JP2004069434A - Image processing method using image with laser light spot, mosaic processing method using the image, and normalization service system using the method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system which obtains an entire image of a concrete structure by obtaining the distance between a several laser light spots, right and left angles, upper and lower angles, objective distance, reduced scale and the like and then automatically normalizing an image using these parameters, and numerically notifies the state of a breakage automatically. <P>SOLUTION: Distant view images are rearranged so that it can be understood to which place of a surface 1 to be investigated the images each correspond. One distant view image selected from multiple first distant view images is displayed on a screen, all laser Pi points in the distant view image are designated in a predetermined direction, and then the reduced scale ratio of the image is obtained based on the vertical and lateral coordinates of four points of a second laser projectile. Next, the aspect ratio of one distant view image on the screen is corrected based on the reduced scale, objective distance, right and left angles, and upper and lower angles, adjacent distant view images are sequentially displayed, and the same point on these images is designated for each display, thereby obtaining a mosaic image which is an entire image of the surface 1 to be investigated. <P>COPYRIGHT: (C)2004,JPO

Description

の各式▲１▼，▲２▼，▲３▼，▲４▼，▲５▼，▲６▼が成り立つ。
【００７１】
ここで、▲１▼，▲２▼は並行光線成分、▲３▼，▲４▼は上下角度付光線成分、▲５▼，▲６▼は左右角度付光線成分を示し、
これら▲１▼，▲２▼，▲３▼，▲４▼，▲５▼，▲６▼の式よりふれ角ｔａｎθと被調査面３まで距離ｘが求められる。
【００７２】
【数２】

一方、図７にあっては、０１〜０２：焦点距離ｆ、０２〜０３：測定距離Ｘレーザビーム間距離ａとした時にθ角カメラ本体７をふった時、０１〜ａ´１：Ｌ１，０１〜ａ´２：Ｌ２，０１〜ｂ´１：Ｌ３（Ｌ５），０１〜ｂ´２：Ｌ４（Ｌ６）から、（直線ビーム）Ｐ_１，Ｐ_２に関し、▲１▼，▲２▼の式が得られる。
【００７３】
【数３】

垂直軸方向角付きビームＰ_３，Ｐ_４に関し、▲３▼，▲４▼の式が得られる。
【００７４】
【数４】

水平軸方向角度付きビームＰ_５，Ｐ_６に関し、▲３▼，▲４▼の式からのＬ_３，Ｌ_４をＬ_５，Ｌ_６とした時、▲５▼，▲６▼の式が得られる。
【００７５】
【数５】

したがって、▲１▼，▲２▼より距離は、
【数６】

：▲３▼，▲４▼より垂直方向の傾きは、
【数７】

で求められる。
【００７６】
さらに、鎖線で示すようにコーナ部に並行光線を発射する一対の測定光線発射体を加え、前部で８つの測定光線を発射することで、標定要素となる８つの輝点から、例えば、斜め方向へふった時の斜め方向のふれ角に基づく写真画像データおよびトンネル内の円弧状の写真画像データが得られるようにすることも可能となる。
【００７７】
したがって、このように構成された写真撮影装置１によれば、取手１５を両手に持つことで、安定した写真撮影が行なえるようになる。この時、取手１５を両手で持ったまま、何等支障なくレリーズ操作釦１９の操作が行なえるようになり、シャッタ操作が可能となる。この時、撮影場所、撮影アングルに何等制約されることなく写真撮影が行なえるようになる。この写真撮影時において、被調査面３には、２つの並行光線ａ−１と角度付光線ｂ−１による６つの輝点Ｐ１，Ｐ２，Ｐ３，Ｐ４，Ｐ５，Ｐ６が写し出され、それら各輝点Ｐ１，Ｐ２，Ｐ３，Ｐ４，Ｐ５，Ｐ６は、被調査面１までの距離ｘとふれ角θを求める標定要素がとり込まれた写真画像データとしてカメラ本体７内のフロッピー（登録商標）ディスク等に収められる。
【００７８】
一方、写真撮影して持ち帰った写真画像データが収められたフロッピー（登録商標）ディスク等を図外の写真画像解析装置にかけることで解析が行なわれる。この時、フロッピー（登録商標）ディスクによるデジタル処理によるため一貫した処理が可能になると共に真正面から正対する位置で写真撮影した時と同じ写真画像が迅速に得られる。
【００７９】
したがって、前記のようにして得られた写真画像の上下、左右の画像が一致するよう順次配列することで、正確な被調査面及び被調査面を含む全体構造の資料が容易に得られるようになる。
【００８０】
また、焦点距離が長くなる望遠レンズに交換することで、５５ｍｍのカメラレンズ３６の時と同じように被調査面３の拡大した標定要素を備えた写真画像データが得られるようになる。
【００８１】
次に、図１３の遠景の正規化処理を説明するフローチャートを用いて、本実施の形態を説明する。図１３に示すように、図１０（ａ）の画像ｅｉ（第１の遠景画像）が指定される毎に、その画像ｅｉの画像データをファイルから読み出し（Ｓ１）、指定された下段の格子ｎｉに移動させる遠景画像のモザイク位置の決定を行う（Ｓ２）。
【００８２】
次に、作業員が「００」の並び替えＯＫとするコマンドを入力すると、図１４に示すように、「００」の遠景画像（ポイント付き）が表示される。
【００８３】
この遠景画像のポイントＰｉ（Ｐ１、Ｐ２、…）を右回りでクリックして保存ボタン（図示せず）を選択する（Ｓ３）。
【００８４】
例えば、図１５（ａ）に示すように、画面における各点の座標Ｐ１…Ｐ６の座標が保存される。
【００８５】
そして、フレームに対するカメラの取り付け状態をチェックする。このチェックは、カメラ中心（フレーム中心と一致しているかどうかをチェックするもので誤差論から見て）が規定を満たしているかどうかを調べる。図１５（ｂ）においてはＰ１とＰ３とＰ４とＰ６とを用いてＸ軸、Ｙ軸の中心位置を調べる（カメラ中心がフレーム中心に来ていること）。この中心位置はＮＩＫＯＮＮ　Ｄ１では、数８に示すピクセル内におさまっていなければならない。
【００８６】
【数８】

