain 13 or the like, and the entire inspection object is photographed.

The image captured by the camera 2 is output to the color monitor 4 and the computer 5, and the image is displayed on the color monitor 4. At the same time, the computer 5 detects and determines the corroded portion. Is automatically recorded on the video tape recorder 3. In this case, the video tape recorder 3 records only a few seconds before and after the video that is equal to or more than the reference value, and the video that does not cause a problem does not pass through and is not recorded. In addition to being avoided, the work of confirming the detection result can be performed efficiently. In addition, the finally obtained corrosion spots and the judgment results are
Saved in the order of material number, displayed and printed out.

Note that all images captured by the camera 2 may be recorded on the video tape recorder 3, and then the recorded images may be reproduced and the computer 5 may detect and determine the location of corrosion.

In the above description, the image of the corroded portion is once recorded on a video tape, a magneto-optical disk, or the like. However, the image captured by the camera 2 can be taken into the computer 5 and processed on the spot in real time. In this case, the video obtained by the camera 2 is taken into the ground computer 5 via a video transmission cable (can be wireless and non-cable), analyzed by the computer 5, and then analyzed by a magneto-optical disk of a magneto-optical disk system. Record the image of the corroded area and the result of the judgment, and output the form using a printer. Further, the sensor head itself may have a function of making a determination. In this case, it is not necessary to send the judgment to the computer, and the system can be made compact.

Further, a still image (photograph) captured by an ordinary film camera is converted to a computer 5 by an image scanner.
Or a still image captured by a digital camera or the like (CCD camera, etc.) into the computer 5,
The processing can be directly performed by a computer to detect and determine the corroded portion, or the computer can detect and determine the corroded portion and record the corroded portion on a recording medium (video tape, magneto-optical disk, or the like).

Next, the image processing will be described.

(1) Extraction of Inspection Object Only FIG. 4 (a) shows an example of an original image obtained by the camera 2. From this original image, a brightness image shown in FIG. 4 (b) is obtained. (Described later), and a noise removal process
Edge extraction is performed and binarization is performed. For example, a median filter is used for noise removal, and a Laplacian filter is used for edge extraction. Hough transform is performed on the binary image shown in FIG. 4C to detect straight lines. The Hough transform is a method of obtaining a straight line equation from a set of points in image analysis / recognition. An approximate straight line is obtained from a set of low density points of the data of the binary image, and this is set as the outline of the member. A region surrounded by the straight line is defined as a member to be inspected.

(2) Features of the object to be measured (structural member) and the acquired image The structural member has been subjected to hot-dip galvanizing or zinc spraying .... Achromatic (gray) Corrosion area Initial ... red (red rust) Middle ... black (However, the surrounding area is red.) Late stage: hole (however, the surrounding area is black and the outside is red)

(3) Extraction of red rust area as chromatic color 3-1) In order to extract red rust area as chromatic color, lightness (brightness): L (the degree of color contrast) Saturation: C ( Hue: H (red to orange to green to blue) is used.

3-2) The signal output from the camera is converted into a Yuv or RGB signal by analog processing in a normal transmission system. The converted signal is digitized and subjected to orthogonal transformation (RGB → Lab or Yuv → RGB → La
According to b), conversion into Lab color is performed. Note that this La
In addition to the b-color method, the evaluation can be directly performed using a transmission signal such as Yuv. Here, the RGB signals are converted into the Lab model shown in FIG. 5 using the following equations (1) and (2), and the saturation C and the hue H are obtained from the equations (3) and (4). In the case of Yuv, the color model shown in FIG. 6 is used.

[0026]

(Equation 1)

[0027] 3-3) as the feature amount evaluation parameters of erosion area, also to the difference from the respective reference regions, ^{lightness: L *, ΔL * = L} * -L 0 * ...... (5) Saturation: ^{C *, ΔC * = C *} -C 0 * ...... (6) ^{hue: H *, ΔH * = H} * -H 0 * using three sets of six ... (7). Note that L ₀ ^* , C ₀ ^* , H ₀ ^*
Is a reference value.

3-4) Specific Detection Judgment Method (Example) As shown in FIG. 7, the brightness L ^* .color is calculated from the three RGB images of the original image captured by the camera according to equations (1) to (4). An image of degree C ^* and hue H ^* is obtained. The following processing is performed on such an image in order to determine a corroded area that is not easily affected by noise. The image of L ^* , a ^* , b ^* is divided into m × n blocks (pixels may be used).
To find the mode value of the pixels in each block, L
A histogram in each block for each of ^* , a ^* , and b ^* is created (see FIG. 8), and values at which the frequency of the histogram is maximum are set as mode values L ^* _p , a ^* _p , and b ^* _p .
Using these modes L ^* _p , a ^* _p , and b ^* _p , L ^* ,
A C ^* , H ^* corrosion determination map is created (see FIG. 9).
In this corrosion determination map (example), the area between the minimum value and the maximum value of the mode is divided into 16 equal parts and displayed in shades.

