JPH0961138A - Crack extraction apparatus - Google Patents

Crack extraction apparatus

Info

Publication number
JPH0961138A
JPH0961138A JP21582695A JP21582695A JPH0961138A JP H0961138 A JPH0961138 A JP H0961138A JP 21582695 A JP21582695 A JP 21582695A JP 21582695 A JP21582695 A JP 21582695A JP H0961138 A JPH0961138 A JP H0961138A
Authority
JP
Japan
Prior art keywords
crack
road surface
image
area
dark
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
JP21582695A
Other languages
Japanese (ja)
Inventor
Shingo Ito
慎悟 伊藤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mitsubishi Heavy Industries Ltd
Original Assignee
Mitsubishi Heavy Industries Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Mitsubishi Heavy Industries Ltd filed Critical Mitsubishi Heavy Industries Ltd
Priority to JP21582695A priority Critical patent/JPH0961138A/en
Publication of JPH0961138A publication Critical patent/JPH0961138A/en
Withdrawn legal-status Critical Current

Links

Landscapes

  • Image Processing (AREA)
  • Image Analysis (AREA)
  • Road Repair (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

PROBLEM TO BE SOLVED: To provide a crack extraction apparatus by which a crack on a road surface can be extracted with high accuracy. SOLUTION: A density-value histogram is created on the basis of a road- surface original image 12 which has been imaged by a road-surface imaging part 11. When a clear and bright part and a clear and dark part exist inside the original image 12, the original image is divided into a bright-part area and a dark-part area by using a threshold value. A crack-extraction processing operation by binarization is performed completely independeritly regarding every area by a shade-removing and crack-extraction part 131. In the crack- extraction processing operation regarding every area, a fixed threshold value which can extract a crack from a road surface is decided, the crack is extracted, a binary image regarding every area is composed finally, the whole image is restored, and the crack is extracted from the image as a whole.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【発明の属する技術分野】本発明は、路面性状測定の一
要素であるひび割れ抽出を画像処理によって行うひび割
れ抽出装置に関するものである。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a crack extraction device for performing crack extraction, which is one element of road surface property measurement, by image processing.

【0002】[0002]

【従来の技術】従来、ある路面区間内のひび割れ度合の
計測を画像処理により行う路面ひび割れ計測装置とし
て、図6に示すようなものがある。この場合、路面を撮
像する路面撮像部1から連続的に出力される路面原画像
2は、ディジタル化され路面ひび割れ計測装置3に入力
される。この路面ひび割れ計測装置3では、まず、ひび
割れ抽出部31の二値化処理によって路面原画像2から
ひび割れを抽出し、ひび割れが抽出された二値画像32
を路面ひび割れ度合算出部33に入力して、各画像内の
ひび割れの数量を算出する。そして、各画像内のひび割
れ数量に基づいて舗装試験法便覧で定義されている計算
式で計測区間内の路面のひび割れ度合を算出する。ここ
で、舗装試験法便覧で定義されている計算式を図7に示
している。
2. Description of the Related Art Conventionally, there is a road surface crack measuring device as shown in FIG. 6 which measures the degree of cracking in a certain road surface section by image processing. In this case, the road surface original image 2 continuously output from the road surface imaging unit 1 that images the road surface is digitized and input to the road surface crack measuring device 3. In the road surface crack measuring device 3, first, a crack is extracted from the road surface original image 2 by the binarization processing of the crack extraction unit 31, and the binary image 32 in which the crack is extracted.
Is input to the road surface crack degree calculation unit 33 to calculate the number of cracks in each image. Then, based on the number of cracks in each image, the degree of cracking of the road surface in the measurement section is calculated by the calculation formula defined in the Pavement Test Manual. Here, the calculation formula defined in the Pavement Test Method Handbook is shown in FIG.

【0003】ところで、路面原画像2中に影などのため
にはっきりとした暗部と明部が存在するような場合、ひ
び割れ抽出部31での二値化処理は、固定しきい値で全
画素を二値化すると、明部と暗部の明度の差が大きいた
め、明部と暗部のひび割れを同時に抽出できないことが
あり、このため、路面原画像2内の1画素毎に、周りの
近傍画素を考慮に入れて、しきい値を決定する、いわゆ
る動的しきい値決定により二値化を行うようにしてい
る。
By the way, when the road surface original image 2 has a clear dark part and a bright part due to a shadow or the like, the binarization processing in the crack extraction part 31 uses all pixels with a fixed threshold value. When binarized, the difference in brightness between the bright and dark areas is large, and therefore cracks in the bright and dark areas may not be extracted at the same time. Therefore, for each pixel in the road surface original image 2, surrounding neighboring pixels may be extracted. Taking this into consideration, the binarization is performed by the so-called dynamic threshold determination that determines the threshold.