次に、Ｐ２とＰ５とを用いて、縮尺の計算を行う。セル（カメラのフイルムの大きさ）上のＰ２とＰ５間の対向距離ｌ（Ｐ２とＰ５は対角に位置しており、ｘ，ｙ成分の変位置にそのまま、追随するため、数９の式で縮尺系数を求めることができる）は
【数９】

として計算する。
【００８７】
次に、左右角（ｘ方向）の傾きを計算する（Ｓ６）。
【００８８】
すなわち、図１７に示すように
【数１０】

とする。
【００８９】
この条件の元で、
ｍ＞ｎの場合は、
右回転＋θ１°
ｍ＝ｎの場合は、
回転なし
ｍ＜ｎの場合は、
左回転−θ１°
と判定する。
【００９０】
そして、角度θ１は、前述の　ｍ＞ｎの場合、　ｍ＝ｎの場合、　ｍ＜ｎの場合に応じて、図１６に示すように求める。
【００９１】
次に、上下角（ｙ方向の傾き）を図１８に示すように検出する（Ｓ６）。
【００９２】
【数１１】

とする。
【００９３】
この条件の元で、
ｕ＞ｖの場合は、
カメラ下向き−θ１°
ｕ＝ｖの場合は、
回転なし
ｕ＜ｖの場合は、
カメラ上向き＋θ１°
と判定する。
【００９４】
次に、ｘ方向、ｙ方向の画像補正を行う（Ｓ７、Ｓ８）。
【００９５】
例えば、図１９に示すように、撮影中心距離Ｌ、縮尺計数Ｒ、左右回転角±θ１、上下回転角±θ２、カメラの左右画角（半角）ｂ、カメラの上下画角（半角）ｃとして求められた場合において、原点を画像中心に対して左右にシフトした場合は、図１９の（ｂ）及び図２０に示すようにして求める。
【００９６】
図２１はｘ方向の傾きの補正を説明するための補正係数の求めた方の具体例を示した説明図であり、例えば近景＝０．９４
遠景＝１．０６
とした補正係数であり、図２２及び図２３はこれらの補正係数を基づいて画面のｘｙを補正していることを示している。
【００９７】
次に、指定画像がｎがマックスに到達したかどうかを判断し（Ｓ１０）、到達しない場合はステップＳ１に戻す。
【００９８】
また、指定画像がｎがマックスに到達した場合は、各画像の補正係数、縮尺データ等を保存して（Ｓ１１）、モザイクルーチンに処理を戻す。
【００９９】
このモザイクルーチン（モザイク化部）は、図２４に示すように、隣合う画像を画面に表示し（ｅ１とｅ２、ｅ３とｅ４）、同じとする基準点同士を指定して合成する。例えば、図２４においては、画像ｅ１（正規化）と画像ｅ２とに写し出されているＴｐ１、Ｔｐ３を指定して合成させる。
【０１００】
この合成はバンドル標定を用いている。このような処理を順次行って図２５のようなダムの１枚の画像を得る。
【０１０１】
そして、オペレータは近接画像の割付を行う。この近接画像の割付は、割付ボタンを選択することによって起動する。近接画像近接画像は、図２５に示す近接画像のサブ画面を開く。オペレータは、このサブ画面が遠景画像（モザイク画像）のどの部分に当たるかを把握しており、図２５に示す枠を指定して近接画像を割り付ける。
【０１０２】
この近接画像において、枠を指定すると、解析部が起動する。解析部は面積計算、線長計算、スケール計測等を備えており、図２６に示すようにひび割れの２点間を指定すると、ひび割れの幅を計算すると共に、所定の長さ単位のスケールを表示する。
【０１０３】
＜実施の形態２＞
図２７は実施の形態２のシステムの概略構成図である。図２７に示すように、本実施の形態はモバイルパソコン側にカメラで撮影した画像データを記憶したフラッシュメモリを格納する機能、ブラウザ機能、無線インターネット機能（ＰＨｃカード）等を備えると共に、環境設定部３０、画像登録部３３、画像割付データ送信部５１、解析指示部５２、無線インターネットの送受信部５３などを備え、モバイルパソコン５０で生成した遠景画像、近接画像、各種パラメータ等をインターネットを介してサービスサイトのサーバ５５に送信し、サービスサイトのサーバ５５が遠景画像の正規化、近接画像の正規化及びモザイク処理等を行って、その結果をモザイクの結果をモバイル側に送信し、ひび割れ解析等のコマンドを受信したときに、ひび割れを解析してモバイル側に送信するものである。
【０１０４】
すなわち、図２８のシーケンス図に示すように、現場側の作業員が環境設定部３０を用いてカメラパラメータ等をファイル３８に保存させ、かつ画像登録部３３を起動させて遠景画像と近接画像とをメモリ領域３２ａ、３２ｂに保存した後に並べ替部４０によって並べ替える。そして、各種パラメータと並べ替えファイル（図示せず）のデータとを送信させる（ｄ１、ｄ２）。
【０１０５】
次に、画像登録部３３を起動させて、近接画像を選択して、送受信部５３を用いて送信させる（ｄ３）。
【０１０６】
一方、サービスサイトのサーバ５５は、送受信部５７によって並べ替えのデータ（遠景）と近接画像とを受信し、上記と同様な遠景画像の正規化処理、近接画像の正規化処理を行ってメモリ領域（５８ａ、５８ｂ）に保存して、正規化された各遠景画像を用いてのモザイク処理を行う（ｄ４）。
【０１０７】
そして、モザイク画像（遠景画像）と、正規化された近接画像とを送受信部５７を用いてモバイルパソコン５０に送信する（ｄ５）。
【０１０８】
一方、モバイルパソコン５０側はサーバ５５からのモザイク画像をメモリ領域３２ｄ、正規化された近接画像をメモリ領域３２ｅに保存する。
【０１０９】
モバイル側の作業員は、画面にモザイク画像を表示した後に、正規化された近接画像を表示し、近接画像がモザイク画像のどのエリアのものかを割り付ける（割付部の処理）。
【０１１０】
この割付データを作業員はサーバ５５に送信する（ｄ６）。サーバ５５は、画像割付データファイル５６に保存する。
【０１１１】
すなわち、サーバ５５側は、モザイク画像のどのエリアにどのような近接画像が割り付けられたを把握している。
【０１１２】
一方、モバイルパソコン５０の作業員は、解析指示部５２を起動させて、ばひび割れ計測の指示（例えば幅）を行う（ｄ６）。このコマンドは無線インターネットでサーバ５５に送信され、サーバの解析部３７が指定の近接画像のひび割れ幅を計測し、スケールと共にモバイルパソコンに送信する（ｄ７）。
【０１１３】
従って、無線インターネットが繋がる箇所では、現場側で直ぐにひび割れ結果が判断できるので、より早く現場側で必要の処置を指示又は対処できる。
【０１１４】
なお、上記実施の形態では、ダムを例にして説明したが、ビル、橋の土台等のコンクリート構造物であってもよい。
【０１１５】
【発明の効果】
以上のように本発明のレーザ光点付き画像を用いた画像処理方法によれば、第１のレーザ発射体と第２のレーザ発射体とを備えた写真装置で得たレーザ光点付き画像の内で第２のレーザ発射体で得られた４点のピクセル座標に基づいて撮影時のカメラの中心を求め、この中心と４点間の距離から左右角、上下角を求めることができると共に、第１のレーザ発射体で得られた２点のピクセル座標（２点の距離はあまり変化しない）とから縮尺率が求めるので、自動的に正面から撮影した画像い容易に変換できるという効果が得られている。
【０１１６】
また、本発明のレーザ光点付き画像を用いたモザイク処理方法によれば、第１遠景画像を作業員が撮影したときの調査面に対応させて並べ替え、この並べ替えた第２遠景画像の１個毎に、正規化を行って合成した１枚の画像を得るので、正確に調査面のモザイク画像を得ることができるという効果が得られている。
【０１１７】
さらに、本発明のレーザ光点付き画像を用いた正規化サービスシステムによれば、サーバはが端末からのレーザ光点付き画像と前記端末の画面の第２のレーザ発射体の４点のピクセル座標と第１のレーザ発射体の２点のピクセル座標とに基づいて、左右角、上下角、縮尺率を求めて正規化して端末に送るので、現場側では小さい容量のパソコンでも正面から撮影させた画像を得ることが容易にできる。特に多数の遠景画像を端末側で撮影したときは、サーバ側から正規化されたモザイク画像が送信され、また近接画像を端末で割り付けたときは、その近接画像内のひび割れの数値等が送信されるので、現場側で調査面の全体像とひび割れ箇所の状況とを直ぐに判断できるという効果が得られている。
【図面の簡単な説明】
【図１】本実施の形態１のレーザ光点付き画像を用いたモザイク処理システムの概略構成図である。
【図２】建物、橋梁、道路、ダム等のコンクリート構造物等の被調査面１の写真を写す損壊写真撮影用カメラ２の概要斜視図を示している。
【図３】カメラ本体７を真正面からの起点の状況を説明する説明図である。
【図４】カメラ本体７を真正面からの起点の状況を説明する説明図である。
【図５】撮影の仕方を説明する説明図である。
【図６】メニュー画面の説明図である。
【図７】カメラパラメータの説明図である。
【図８】画像登録の説明図である。
【図９】画像の正規化の初期画面の説明図である。
【図１０】並べ替えの説明図である。
【図１１】ポイントＰｉ（Ｐ１、Ｐ２、…）付きの写真画像でダムを正面から撮影したときの画像に変換できる原理説明である。
【図１２】ポイントＰｉ（Ｐ１、Ｐ２、…）付きの写真画像でダムを正面から撮影したときの画像に変換できる原理説明である。
【図１３】本発明に係わる正規化処理を説明するフローチャートである。
【図１４】ポイントＰｉ（Ｐ１、Ｐ２、…）付きの写真画像の説明図である。
【図１５】ポイントＰｉ（Ｐ１、Ｐ２、…）付きの座標認識を説明する説明図である。
【図１６】縮尺、対物の算出を説明する説明図である。
【図１７】左右角の検出と振れ角の算出を説明する説明図である。
【図１８】上下角の検出を説明する説明図である。
【図１９】左右の画像の補正を説明する説明図である。
【図２０】上下の画像の補正を説明する説明図である。
【図２１】画像の正規化の説明図である。
【図２２】画像の正規化の説明図である。
【図２３】画像の正規化の説明図である。
【図２４】画像の合成を説明する説明図である。
【図２５】近接画像の割付を説明する説明図である。
【図２６】ひび割れ計測の説明図である。
【図２７】実施の形態２のシステム構成図である。
【図２８】実施の形態２のシーケンス図である。
【符号の説明】
３０　環境設定部
３３　画像登録部
３４　遠景画像正規化部
３５　モザイク化部
３６　近接画像正規化部
３７　画像割付部
３８　解析部
３２ａ　遠景用メモリ領域
３２ｃ　並替用メモリ領域
３２ｄ　遠景正規化用メモリ領域
３２ｅ　近接正規化用メモリ領域
３２ｆ　解析結果用メモリ領域[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention supports the periphery of a digital camera with a support frame, and emits parallel rays to an optical axis passing through the lens center at the upper corner of the left frame and the lower corner of the right frame of the support frame. A surface to be inspected by a photographic apparatus having a first laser emitter and a second laser emitter that emits a light beam at a predetermined angle outward with respect to the optical axis on the upper, lower, left and right frames of the support frame. The present invention relates to an image processing method using a laser light spotted image that is normalized using a laser light spotted image obtained by photographing the image.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, cracking of concrete structures has been mostly performed by the following two methods.
[0003]
(1) Preparation of survey data based on actual measurement (sketch of cracks in concrete structure, actual measurement of cracks)
(2) Three-dimensional measurement using photographs of the site, and subsequent creation of a plotting model (situation map of the structure)
On the other hand, in photography, conventionally, photography is performed by one camera with a parallel movement of 60% overlap (to obtain an orientation element), or photography with a fixed base length is performed by two cameras, and the film is developed. Three-dimensional data was created using a plotter, and cracks were analyzed.
[0004]
Further, with respect to the orientation element, in the conventional method, an orientation point (point serving as a position reference) is set at the time of photographing to be an orientation element, or an orientation parameter is set by calculating from a fixed base line length.
[0005]
However, as described above, damage investigations such as cracks have been carried out by on-site actual measurement methods or photographing measurement methods, but in the former case, a high degree of skill is required of workers, and cracks and the like are required. It is difficult to accurately sketch the position and width of each person without individual differences. Further, it takes a long time to work on site, which is not preferable in terms of workability. In addition, there is a problem that work equipment such as a scaffold is required in a place where a measurement point is high, and there is a danger in work.
[0006]
In the latter case, photo analysis techniques have advanced and photography measurement methods are becoming mainstream.However, even in this case, the size of the actual surveyed surface differs from that of the photographed image. It is necessary to copy a ruler such as a standard sword and a scale serving as an orientation element.