3-5) More Accurate Detection of Corroded Area <Setting of Region of Interest Based on Lightness L ^* (Example)> In the corrosion determination map of chroma C ^* , even though lightness L ^* is small, chroma since C ^* is not greater corrosion area may be scattered noise-like becomes white block, attention block for corrosion determined using a threshold L ^* _th lightness as shown in the following equation ( x, y) is restricted to brighter blocks. (x, y): {g (x, y)> L ^* _th } (L ^* _th = 55) (8)

<Corrosion Judgment Based on Saturation C ^* > Since the area of the corrosion area includes a low saturation area and a high saturation area, the cumulative frequency of the right tail in the histogram of the saturation C ^* is S%. the saturation C ^* of the value of time that exceeds C ^* _pS
And a block exceeding the threshold value C ^* _th is determined as a candidate for a corroded area.

3-6) Use of Hue (Example) A hue histogram for each block is obtained using a hue corrosion determination map (see FIG. 8). 0 ° -9
Since there are pixels having a hue H ^{* in} the range of 0 °, 0
A block in which the number of pixels in the range of ° to 90 ° exceeds the threshold value H ^* _th is defined as a corroded portion. The saturation determination of each block is performed based on the saturation histogram of each block, and the above-described hue determination is performed on the blocks that have been determined as corrosion candidate regions in the saturation determination.

(4) Evaluation of Deterioration Degree Using Evaluation Parameters (Example) FIG. 8 is a qualitative examination result of a corrosion test piece cut out from an angle steel of a steel tower.
It was found that there was a difference in color characteristics depending on the deterioration level. In FIG. 8, the deterioration level I is gray in the original color image because the zinc layer remains, and the deterioration level IV is mostly red rust, and the lower part is black rust due to iron erosion. I have. Therefore, the deterioration level IV is higher in the pixels having a larger saturation C ^* than in the deterioration level I, and the hue H
^{In *} , the pixels in the range showing brown (around 90 °) tend to increase, and the lightness L ^* tends to decrease. In addition, in the case of the angle iron of the steel tower, the degradation tends to progress more on the mountain side than on the valley side,
In the width direction, the blade side at the leading end tends to deteriorate.

(5) Example of Deterioration Degree Determination FIG. 9 shows an example in which the deterioration level of the corrosion test piece of FIG. 8 is determined, and shows a corrosion determination map at the degradation levels I and IV. In FIG. 9, (a) is a corrosion determination map of saturation C ^* for red rust, and (b) is a corrosion determination map of lightness L ^* for black rust. Here, in the corrosion determination map of the saturation C ^* , the corroded area is light brown to dark brown (chromatic), and is therefore displayed brightly. Lightness L
^In the corrosion judgment map marked with ^* , both gray (achromatic) and brown (chromatic) are displayed brightly.

At the deterioration level I shown in FIG. 9A, since the zinc layer is entirely gray and corrosion is partially observed at a part of the lower end, the saturation C ^* in FIG. The red rust portion is displayed in white, and the brightness L ^* in (b) is substantially white as a whole because there is no black rust. At the deterioration level IV in FIG. 9 (b), since the whole is in a red rust state and the lower part is in a black rust state, the red rust part is displayed white and wide in the saturation C ^* of (a), and the lower black rust part is It is displayed in black and (b)
At the lightness L ^* , the lower black rust portion is displayed in black.

Therefore, the presence / absence of red rust and black rust, that is, the deterioration level can be determined by the corrosion determination map of the saturation C ^* and the lightness L ^* in FIG. Further, in the hue corrosion determination map, since the red rust portion has pixels having the hue H ^{* in} the range of 0 ° to 90 °, the number of pixels in the range of 0 ° to 90 ° is the threshold value H ^*. Blocks exceeding _th can be used as red rust parts. Further, the areas of the red rust region and the black rust region are obtained, and the deterioration level can be determined from these sizes or the area ratio between the red rust portion and the black rust portion.

(6) Judgment of Deterioration Level Although the case of the corrosion test piece has been described above, the deterioration level is determined by performing the above-described processing on an image of the actual outer surface of the steel material. be able to. Also, based on the results of the corrosion test pieces described above, the deterioration reference degrees (saturation C ^* , lightness L ^* , hue H ^* ,
It is also possible to determine a deterioration level by setting in advance a reference value such as an area or an area ratio) and comparing the reference value with an actual detection result of a steel material.