【0004】図8は、このような動的しきい値決定によ
るひび割れ抽出部31での処理フローを示すもので、ま
ず、ステップ801で、二値化すべき画素および上下左
右の画素分(近傍画素)の(2a+1)2 個の濃度の平
均値を算出し、次いで、ステップ802で、濃度の平均
値と二値化すべき画素の濃度を比較して二値化を行い、
さらに、ステップ803で、原画像内の全画素二値化が
終了したか判断し、終了していなければステップ804
で、別画素選択してステップ801に戻り、以下、同様
な動作を繰り返すことにより、明部、暗部およびその境
界わけ隔てなく画像内の全画素に対して処理が行われる
ようになる。
FIG. 8 shows a processing flow in the crack extraction unit 31 based on such dynamic threshold value determination. First, in step 801, pixels to be binarized and upper, lower, left and right pixels (neighboring pixels). ), The average value of the (2a + 1) 2 densities is calculated, and in step 802, the average value of the densities is compared with the densities of the pixels to be binarized to perform binarization.
Further, in step 803, it is determined whether binarization of all pixels in the original image is completed, and if not, step 804.
Then, another pixel is selected, the process returns to step 801, and thereafter, the same operation is repeated, so that the processing is performed on all pixels in the image without separating the bright portion, the dark portion, and the boundaries thereof.

【0005】[0005]

【発明が解決しようとする課題】ところで、このような
動的しきい値決定による二値化処理によると、上述した
影などによりはっきりとした明部と暗部が存在する路面
原画像からひび割れを二値抽出するような場合、二値化
すべき画素および近傍画素が完全に明部内あるいは暗部
内にあるような場合は、ひび割れは路面より黒いという
想定のもとに、確実に抽出される。
By the way, according to the binarization processing by such dynamic threshold value determination, it is possible to detect cracks from the road surface original image in which there are clear bright parts and dark parts due to the above-mentioned shadows and the like. In the case of value extraction, when the pixel to be binarized and the neighboring pixel are completely in the bright part or the dark part, the crack is surely extracted under the assumption that the crack is blacker than the road surface.

【0006】ところが、明部と暗部の境界付近の画素を
二値化するような場合、近傍画素が明部によって濃度平
均値が上がるために、ひび割れ部分の画素と同様にして
暗部路面の画素も抽出されてしまうことがあるため、ひ
び割れ部分が、明部と暗部の境界付近にあるような場
合、このひび割れ部分は、抽出された暗部路面に埋もれ
てしまい抽出できなくなる。
However, in the case of binarizing the pixels near the boundary between the bright portion and the dark portion, the density average value of the neighboring pixels increases due to the bright portion, so that the pixels on the dark road surface are the same as the pixels at the crack portion. Since the cracked portion may be extracted, if the cracked portion is near the boundary between the bright portion and the dark portion, the cracked portion is buried in the extracted dark road surface and cannot be extracted.

【0007】このことから、影などによりはっきりとし
た明部と暗部が存在する路面原画像の場合は、影部分が
原因で正確なひび割れ抽出ができないことがあり、ひび
割れ度合の精度が下がってしまうという問題点があっ
た。本発明は、上記事情に鑑みてなされたもので、精度
の高い路面ひび割れ抽出を可能にしたひび割れ抽出装置
を提供することを目的とする。
Therefore, in the case of the road surface original image in which the bright and dark portions are clearly present due to the shadow or the like, the accurate extraction of the crack may not be possible due to the shadow portion, and the accuracy of the crack degree is lowered. There was a problem. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a crack extraction device that enables highly accurate road surface crack extraction.

【0008】[0008]

【課題を解決するための手段】請求項1記載の発明は、
路面原画像を撮像する撮像手段と、この撮像手段により
撮像された路面原画像内に明部と暗部があるかを判断す
る明暗部判断手段と、この手段により判断された明暗部
を明部エリアと暗部エリアに分けてそれぞれ二値化によ
るひび割れ抽出処理を行うひび割れ抽出手段とにより構
成している。
According to the first aspect of the present invention,
An image pickup means for picking up an original road surface image, a light / dark part decision means for judging whether there is a bright portion or a dark portion in the original road surface image picked up by this image pickup means, and a light / dark portion decided by this means is a bright area. And a dark area and a crack extracting unit that performs a crack extracting process by binarization.

【0009】請求項2記載の発明では、請求項1記載に
おいて、各エリアでのひび割れ抽出処理は、路面からひ
び割れを抽出できる固定のしきい値を決定してひび割れ
抽出処理を行うようにしている。
According to a second aspect of the present invention, in the first aspect, the crack extraction processing in each area is performed by determining a fixed threshold value for extracting the crack from the road surface. .

【0010】請求項3記載の発明では、請求項2記載に
おいて、さらに、各エリアの二値画像を合成して、全体
画像を復元し画像全体からひび割れを抽出するようにし
ている。
According to a third aspect of the present invention, in addition to the second aspect, the binary images of each area are combined to restore the entire image and the cracks are extracted from the entire image.