[0007]
This makes it possible to determine the scale of the photographic image from the scale that has been captured and to analyze numerical values such as the length and width of the crack based on the scale. Since the work of sticking to the survey surface is indispensable, for example, when the survey surface is at a high position, a high scaffold for setting the scale is required in the same manner as the on-site actual measurement method.
[0008]
In order to solve this problem, for example, it is possible to reduce the size of the camera so that it can be held in a hand, and to face the camera directly from the front without being limited by the conditions such as the shooting location and the shooting angle (angle of shooting). There is a destructive photographing apparatus for a concrete structure capable of obtaining photographic data having orientation elements for obtaining the same accurate photographic image.
[0009]
[Problems to be solved by the invention]
However, at present, it is necessary to manually measure the distance between several laser light spots projected on the image, obtain the horizontal angle, vertical angle, objective distance, scale, etc., and normalize the image with these parameters. Was most.
[0010]
The present invention obtains the distance between several laser light points automatically projected on an image, the right and left angle, the vertical angle, the object distance, the scale, and the like, and automatically normalizes with these parameters, concrete structure And a system that automatically informs the status of the damaged part by numerical values.
[0011]
[Means for Solving the Problems]
An image processing method using an image with laser light spots according to the present invention includes supporting the periphery of a digital camera with a support frame, and applying the digital image to a corner above a left frame and a corner below a right frame of the support frame. A first laser projecting body that emits a parallel light beam with respect to an optical axis passing through the center of the lens of the camera is provided, and light beams having a predetermined angle outward with respect to the optical axis line are provided on upper, lower, left and right frames of the support frame. This is an image processing method using an image with laser light spots, in which a surface to be inspected is photographed by a photographic device provided with each of the second laser projectiles to be emitted, and normalized using the image with laser light spots.
[0012]
A computer captures the image with the laser light spot taken by the photographic device, and displays the image on the screen.Each time there is a designation of the position of the laser light spot in the image with the laser light spot displayed on the screen Recognizing the pixel coordinates on the screen, and determining a center based on the four pixel coordinates obtained by the second laser projectile; and an x-axis from the center to the four pixel coordinates. a step of obtaining distances on the y-axis and obtaining the vertical and horizontal shake angles of the digital camera from the ratio of these distances; and the pixel coordinates of two points obtained by the first laser projectile. Obtaining an object distance between the film of the digital camera and the surface to be inspected based on the obtained camera parameters, and obtaining a scale factor of an image on the film from the object distance; Vertical, is summarized in that comprising the step of normalizing the image upon shooting laser spot with the image on the screen from the front with the left and right deflection angle and the scale factor.
[0013]
A mosaic processing method using an image with laser light spots according to the present invention, the supporting frame supports the periphery of the digital camera, the upper corner of the left frame and the lower corner of the right frame of the digital, the digital A first laser projecting body that emits a parallel light beam with respect to an optical axis passing through the center of the lens of the camera is provided, and light beams having a predetermined angle outward with respect to the optical axis line are provided on upper, lower, left and right frames of the support frame. This is a mosaic processing method using an image with laser light spots, in which a to-be-examined surface is photographed by a photographic device equipped with each of the second laser projectiles to be emitted, and a mosaic image is generated using the image with laser light spots.
[0014]
The computer reads a large number of first distant images obtained by sequentially photographing predetermined areas adjacent to each other on the inspected surface at an arbitrary inclination while keeping the focal length of the digital camera constant with respect to the inspected surface. And displaying a photographing place piece table in which the surveyed face is divided by a mesh at a fixed interval and a number of the surveyed face is assigned, and a piece of the photographing place piece table is designated, and the When a distant view image among a large number of first distant view images displayed on the screen is designated, the designated second distant view image is moved to the designated piece to determine where the distant view image on the survey plane is located. And displaying one distant view image from the plurality of first distant view images on a screen, and specifying all the laser light spots in the distant view image in a predetermined direction; Pixel coordinates of A step of recognizing and obtaining a center based on the vertical and horizontal coordinates of the four points of the second laser projectile; obtaining the vertical and horizontal distances of the four points from the center; Determining the right and left angles, the object coordinates on the film of the digital camera and the object surface based on the pixel coordinates of the two points obtained by the first laser projectile and the camera parameters input in advance. Determining a distance, and determining a scale of the image on the film from the objective distance and the ratio; and an aspect ratio of one distant view image on the screen based on the scale, the object distance, and the horizontal and vertical angles. To obtain one distant image that assumes that the object to be investigated is photographed from the front, and sequentially displays adjacent distant images, and designates the same spot on these images for each display. Wherein performing bundle orientation is summarized in that comprising the steps of obtaining a mosaic image which is an object to be investigated surface entire image.
[0015]