[0037] (7) determines the degree of deterioration of each member unit from deterioration level repair resulting member, checks whether members deterioration level is distributed to the throat throughout the object. Based on this deterioration level distribution, the degree of deterioration of the entire inspection object is evaluated, and a repair plan is made.

It should be noted that the above description is for detecting corrosion of steel tower components.
Although the determination has been described, it is needless to say that the present invention is not limited to this and can be applied to the detection and determination of corrosion of other structures.

[0039]

Since the present invention has the above-described structure, the following effects can be obtained.

(1) The corrosion state of the outer surface of members constituting a structure such as a steel tower can be automatically detected, and the conventional visual inspection is eliminated, thereby greatly reducing labor and efficiency of work. Can be planned. In addition, detection of corrosion points, corrosion rate,
The determination of the deterioration level becomes objective, and the deterioration and corrosion can be reliably evaluated.

(2) Rust color characteristics (brightness, saturation,
(Hue, area, area ratio, etc.), the degree of corrosion or deterioration level of the corroded portion can be reliably determined without erroneous detection.

(3) The progress of the corroded portion can be easily known from the color of the corroded portion, and a repair plan such as repair time prediction and repair process management can be made.

[Brief description of the drawings]

FIG. 1 is an example in which a deterioration / corrosion detection / judgment system of the present invention is applied to a steel tower. FIG. 1 (a) is a mounting state diagram and a configuration diagram of an apparatus in the case of partially photographing using a lifting robot. )
FIG.

FIG. 2 is an example in which the deterioration / corrosion detection / judgment system of the present invention is applied to a steel tower. FIG. 2 (a) is an installation state diagram / configuration diagram of an apparatus for photographing the whole with a helicopter or the like, and FIG. It is the flowchart.

3A is a flowchart of photographing and member extraction in FIG. 1, and FIG. 3B is a flowchart of photographing and member extraction in FIG.

FIG. 4 is an image example of a steel tower used in the image processing of the present invention, wherein (a) is an original image, (b) is a brightness image, and (c) is a binary image.

FIG. 5 shows an RGB model and L ^* a ^* used for image processing.
It is explanatory drawing of a b ^* model.

FIG. 6 is an explanatory diagram of a Yuv model used for image processing.

FIG. 7 is a diagram showing a schematic procedure of image processing, showing an original image and each analysis image.

FIG. 8 shows sample images at each deterioration level according to the present invention;
The chroma, hue, and brightness histogram (example) are shown.

FIG. 9 is a corrosion determination map (example) in the image of FIG. 8;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Deterioration / corrosion detection judgment system 2 ... Camera 3 ... Video tape recorder 4 ... Color monitor 5 ... Computer 6 ... Image processing board 10 ... Elevating robot 11 ... Elevating auxiliary equipment 12 ... Helicopter 13 ... Mountain

──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Masatoshi Kikuchi 6-2-10 Ginza, Chuo-ku, Tokyo Stock Company Tomoe Corporation (72) Inventor Shinsaburo Takao 6-2-10 Ginza, Chuo-ku Tokyo In Tomoe Corporation (72) Inventor Hiroo Matsusue 6-2-10 Ginza, Chuo-ku, Tokyo, Japan Stock Company in Tomo Corporation (72) Inventor Norio Masaoka 6-2-10 Ginza, Chuo-ku, Tokyo In Stock Company in Tomoe Corporation (72) Inventor Takeo Tsushima 6-2-10 Ginza, Chuo-ku, Tokyo Inside Tomoe Corporation Co., Ltd. (56) References JP-A-7-318510 (JP, A) JP-A-9-229869 (JP, A) JP-A-7-120406 (JP, A) JP-A-11-37950 (JP, A) JP-A-8-170947 (JP, A) -260707 (JP, A) JP flat 9-229869 (JP, A) JitsuHiraku flat 3-125245 (JP, U) (58 ) investigated the field (Int.Cl. ^7, DB name) G01N 21/84 - 21/958 G01B 11/00-11/30 G06T 1/00

Claims (1)

(57) [Claims] 1. An imaging means capable of imaging an external surface of a member constituting a structure is installed on a robot or a flying machine moving along a structural member of the structure, or is installed at a distant place, and the imaging means is used for the structure. Shoot whole or partial objects,
The obtained color video is input to a computer equipped with an image processing function, and by this image processing, the inspection object is extracted, and the brightness, saturation, and hue obtained from the color original image are obtained.
At least the brightness and saturation of the inspection object from the three images
Determine the hue and use it for image processing from corrosion test specimens of each deterioration level
Deterioration at least consisting of lightness, saturation and hue
A brightness reference value, at least lightness and saturation of the inspection object
A deterioration / corrosion detection / judgment method for a surface-treated steel material , wherein a hue is compared to judge the degree of corrosion or deterioration level of the external surface of the inspection object.