【0011】この結果、本発明によれば、影などにより
はっきりとした明部と暗部が存在する路面原画像の場合
でも、暗部と明部を完全に分けて二値化処理を行うこと
から、ひび割れ部分が明部と暗部の境界付近にあるよう
な場合も、明部と暗部の濃度差により、ひび割れが暗部
路面に埋もれるといったことはなく、これら境界付近の
ひび割れも正確に抽出できる、精度の高い路面ひび割れ
抽出を行うことができる。
As a result, according to the present invention, even in the case of a road surface original image in which a bright portion and a dark portion are clearly present due to a shadow or the like, the dark portion and the light portion are completely separated and the binarization processing is performed. Even if the cracked portion is near the boundary between the light and dark areas, the difference in density between the light and dark areas will not cause the crack to be buried in the road surface in the dark area. It is possible to perform high road crack extraction.

【0012】[0012]

【発明の実施の形態】以下、本発明の一実施の形態を図
面に従い説明する。この場合、図1は、本発明を適用し
た路面ひび割れ計測装置の概略構成図、図2は、ひび割
れ抽出部および影除去ひび割れ抽出部での処理フロー、
図3は原画像から得られる濃度ヒストグラムの概念図、
図4および図5は、路面ひび割れ度合算出部の処理フロ
ーである。
DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. In this case, FIG. 1 is a schematic configuration diagram of a road surface crack measuring device to which the present invention is applied, and FIG. 2 is a processing flow in a crack extraction unit and a shadow removal crack extraction unit,
3 is a conceptual diagram of a density histogram obtained from the original image,
4 and 5 are process flows of the road surface crack degree calculating unit.

【0013】まず、図1において、11は路面撮像部
で、この路面撮像部11は、ある計測区間内で連続的に
路面を撮像し、この撮像された2次元の路面原画像12
をディジタル化して順次路面ひび割れ計測装置13に入
力するようにしている。
First, in FIG. 1, reference numeral 11 denotes a road surface image pickup unit, and this road surface image pickup unit 11 continuously picks up an image of the road surface within a certain measurement section, and the imaged two-dimensional road surface original image 12 is taken.
Are digitized and sequentially input to the road surface crack measuring device 13.

【0014】路面ひび割れ計測装置13は、影除去ひび
割れ抽出部131、ひび割れ抽出部132、路面ひび割
れ度合算出部133を有するものである。そして、路面
ひび割れ計測装置13に入力された路面原画像12は、
影除去ひび割れ抽出部131またはひび割れ抽出部13
2に入力され、二値化によるひび割れ抽出処理され、そ
の後、これら影除去ひび割れ抽出部131またはひび割
れ抽出部132からの二値画像134を路面ひび割れ度
合算出部133に入力するようにしている。
The road surface crack measuring device 13 has a shadow removal crack extracting portion 131, a crack extracting portion 132, and a road surface crack degree calculating portion 133. Then, the road surface original image 12 input to the road surface crack measuring device 13 is
Shadow removal crack extraction unit 131 or crack extraction unit 13
2, the binary image 134 from the shadow removal crack extracting unit 131 or the crack extracting unit 132 is input to the road surface crack degree calculating unit 133.

【0015】また、路面ひび割れ度合算出部133は、
入力された二値画像134内のひび割れの数量あるいは
パッチング面積を算出するもので、最終的に、上述した
図7に示される舗装試験法便覧により定義されている計
算式で計測区間内の路面のひび割れ度合を算出するよう
にしている。
Further, the road surface crack degree calculation unit 133 is
The number of cracks or the patching area in the input binary image 134 is calculated. Finally, the road surface in the measurement section is calculated by the calculation formula defined by the Pavement Test Method Manual shown in FIG. 7 described above. The degree of cracking is calculated.

【0016】次に、路面ひび割れ計測装置13を構成す
る影除去ひび割れ抽出部131またはひび割れ抽出部1
32の具体的な処理内容を図2に示す処理フローにより
説明する。
Next, the shadow removal crack extracting section 131 or the crack extracting section 1 which constitutes the road surface crack measuring device 13
The specific processing contents of 32 will be described with reference to the processing flow shown in FIG.

【0017】まず、ステップ201で、撮像された2次
元の路面原画像12を用いて、画像内の画素の濃度(明
度を表わすディジタルデータ)と度数(画素数)を示す
濃度値ヒストグラムを作成する。ここで、図3は、ヒス
トグラムの概念図を示すもので、仮に、同図(a)に示
すように原画像12内に明確な明部と暗部が存在するよ
うな場合は、同図(b)に示すようにヒストグラムに複
数の山が現れるようになる。
First, in step 201, using the captured two-dimensional original road surface image 12, a density value histogram showing the density (digital data representing lightness) and the frequency (number of pixels) of pixels in the image is created. . Here, FIG. 3 is a conceptual diagram of a histogram. If there are clear bright and dark parts in the original image 12 as shown in FIG. ), Multiple peaks appear in the histogram.