The normalization service system using the image with laser light spots of the present invention supports the periphery of the digital camera with a support frame, and the upper frame and the lower corner of the right frame of the support frame are A first laser projecting body that emits a parallel ray with respect to an optical axis passing through the center of the lens of the digital camera is provided, and light beams having a predetermined angle outward with respect to the optical axis are provided on upper, lower, left and right frames of the support frame. The image with the laser light spot obtained by photographing the surface to be inspected by a photographic device equipped with a second laser projectile that emits light is stored in a terminal, and is normalized by communicating with the terminal via a computer network system. This is a service system using an image with a laser light spot, which provides the obtained distant view image and proximity image from a server.
[0016]
The server, for the terminal, the image with the laser light spot photographed by the photographing device, the pixel coordinates of four points of the second laser projectile on the screen of the terminal, and the two points of the first laser projectile. Means for transmitting the pixel coordinates of a point; receiving the image with the laser light spot and the pixel coordinates; obtaining a center based on the vertical and horizontal coordinates of the four points on the image; Means for determining the vertical and horizontal distances, the vertical and horizontal angles of the photographing apparatus based on the ratio of these distances, pixel coordinates of two points obtained by the first laser projectile, and camera parameters input in advance Means for obtaining an object distance between the film of the digital camera and the surface to be inspected on the basis of the above, a means for obtaining a scale of an image on the film from the object distance and the ratio, the scale, the object distance, and the right and left angles. Means for correcting the aspect ratio of one distant view image on the screen based on the vertical angle and the vertical angle to obtain a single normalized distant view image that assumes that the subject is photographed from the front. Means for providing one distant view image to the terminal;
The gist is that it is provided.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
<Embodiment 1>
FIG. 1 is a schematic configuration diagram of a mosaic processing system using an image with laser light spots according to the first embodiment. Before describing the configuration of each part of the present embodiment, a camera (hereinafter referred to as a camera for taking a damaged image) used in this system will be described.
[0018]
FIG. 2 is a schematic perspective view of a camera 2 for taking a damaged photograph for photographing a surface 1 to be inspected, such as a concrete structure such as a building, a bridge, a road, or a dam.
[0019]
The damaged photographing camera 2 includes a camera body 7 (digital camera with a built-in flash memory) set in a support frame 5 framed in a rectangular shape, and six orientation elements on a surface 1 to be inspected such as a concrete structure. It has first laser emitters 9, 10, 11, 12 for projecting bright points P1, P2, P3, P4, P5, P6 and second laser emitters 13, 14. Further, even if the surface 1 to be inspected is curved (for example, a tunnel), the third laser beam emitter 15 for projecting the bright spots P5 and P6 in order to correct the image as viewed from the front and to measure the cracks. , 16 are provided. In addition, a total of eight bright points P1, P2, P3, P4, P5, P6, P7, and P8 may be obtained so that even if the surface 1 to be inspected is curved, it can be corrected to an image viewed directly from the front. . In the present embodiment, a description will be given assuming that there are six.
[0020]
The support frame 5 is made of aluminum to reduce the weight, and is integrally formed by cutting a single thick plate into a rectangular shape. It has a frame structure with.
[0021]
The support frame 5 is provided with a screw hole (not shown) at the bottom that can be set on a tripod or the like, while a handle 15 is provided on both left and right sides so that it can be held by both hands.
[0022]
One of the handles 15 is provided with a measuring beam operation switch 17 for turning on / off the first laser projecting bodies 9, 10, 11, 12 and the second laser projecting bodies 13, 14. When the operation switch 17 is turned on, the measurement light beam (laser) is continuously emitted from the first laser projectiles 9, 10, 11, 12 and the second laser projectiles 13, 14. ing.
[0023]
Further, a release operation button 19 for controlling the operation of the shutter of the camera body 7 is provided on the other side of the handle 15 so that a shutter operation can be performed while holding the handle 15 with both hands.
[0024]
The camera body 7 converts an image into an electric signal using an image pickup device such as a CCD (Charge Coupled Device) instead of a film, digitizes the data, and converts the data into, for example, a semiconductor memory, a floppy (registered trademark) disk, or a flash. It is a digital camera that records photographic data in a memory or the like.
[0025]
Note that the camera body 7 is not necessarily a digital type, and may be an optical camera using a normal film. In this case, the film is read by a scanner and processed. The bottom of the camera body 7 is fixedly supported on the support frame 5 via a camera body setting member 21.
[0026]
In addition, the first laser projectiles 9, 10, 11, 12 and the second laser projectiles 13, 14, and the third laser projectiles 15, 16 supply currents inside the projectile mounting cases 23a... 23h. A semiconductor element (not shown) that outputs a laser beam from the case window hole by flowing is provided. A current is supplied to the semiconductor element from a power supply unit such as a battery disposed above the support frame 5 by turning on the measurement light operation switch 17 described above.
[0027]
Further, the first laser light emitters 9 and 10 are for emitting a set of parallel light rays for emitting measurement light rays in parallel to the optical axis Y of the camera lens 36. The mounting positions are diagonally arranged at the corners of the support frame 5, and the projectile mounting case is directly and detachably fixed and supported by case fixing screws.
[0028]
The first laser projecting bodies 11 and 12 and the second laser projecting bodies 13 and 14 are separated from each other by a predetermined angle (including an angle 0) with respect to the optical axis Y of the camera lens 36 (left and right angles and vertical angles are separated). B) for emitting a set of angled rays that emit angled rays.
[0029]
The first laser projectiles 11 and 12 are for an angled light beam in the vertical direction, and the second laser projectiles 13 and 14 are for an angled light beam in the left and right direction.
[0030]
Further, the laser projecting bodies 11 and 12 are arranged to face the support frame 5 at the up and down positions so as to pass through the optical axis Y of the camera lens 36, while the second laser reflectors 13 and 14 are arranged on the supporting frame 5 , Which are disposed opposite to each other at right and left positions so as to pass through the optical axis Y of the camera lens 36, and are fixed and supported via mounting members.