【0018】そして、ステップ202で、ヒストグラム
の山が1つかを判断する。原画像12を基に最初に作成
されたヒストグラムの山が1つの場合は、原画像12内
に明部と暗部が存在しないことになるので、エリア分離
する必要はなく、ステップ203以降に進み、ひび割れ
抽出部132による処理を行う。この場合、ステップ2
03で、ヒストグラムからひび割れおよびパッチングの
エッジが抽出される濃度しきい値を決定し、ステップ2
04で、決定されたしきい値を用いて全体画像の二値化
を行い、ひび割れおよびパッチングのエッジ抽出を行う
ようになる。
Then, in step 202, it is judged whether there is one mountain in the histogram. When there is one mountain in the histogram first created based on the original image 12, it means that there is no bright part and dark part in the original image 12, so there is no need to separate the areas, and the process proceeds to step 203 and subsequent steps. Processing by the crack extraction unit 132 is performed. In this case, step 2
At 03, determine the density threshold at which crack and patching edges are extracted from the histogram, and step 2
At 04, the entire image is binarized by using the determined threshold value, and the crack and patching edge extraction is performed.

【0019】一方、ステップ202で、ヒストグラムの
山が2つと判断すると、原画像12内に明部と暗部が存
在することになるので、ステップ205以降に進み、影
除去ひび割れ抽出部131による処理を行う。この場
合、ステップ205で、ヒストグラムを用いて明部エリ
アと暗部エリアを分けるしきい値(図3の符号135)
を決定し、次いで、ステップ206で、原画像の暗部エ
リアのみが処理できる暗部マスクと、明部エリアのみ処
理できる明部マスクを作成する。
On the other hand, if it is determined in step 202 that the histogram has two peaks, it means that the original image 12 has a bright portion and a dark portion. To do. In this case, in step 205, the threshold value for dividing the bright area and the dark area using the histogram (reference numeral 135 in FIG. 3).
Then, in step 206, a dark mask that can process only the dark area of the original image and a bright mask that can process only the bright area are created.

【0020】次に、ステップ207で、原画像12にこ
れら各マスクをかけて、ステップ208で、マスクエリ
ア内のみのヒストグラムを作成して、ステップ209
で、このヒストグラムを用いて、ひび割れおよびパッチ
ングのエッジ抽出のための濃度しきい値を決定し、その
後、ステップ210で、この決定したしきい値を用いて
そのマスクエリア内の二値化処理を行い、二値画像を作
成する。
Next, in step 207, each of these masks is applied to the original image 12, and in step 208, a histogram only in the mask area is created, and in step 209
Then, the histogram is used to determine a density threshold value for edge extraction of cracks and patching, and thereafter, in step 210, the binarization processing in the mask area is performed using the determined threshold value. Do and create a binary image.

【0021】そして、ステップ211で、全マスクエリ
ア内について処理を終了したかを判断し、終了していな
ければ、ステップ207に戻り、以下、同様な動作を繰
り返すことにより、全マスクエリア内について二値画像
を作成し、最終的に、ステップ212で、全エリアの二
値画像を合成し全体画像を再構成するようになる。
Then, in step 211, it is judged whether or not the processing has been completed for all the mask areas. If not completed, the procedure returns to step 207, and the same operation is repeated thereafter, so that the entire mask area is reprocessed. A value image is created, and finally, in step 212, the binary images of all areas are combined to reconstruct the entire image.

【0022】このようにすれば、二値化によるひび割れ
やパッチングのエッジ抽出処理をエリア毎に完全に独立
して行うようになるため、明部と暗部の境界付近での明
部と暗部の濃度差の影響が無くなり、境界付近でもひび
割れやパッチングのエッジが抽出できる。
In this way, since the edge extraction processing for cracking and patching by binarization is performed completely independently for each area, the density of the bright part and the dark part near the boundary between the bright part and the dark part is increased. The influence of the difference is eliminated, and cracks and patching edges can be extracted even near the boundary.

【0023】次に、これら影除去ひび割れ抽出部131
あるいはひび割れ抽出部132から出力された二値画像
は、路面ひび割れ度合算出部133に与えられ、計測区
間内の路面のひび割れ度合が算出されるが、この場合の
具体的な処理を図4および図5に示す処理フローにより
説明する。
Next, these shadow removal crack extraction unit 131
Alternatively, the binary image output from the crack extraction unit 132 is given to the road surface crack degree calculation unit 133, and the crack degree of the road surface in the measurement section is calculated. Specific processing in this case is shown in FIG. 4 and FIG. This will be described with reference to the processing flow shown in FIG.

【0024】まず、ステップ401で、二値画像全体に
現われている路面上の微小ノイズを微小領域除去により
取り除く。次に、ステップ402で、画像内にパッチン
グが存在するかチェックするために、二値画像の縦横に
画素毎に1画素抜き出して、元の画像の1/n×1/n
の縮小二値画像を構成する。この処理は、ひび割れ等の
影響をなるべく少なくしパッチングのみに着目しやすく
するためである。
First, in step 401, minute noise on the road surface appearing in the entire binary image is removed by minute area removal. Next, in step 402, in order to check whether there is patching in the image, one pixel is extracted for each pixel in the vertical and horizontal directions of the binary image to obtain 1 / n × 1 / n of the original image.
A reduced binary image of. This processing is for reducing the influence of cracks and the like as much as possible and making it easy to focus only on patching.

【0025】次に、ステップ403で、縮小二値画像内
の微小ノイズを削除した後、ステップ404で、画像内
の微小線分をつなぎ合わせるための線分連結を行う。そ
して、ステップ405で、パッチング有りかを画像上に
パッチングの輪郭線(エッジ)に対応する線分があるか
否かによりチェックする。ここで、線分がパッチングの
輪郭線かどうかの判断は、線分の形状、線分で形づくら
れる面の大きさ(面積)の特微量により行う。パッチン
グの形状には複雑なものはない(せいぜい円形か四角
形)し、パッチングと称するものの大きさの範囲も限定
できるので、これらのルールを基に判断を行うようにな
る。
Next, in step 403, minute noises in the reduced binary image are deleted, and in step 404, line segment connection for connecting the minute line segments in the image is performed. Then, in step 405, it is checked whether or not there is patching by whether or not there is a line segment corresponding to the contour line (edge) of patching on the image. Here, the determination as to whether the line segment is a contour line for patching is made based on the characteristics of the shape of the line segment and the size (area) of the surface formed by the line segment. Since there is no complicated patching shape (a circle or a rectangle at most) and the size range of what is called patching can be limited, judgment is made based on these rules.

【0026】そして、これらの判断によりパッチングの
輪郭線が見つかった場合、ステップ406で、輪郭線で
形づくられるパッチング面の面積を面内の画素数により
算出してストアする。そして、ステップ407で、縮小
画像でのパッチングの輪郭線(エッジ)データをもとに
元の二値画像からパッチングの輪郭線(エッジ)を除去
して、ステップ408以降の処理に進む。
If a patching contour line is found by these judgments, in step 406, the area of the patching surface formed by the contour line is calculated by the number of pixels in the surface and stored. Then, in step 407, the patching contour line (edge) is removed from the original binary image based on the patching contour line (edge) data in the reduced image, and the process proceeds to step 408 and subsequent steps.

【0027】なお、ステップ405で、パッチングなし
を判断した場合は、ステップ406、ステップ407の
処理を飛ばして直ちにステップ408以降の処理に進
む。ステップ408以降の処理は、ひび割れが画像内に
あるかをチェックするためのもので、まず、ステップ4
08で、元の二値画像内の微小線分をつなぎ合わせるた
めの線分連結を行い、その後、ステップ409で、元の
二値画像内で抽出されている領域の面積、円形度、長さ
といった特微量を算出して、ステップ410で、ひび割
れ以外の物体を除去する。
When it is determined in step 405 that there is no patching, the processing in steps 406 and 407 is skipped and the processing immediately proceeds to step 408 and thereafter. The processing after step 408 is for checking whether a crack is present in the image.
In step 08, line segment connection for connecting the minute line segments in the original binary image is performed. Then, in step 409, the area, circularity, and length of the region extracted in the original binary image are connected. Then, in step 410, the objects other than the cracks are removed.

【0028】そして、ステップ411で、ひび割れ有り
を判断する。ここで、ひび割れ以外の物体を除去した後
に残ったものがあれば、ひび割れが存在すると判断し、
ステップ412で、ひび割れを細線化して幅が1画素に
なるようにし、次いで、ステップ413で、その線分か
らひび割れの数量や長さを算出してストアする。ここ
で、ひび割れの数量は線分の数、ひび割れの長さは線分
の画素数で計算している。
Then, in step 411, it is judged that there is a crack. Here, if there is something left after removing objects other than cracks, it is judged that cracks exist,
In step 412, the crack is thinned to have a width of 1 pixel, and in step 413, the number and length of cracks are calculated from the line segment and stored. Here, the number of cracks is calculated by the number of line segments, and the length of the crack is calculated by the number of pixels of the line segment.

【0029】そして、ステップ414で、計測区間内の
全ての画像について上述の処理が終了したか判断し、終
了していなければ、処理を終了し、計測区間内の全ての
画像について上述の処理が終了したとき、ステップ41
5で、ストアされたデータ(パッチングの面積、ひび割
れの数や長さ)をもとに図7に示される計算式を用いて
路面のひび割れ度合を算出するようになる。
Then, in step 414, it is judged whether the above-mentioned processing has been completed for all the images in the measurement section. If not, the processing is ended and the above-mentioned processing is executed for all the images in the measurement section. When finished, step 41
In step 5, the degree of cracking of the road surface is calculated using the calculation formula shown in FIG. 7 based on the stored data (area of patching, number and length of cracks).