[0031]
The angle of the above-mentioned angled light beam differs depending on the camera lens focal length f. In this embodiment, when the camera lens focal length f is 55 mm, the angle of the angle light beam is 3 degrees.
[0032]
The projectile mounting cases 23a that emit angled light beams are, for example, for 5 degrees and 7 degrees with respect to different camera lens focal lengths f when a telephoto lens or the like that enlarges and photographs the surface to be inspected 1 is used. It is a dedicated unit whose angle is set in advance, such as for use.
[0033]
The parallel light beam a-1 emitted from the first measurement light beam emitters 9 and 10 and the angled light beam b-1 emitted from the second measurement light beam emitters 11, 12, 13, and 14 are emitted from the camera body 7 The four bright points P3, P4, P5, and P6 are displaced by the vertical and horizontal deflection angles θ.
[0034]
FIG. 3 shows a state in which the camera body 7 is moved upward, for example, from a position directly in front of the camera body 7, and the bright points P3 and P4 of the angled light beam b-1 are changed from a solid line (black point) to a dotted line (white point). ), And the camera body 7 can obtain photographic image data corresponding to the swing angle.
[0035]
Further, as shown in FIG. 4, by moving the shake angle to the left by a predetermined angle from the zero state, the bright spots P5 and P6 are projected from the solid line to the position shown by the dotted line, and the camera body 7 has the shake angle. Photo image data at the time of θ can be obtained.
[0036]
P1 and P2 are two bright spots by the parallel ray a-1, P3, P4, P5 and P6 are bright spots by the angled ray b-1, and the bright spots P3, P4 and P5 at the deflection angle θ. It can be seen that P6 (white point) is much wider than between the two bright points (black points) when the deflection angle is zero. Therefore, based on the distance ratio from 01 between the two open points to the photographic image analysis device (not shown), the deflection angle θ is obtained. Therefore, by performing arithmetic processing based on the deflection angle θ, it is possible to correct a normal image, and the same photographic image data as when facing directly from the front can be obtained.
[0037]
As shown in FIG. 5A, such a damaged photographing camera 2 may be rotated only in the horizontal direction at the same position to sequentially photograph a desired area, or FIG. As shown in FIG. 5, the image may be taken by rotating only in the vertical direction, or as shown in FIG. 5C, the image may be taken by rotating only in the vertical and horizontal directions. Alternatively, the operator may hold and hold the hand and sequentially photograph desired areas only in the horizontal, vertical, and vertical and horizontal directions while moving. These are collectively referred to as distant view images. Further, an area where there is a change such as breakage or crack may be photographed in close proximity (referred to as a proximity image).
[0038]
Next, analysis processing of data photographed by the damaged photographing camera 2 as described above using the configuration of FIG. 1 will be described.
[0039]
In the present embodiment, it is assumed that a large number of captured images (with bright spots) of the inspected surface 1 captured by the damaged photographing camera 2 are already stored in the database. In addition, the operator knows what time the captured image is. Further, the operator prepares a drawing or the like in which the surveyed surface 1 is divided into meshes and assigned numbers.
[0040]
As shown in FIG. 1, a mosaic processing / crack measurement system 29 (personal computer) for an image with a laser beam includes an environment setting unit 30, an image registration unit 33, a distant view image normalization unit 34, a mosaic generation unit 35, A proximity image normalizing unit 36, an image allocating unit 37, an analyzing unit 38, a rearranging unit 40, and the like are provided.
[0041]
The environment setting unit 30 displays a screen (not shown) for specifying a location of the image data captured by the damaged photo capturing camera 2, and displays the specified location (designation of the directory of the input file) input on this screen. ) Is saved in the file 31.
[0042]
Further, the environment setting unit 30 displays a screen (not shown) for designating an area for saving normalized data, analysis data, and the like, which will be described later, and stores the saved area (output file) input on this screen. Directory 32), the memory area for distant view 32a, the memory area for proximity 32b, the memory area for rearrangement 32c, the memory area for distant view normalization 32d, the memory area for proximity normalization 32e, and the memory area for analysis result 32f , Mosaic memory area 32g, etc.).
[0043]
Further, the environment setting unit 30 displays screens for setting camera parameters, survey plane parameters, and laser beam parameters, and saves data of these screens in the file 39. The above-mentioned camera parameters include a camera type, a cell size (film size), a focal length, and the like.
[0044]
Further, the survey surface parameters include the dimensions, coordinates (xyz), and the like of the dam. The laser beam parameters include a beam angle value, a camera type, and the like.
[0045]
The image registration unit 33 displays a screen for distributing the photographic image data stored in the input file 31 into a distant view image and a close-up image. To save.
[0046]
The reordering unit 40 displays all the distant view images stored in the distant view image memory area 32a on the screen, and divides the survey surface 1 into meshes and assigns numbers to the screens. Is displayed, and every time the frame and the distant view image are designated, the distant view image is allocated to the frame. Then, the result is stored in the rearrangement memory area 32c.
[0047]
Each time the distant view image data for normalization is designated (using the sorted distant view image data), the distant view image normalizing unit 34 recognizes the specified point Pi on the screen, and The center of the film of the camera, the horizontal rotation angle, the vertical rotation angle, the scale count, and the projection center distance of the film of the camera are obtained from the coordinate values of each point, and normalization for correcting the image as viewed from the front is performed based on these information. . Then, the normalized distant view image is stored in the distant view normalization memory area 32d.
[0048]
The mosaic unit 35 combines the normalized distant images stored in the distant view normalization memory 32d with each other to obtain one image of the entire dam (hereinafter, referred to as a mosaic image). Then, the mosaic image is stored in the mosaic image memory area 32g.
[0049]