【0030】なお、ステップ411で、ひび割れなしを
判断した場合は、ステップ412、ステップ413の処
理を飛ばして直ちにステップ414以降の処理に進む。
従って、このようにすれば、路面撮像部11で撮像され
た路面原画像12から濃度値ヒストグラムを作成し、原
画像12内に明確な明部と暗部がある場合、しきい値を
用いて原画像12を明部エリアと暗部エリアに分けて、
影除去ひび割れ抽出部131により各エリア毎に二値化
によるひび割れ抽出処理を完全に独立して行い、これら
各エリアでのひび割れ抽出処理は、ひび割れを路面から
抽出できる固定のしきい値を決定して、ひび割れを抽出
し、最後に、各エリアの二値画像を合成して、全体画像
を復元し画像全体からひび割れを抽出するようにしてい
る。これにより、影などによりはっきりとした明部と暗
部が存在する路面原画像の場合でも、暗部と明部を完全
に分けて二値化処理を行うことから、ひび割れ部分が明
部と暗部の境界付近にあるような場合でも、明部と暗部
の濃度差により、ひび割れが暗部路面に埋もれるといっ
たことはなく、これら境界付近のひび割れも正確に抽出
できる、精度の高い路面ひび割れ抽出を行うことができ
る。
When it is determined in step 411 that there is no crack, the processing in steps 412 and 413 is skipped and the processing immediately proceeds to step 414 and thereafter.
Therefore, by doing so, a density value histogram is created from the road surface original image 12 imaged by the road surface imaging unit 11, and when there are clear bright and dark parts in the original image 12, the original value is calculated using the threshold value. The image 12 is divided into a light area and a dark area,
The shadow removal crack extraction unit 131 completely and independently performs the crack extraction processing by binarization for each area, and the crack extraction processing in each of these areas determines a fixed threshold value by which the crack can be extracted from the road surface. Then, the cracks are extracted, and finally, the binary images of each area are combined to restore the entire image and the cracks are extracted from the entire image. As a result, even in the case of a road surface image in which there are clear bright and dark areas due to shadows, etc., since the dark and bright areas are completely separated and binarization processing is performed, the cracked area is the boundary between the bright and dark areas. Even if it is in the vicinity, cracks will not be buried in the dark road due to the difference in density between the light and dark areas, and cracks near these boundaries can be extracted accurately, and highly accurate road crack extraction can be performed. .

【0031】[0031]

【発明の効果】以上述べたように本発明によれば、暗部
エリアと明部エリアに分けて、二値化によるひび割れ抽
出を各エリアで独立して行うことから、ひび割れ部分が
明部と暗部の境界付近にあるような場合でも、明部と暗
部の濃度差により、ひび割れが暗部路面に埋もれるとい
ったことはなく、これら境界付近のひび割れも正確に抽
出できる、精度の高い路面ひび割れ抽出を行うことがで
きる。
As described above, according to the present invention, since the crack extraction by binarization is performed independently for each area by dividing it into a dark area and a bright area, the cracked portion is a bright area and a dark area. Even if it is near the boundary, the cracks will not be buried in the dark road surface due to the difference in density between the light and dark areas, and cracks near these boundaries can be accurately extracted. You can

【図面の簡単な説明】[Brief description of drawings]

【図1】本発明の第1実施の形態に係る路面ひび割れ計
測装置の概略構成を示す図。
FIG. 1 is a diagram showing a schematic configuration of a road surface crack measuring device according to a first embodiment of the present invention.

【図2】第1実施の形態のひび割れ抽出部および影除去
ひび割れ抽出部での処理フローを示す図
FIG. 2 is a diagram showing a processing flow in a crack extraction unit and a shadow removal crack extraction unit of the first embodiment.

【図3】濃度値ヒストグラムを説明するための概念図FIG. 3 is a conceptual diagram for explaining a density value histogram.

【図4】第1実施の形態の路面ひび割れ度合算出部での
処理フローを示す図。
FIG. 4 is a diagram showing a processing flow in a road surface crack degree calculating unit according to the first embodiment.

【図5】第1実施の形態の路面ひび割れ度合算出部での
処理フローを示す図。
FIG. 5 is a diagram showing a processing flow in a road surface crack degree calculating unit of the first embodiment.

【図6】従来のひび割れ計測装置の概略構成を示す図。FIG. 6 is a diagram showing a schematic configuration of a conventional crack measuring device.

【図7】舗装試験法便覧に定義されている路面ひび割れ
度合計算式を示す図。
FIG. 7 is a diagram showing a road surface crack degree calculation formula defined in the Pavement Test Method Handbook.

【図8】従来の動的しきい値決定によるひび割れ抽出部
の処理フローを示す図。
FIG. 8 is a diagram showing a processing flow of a crack extraction unit by conventional dynamic threshold determination.