Each time the proximity image data for normalization is designated, the proximity image normalization unit 36 recognizes the designated points Pi on the screen, and, based on the coordinate values of these points, the camera film center, The left-right rotation angle, the up-down rotation angle, the scale count, and the projection center distance of the film are obtained, and normalization for correcting an image viewed from the front is performed based on these information. Then, the normalized proximity image is stored in the proximity normalization memory area 32e.
[0050]
The image allocating unit 38 displays the proximity image stored in the proximity normalization memory area 32e, and displays a frame (defined by coordinates) specified in the mosaic image and the proximity image (displayed proximity) in the proximity normalization memory 32e. (Image: normalized).
[0051]
The analysis unit 38 recognizes a crack location (determined based on density or brightness) of the proximity image (displayed proximity image: normalized), measures the area of the crack portion, the line length of the crack, the crack width, and the like. Are measured, and these results are stored in the measurement result memory area 32f.
[0052]
(Operation explanation)
The operation of the mosaic processing / crack measurement system using the image with laser light spots configured as described above will be described below.
[0053]
First, the computer displays a menu screen shown in FIG. 6 according to an instruction of the worker. This menu screen includes an environment setting button Aa, a digital camera image registration button Ab, a distant view image normalization button Ac, a close image normalization button Ad, an image allocation button Ae, an analysis button Af, and the like.
[0054]
Next, the worker selects the environment setting button Aa to activate the environment setting unit. This operation displays a screen (not shown) for determining where the image of the digital camera is located and where the processed image and conditions are stored. In this way, a large number of camera images are stored (data area of m rows and n columns).
[0055]
In this embodiment, as shown in FIG. 5A, the camera is rotated only in the horizontal direction at the same position using the damaged photographing camera 2, and the desired areas are sequentially photographed and stored in the file 31. It is also assumed that the output file 32 has been created.
[0056]
Also, as shown in FIG. 7, the environment setting unit 30 includes the type of camera, the size / pixel of the cell (the size of the camera), the size (area) of the cell, the size of one pixel of the CCD, and the focal length. , A screen for inputting camera parameters such as right / left angle / half angle and up / down angle / half angle.
[0057]
Then, the user selects the digital camera image registration button Ab. When the digital camera image registration button is selected, the image registration unit 33 is activated, and as shown in FIG. 7, a folder name area Ba indicating a storage location of the digital camera image, a selected image area Bb of the selected folder, and a distant view image designation area A registered image setting screen including Bc and a proximity image designation area Bd is displayed.
[0058]
In the registered image setting screen, the distant view image designation area Bc and the close image designation area Bd shown in FIG. 8 are composed of a shooting position, a date and time, and a film number.
[0059]
In other words, the image of the registered folder (DSC_0005.jpg, DSC_0006.jpg,...) Can be determined so that it is the number of the film, and it can be determined later whether the worker has taken the image in the near or distant view. I have.
[0060]
The images of the folder are displayed in the area Bb by sequentially selecting the above-described folders, and the images are sorted into the distant view or the close view (the shooting position and the date and time are set in the distant view image specification area Bc or the close image specification area Bd). And the film number).
[0061]
Then, for example, when the distant view normalization button Bf is selected, a preparation screen for distant view image normalization shown in FIG. 9A is displayed.
[0062]
On this preparation screen, a dial log of the position and a dial log of the attribute information are displayed. At the position 246, there are four images of the dam taken on February 5, 2002, which are not processed (normalized). No).
[0063]
When the normalization process Ca is selected on this screen, a dial log shown in FIG. 9B is displayed. The operator selects the rotation rearrangement button Da. The rearranged image shown in FIG. 10 is displayed according to the selection of the rotation rearrangement button Da. This image displays the four unprocessed images e1, e2, e3, and e4 of FIG. 10A in the upper row, and displays a rearrangement screen in the lower row. This rearrangement screen corresponds to a drawing in which the surveyed surface 1 is divided into meshes and numbers are assigned.
[0064]
Next, the worker determines which of the lower-level grids (frames) each image ei (also referred to as a first distant view image) in the upper row corresponds to, and the corresponding grid ni (n1, n2,...) Into the image ei. This is referred to as determining the mosaic position of the distant view image.
[0065]
That is, the rearrangement unit 40 displays all the first distant images stored in the distant image memory area 32a on the screen, and divides the surveyed surface 1 into meshes and assigns numbers to the drawings. Is displayed, and each time a frame and a distant view image are designated, a distant view image is allocated to the frame. Then, the result is stored in the rearrangement memory area 32c.
[0066]
Then, when the operator inputs a command to rearrange “00”, the image of “00” (with points) is displayed as shown in FIG.
[0067]
Click on the point Pi (P1, P2,...) Clockwise in this image and select a save button (not shown). In this selection, the computer recognizes each point Pi and performs a normalization process (distant view) described below.
[0068]
Here, before the description of the normalization processing, the principle of converting the above-mentioned photographic image with the points Pi (P1, P2,...) Into an image when the dam is photographed from the front will be described with reference to FIGS. 11 and 12. explain.
[0069]
11 and FIG. 12, FIG. 11 shows the inspection surface 3 where the deflection angle θ of the camera body 7 is zero when the camera body 7 is photographed directly from the front, with a dotted line, and when the deflection angle of the camera body 7 is θ. FIG. 11 shows, for the sake of explanation, the surveyed surface 3 as a solid line with an inclination. FIG. 12 shows the camera body 7 with an inclination of θ.
[0070]
Here, a-1 is a parallel ray, b-1 is an angled ray having α, f is a focal length of a camera lens, a is a distance between measurement rays, x is a distance to a surface to be inspected 3, and 01 is a camera. In FIG. 11, when the lens center is set, L1, L2, L3, L4, L5, and L6 obtained when the deflection angle obtained on the image plane 59 is zero and when the predetermined angle is shifted are θ.
(Equation 1)