【符号の説明】[Explanation of symbols]

11…路面撮像部、 12…路面原画像、 13…路面ひび割れ計測装置、 131…影除去ひび割れ抽出部、 132…ひび割れ抽出部、 133…路面ひび割れ度合算出部、 134…二値画像、 135…しきい値。 11 ... Road surface imaging unit, 12 ... Road surface original image, 13 ... Road surface crack measuring device, 131 ... Shadow removal crack extraction unit, 132 ... Crack extraction unit, 133 ... Road surface crack degree calculation unit, 134 ... Binary image, 135 ... Threshold.

Claims (3)

【特許請求の範囲】[Claims] 【請求項1】 路面原画像を撮像する撮像手段と、 この撮像手段により撮像された路面原画像内に明部と暗
部があるかを判断する明暗部判断手段と、 この手段により判断された明暗部を明部エリアと暗部エ
リアに分けてそれぞれ二値化によるひび割れ抽出処理を
行うひび割れ抽出手段と、 を具備したことを特徴とするひび割れ抽出装置。
1. An image pickup means for picking up an original road surface image, a light and dark part decision means for judging whether there is a bright portion and a dark portion in the original road surface image picked up by the image pickup means, and a light and dark portion decided by this means. A crack extracting device comprising: a crack extracting unit that divides a portion into a bright area and a dark area and performs a crack extracting process by binarization.
【請求項2】 各エリアでのひび割れ抽出処理は、路面
からひび割れを抽出できる固定のしきい値を決定してひ
び割れを抽出処理を行うことを特徴とする請求項1記載
のひび割れ抽出装置。
2. The crack extracting device according to claim 1, wherein in the crack extracting process in each area, the crack extracting process is performed by determining a fixed threshold value capable of extracting the crack from the road surface.
【請求項3】 さらに、各エリアの二値画像を合成し
て、全体画像を復元し画像全体からひび割れを抽出する
ことを特徴とする請求項2記載のひび割れ抽出装置。
3. The crack extracting apparatus according to claim 2, further comprising: combining the binary images of the respective areas to restore the entire image and extracting the crack from the entire image.
JP21582695A 1995-08-24 1995-08-24 Crack extraction apparatus Withdrawn JPH0961138A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP21582695A JPH0961138A (en) 1995-08-24 1995-08-24 Crack extraction apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP21582695A JPH0961138A (en) 1995-08-24 1995-08-24 Crack extraction apparatus

Publications (1)

Publication Number Publication Date
JPH0961138A true JPH0961138A (en) 1997-03-07

Family

ID=16678906

Family Applications (1)

Application Number Title Priority Date Filing Date
JP21582695A Withdrawn JPH0961138A (en) 1995-08-24 1995-08-24 Crack extraction apparatus

Country Status (1)

Country Link
JP (1) JPH0961138A (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100443667B1 (en) * 2002-01-23 2004-08-09 주식회사 비젼넷 Method for detecting a crack of pavement
JP2006337237A (en) * 2005-06-03 2006-12-14 Honda Motor Co Ltd Method and apparatus for inspecting bonded state of structure
KR100826153B1 (en) * 2006-11-29 2008-04-30 한국표준과학연구원 Width measurement method of the crack by using the depth value in histogram of image
JP2011058940A (en) * 2009-09-09 2011-03-24 Panasonic Electric Works Co Ltd Apparatus and method for visual inspection
CN103161121A (en) * 2013-03-19 2013-06-19 南京道润交通科技有限公司 Automatic mileage stake mark calibration method, system and method
CN105320962A (en) * 2015-10-21 2016-02-10 东南大学 Pavement damage type identification method based on classifier ensemble
JP2016045508A (en) * 2014-08-19 2016-04-04 インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation Method and algorithm for estimating degree of degradation of road surface (crack detection method using gabor filter output image of multi-resolution image)
CN106223176A (en) * 2016-09-07 2016-12-14 吉林大学 Vehicle-mounted road surface flaw detection system light shielding device based on flexible cover plate
CN106223177A (en) * 2016-09-07 2016-12-14 吉林大学 Vehicle-mounted road surface flaw detection system light shielding device based on cam space connecting-rod
WO2018122819A1 (en) * 2016-12-30 2018-07-05 同济大学 Binocular image analysis-based asphalt road surface damage detection system
JP2019045922A (en) * 2017-08-30 2019-03-22 株式会社ネクスコ・エンジニアリング北海道 Deformed spot detection device, deformed spot detection program and deformed spot detection method
KR20200021720A (en) * 2018-08-21 2020-03-02 한국철도기술연구원 Apparatus and method for recognizing crack in sturcutre
KR20200025480A (en) * 2018-08-30 2020-03-10 한국철도기술연구원 Boundary based structure crack detection apparatus and method

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100443667B1 (en) * 2002-01-23 2004-08-09 주식회사 비젼넷 Method for detecting a crack of pavement
JP2006337237A (en) * 2005-06-03 2006-12-14 Honda Motor Co Ltd Method and apparatus for inspecting bonded state of structure
JP4551821B2 (en) * 2005-06-03 2010-09-29 本田技研工業株式会社 Method and apparatus for inspection of bonding state of structure
KR100826153B1 (en) * 2006-11-29 2008-04-30 한국표준과학연구원 Width measurement method of the crack by using the depth value in histogram of image
JP2011058940A (en) * 2009-09-09 2011-03-24 Panasonic Electric Works Co Ltd Apparatus and method for visual inspection
CN103161121A (en) * 2013-03-19 2013-06-19 南京道润交通科技有限公司 Automatic mileage stake mark calibration method, system and method
JP2016045508A (en) * 2014-08-19 2016-04-04 インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation Method and algorithm for estimating degree of degradation of road surface (crack detection method using gabor filter output image of multi-resolution image)
US9390343B2 (en) 2014-08-19 2016-07-12 International Business Machines Corporation Estimating degree of deterioration of road surface
CN105320962A (en) * 2015-10-21 2016-02-10 东南大学 Pavement damage type identification method based on classifier ensemble
CN106223177A (en) * 2016-09-07 2016-12-14 吉林大学 Vehicle-mounted road surface flaw detection system light shielding device based on cam space connecting-rod
CN106223176A (en) * 2016-09-07 2016-12-14 吉林大学 Vehicle-mounted road surface flaw detection system light shielding device based on flexible cover plate
WO2018122819A1 (en) * 2016-12-30 2018-07-05 同济大学 Binocular image analysis-based asphalt road surface damage detection system
CN109716108A (en) * 2016-12-30 2019-05-03 同济大学 A kind of Asphalt Pavement Damage detection system based on binocular image analysis
CN109804119A (en) * 2016-12-30 2019-05-24 同济大学 A kind of asphalt pavement crack development degree detection method based on infrared thermal imagery map analysis
GB2571016A (en) * 2016-12-30 2019-08-14 Du Yuchuan Binocular image analysis-based asphalt road surface damage detection system
GB2571016B (en) * 2016-12-30 2020-09-02 Du Yuchuan A system for detecting crack growth of asphalt pavement based on binocular image analysis
CN109804119B (en) * 2016-12-30 2021-03-19 同济大学 Asphalt pavement crack development degree detection method based on infrared thermography analysis
CN109716108B (en) * 2016-12-30 2021-07-13 同济大学 Bituminous paving disease detecting system based on two mesh image analysis
US11221107B2 (en) 2016-12-30 2022-01-11 Yuchuan DU Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing
JP2019045922A (en) * 2017-08-30 2019-03-22 株式会社ネクスコ・エンジニアリング北海道 Deformed spot detection device, deformed spot detection program and deformed spot detection method
KR20200021720A (en) * 2018-08-21 2020-03-02 한국철도기술연구원 Apparatus and method for recognizing crack in sturcutre
KR20200025480A (en) * 2018-08-30 2020-03-10 한국철도기술연구원 Boundary based structure crack detection apparatus and method

Similar Documents

Publication Publication Date Title
CN112419250A (en) Pavement crack digital image extraction, crack repair and crack parameter calculation method
TWI412739B (en) Defect detection method and defect detection apparatus
CN110675346A (en) Image acquisition and depth map enhancement method and device suitable for Kinect
JPH0961138A (en) Crack extraction apparatus
JP2018128309A (en) Crack detection method
JP2001184510A (en) Defect detecting method
JPH03160349A (en) Device for detecting crack
JP4954469B2 (en) Appearance inspection method
JP3572750B2 (en) Automatic evaluation method for concrete defects
JP2013002839A (en) Crack detection method
JP2006170922A (en) Visual inspection method and its apparatus
JP3333568B2 (en) Surface defect inspection equipment
JP3330829B2 (en) Automatic detection method of evaluable area in images of machine parts
JP2005069887A (en) Defect inspection method and apparatus
JP4104213B2 (en) Defect detection method
KR100754969B1 (en) Image inspecting apparatus, image inspecting method, and computer-readable storage medium
JPH1125222A (en) Method and device for segmenting character
CN112465817B (en) Pavement crack detection method based on directional filter
JP5157575B2 (en) Defect detection method
JP2711649B2 (en) Surface scratch detection method for inspection object
JP4635760B2 (en) Image processing apparatus and method
JP3753234B2 (en) Defect detection method
JPH0723845B2 (en) Defect detection method
JP3409695B2 (en) Surface defect inspection method
JP3190173B2 (en) Semiconductor device appearance inspection method and appearance inspection device

Legal Events

Date Code Title Description
A300 Withdrawal of application because of no request for examination

Free format text: JAPANESE INTERMEDIATE CODE: A300

Effective date: 20021105