Equations (1), (2), (3), (4), (5), and (6) hold.
[0071]
Here, (1) and (2) indicate parallel ray components, (3) and (4) indicate up-down angled ray components, (5) and (6) indicate left-right angled ray components,
From these formulas (1), (2), (3), (4), (5), and (6), the deflection angle tan θ and the distance x to the surface to be inspected 3 are obtained.
[0072]
(Equation 2)

On the other hand, in FIG. 7, when the θ angle camera body 7 is shaken when 01 to 02: focal length f, 02 to 03: measurement distance X distance between laser beams, 01 to a′1: L1, 01 to a'2: L2, 01 to b'1: L3 (L5), 01 to b'2: L4 (L6), (linear beam) P ₁ , P ₂ , The following equations (1) and (2) are obtained.
[0073]
[Equation 3]

Vertical angle beam P ₃ , P ₄ With respect to, the expressions of (3) and (4) are obtained.
[0074]
(Equation 4)

Angled beam P in horizontal axis direction ₅ , P ₆ , From the formulas (3) and (4), L ₃ , L ₄ To L ₅ , L ₆ Then, the expressions (5) and (6) are obtained.
[0075]
(Equation 5)

Therefore, the distance from (1) and (2) is
(Equation 6)

: From (3) and (4), the inclination in the vertical direction is
(Equation 7)

Is required.
[0076]
Further, a pair of measurement beam emitters that emit parallel rays at the corners as shown by the dashed lines are added, and eight measurement rays are emitted at the front part. It is also possible to obtain photographic image data based on a tilt angle in a diagonal direction when the player shakes in the direction and arc-shaped photographic image data in the tunnel.
[0077]
Therefore, according to the photographing apparatus 1 configured as described above, by holding the handle 15 in both hands, stable photographing can be performed. At this time, the release operation button 19 can be operated without any trouble while holding the handle 15 with both hands, and the shutter operation can be performed. At this time, the photographing can be performed without being restricted by the photographing place and the photographing angle. At the time of this photographing, six luminous points P1, P2, P3, P4, P5, and P6 due to two parallel light rays a-1 and angled light rays b-1 are projected on the inspected surface 3, and the respective luminous points are shown. The points P1, P2, P3, P4, P5, and P6 are the floppy (registered trademark) disks in the camera body 7 as photographic image data in which the orientation elements for obtaining the distance x and the deflection angle θ to the surface to be inspected 1 are captured. And so on.
[0078]
On the other hand, analysis is performed by applying a floppy (registered trademark) disk or the like containing photographic image data brought back by taking a photograph to a photographic image analysis device (not shown). At this time, since the digital processing is performed by the floppy (registered trademark) disk, the processing can be performed consistently, and the same photographic image as when a photograph is taken at a position facing directly from the front can be obtained quickly.
[0079]
Therefore, by sequentially arranging the upper and lower images and the left and right images of the photographic image obtained as described above so as to coincide with each other, it is possible to easily obtain a material of the entire structure including the accurate inspected surface and the inspected surface. Become.
[0080]
Further, by replacing the telephoto lens with a longer focal length, photographic image data having an enlarged orientation element of the inspection target surface 3 can be obtained as in the case of the camera lens 36 of 55 mm.
[0081]
Next, the present embodiment will be described with reference to the flowchart of FIG. As shown in FIG. 13, every time the image ei (first distant image) of FIG. 10A is designated, the image data of the image ei is read from the file (S1), and the designated lower grid ni is designated. The mosaic position of the distant view image to be moved to is determined (S2).
[0082]
Next, when the operator inputs a command to rearrange “00” to be OK, as shown in FIG. 14, a distant view image (with points) of “00” is displayed.
[0083]
Clicking the point Pi (P1, P2,...) Clockwise in this distant view image and selecting a save button (not shown) (S3).
[0084]
For example, as shown in FIG. 15A, coordinates P1 to P6 of each point on the screen are stored.
[0085]
Then, the state of attachment of the camera to the frame is checked. This check checks whether the camera center (checks whether it is coincident with the frame center and from the viewpoint of error theory) satisfies the rules. In FIG. 15B, the center positions of the X axis and the Y axis are checked using P1, P3, P4, and P6 (the center of the camera is located at the center of the frame). In the NIKONN D1, this center position must fall within the pixel shown in Expression 8.
[0086]
(Equation 8)

Next, the scale is calculated using P2 and P5. The opposing distance 1 between P2 and P5 on the cell (the size of the camera film) (P2 and P5 are located diagonally and follow the change position of the x and y components as they are. Can be used to calculate the scale factor)
(Equation 9)

Is calculated as
[0087]
Next, the inclination of the left and right angles (x direction) is calculated (S6).
[0088]
That is, as shown in FIG.
(Equation 10)

And
[0089]
Under this condition,
If m> n,
Clockwise rotation + θ1 °
If m = n,
No rotation
If m <n,
Left rotation -θ1 °
Is determined.
[0090]
Then, the angle θ1 is obtained as shown in FIG. 16 according to the above-described cases where m> n, m = n, and m <n.
[0091]
Next, the vertical angle (inclination in the y direction) is detected as shown in FIG. 18 (S6).
[0092]
[Equation 11]

And
[0093]
Under this condition,
If u> v,
Camera downward -θ1 °
If u = v,
No rotation
If u <v,
Camera upward + θ1 °
Is determined.
[0094]
Next, image correction in the x and y directions is performed (S7, S8).
[0095]
For example, as shown in FIG. 19, as the photographing center distance L, the scale factor R, the left / right rotation angle ± θ1, the up / down rotation angle ± θ2, the left / right angle of view (half angle) b, and the up / down angle of view (half angle) c of the camera If the origin is shifted to the left or right with respect to the center of the image in the obtained case, it is obtained as shown in FIG. 19B and FIG.
[0096]
FIG. 21 is an explanatory diagram showing a specific example of a method of obtaining a correction coefficient for explaining the correction of the inclination in the x direction. For example, foreground = 0.94
Distant view = 1.06
22 and 23 show that xy of the screen is corrected based on these correction coefficients.
[0097]
Next, it is determined whether or not n of the designated image has reached Max (S10). If not, the process returns to step S1.
[0098]
When n of the designated image reaches the maximum, the correction coefficient and the scale data of each image are stored (S11), and the process returns to the mosaic routine.
[0099]
As shown in FIG. 24, the mosaic routine (mosaicking unit) displays adjacent images on the screen (e1 and e2, e3 and e4), and specifies and combines the same reference points. For example, in FIG. 24, Tp1 and Tp3 appearing in the image e1 (normalized) and the image e2 are designated and combined.
[0100]
This composition uses bundle orientation. By sequentially performing such processing, one image of the dam as shown in FIG. 25 is obtained.
[0101]
Then, the operator assigns a proximity image. The allocation of the proximity image is activated by selecting an allocation button. The close-up image opens a sub-screen of the close-up image shown in FIG. The operator knows which part of the distant view image (mosaic image) this sub-screen corresponds to, and assigns a nearby image by designating a frame shown in FIG.
[0102]
When a frame is designated in this proximity image, the analysis unit starts. The analysis unit is provided with area calculation, line length calculation, scale measurement, etc. When specifying between two points of a crack as shown in FIG. 26, the width of the crack is calculated and the scale of a predetermined length unit is displayed. I do.
[0103]
<Embodiment 2>
FIG. 27 is a schematic configuration diagram of a system according to the second embodiment. As shown in FIG. 27, the present embodiment has a function of storing a flash memory storing image data photographed by a camera on the mobile personal computer side, a browser function, a wireless Internet function (PHc card), and the like. 30, an image registration unit 33, an image allocation data transmission unit 51, an analysis instruction unit 52, a wireless Internet transmission / reception unit 53, etc., and provide a distant view image, a proximity image, various parameters, etc. generated by the mobile personal computer 50 via the Internet. The server 55 of the service site transmits the result of normalization of the distant view image, the normalization of the close image, the mosaic processing, and the like to the server 55 of the service site. When a command is received, cracks are analyzed and transmitted to the mobile side.
[0104]
That is, as shown in the sequence diagram of FIG. 28, the worker at the site uses the environment setting unit 30 to save the camera parameters and the like in the file 38 and activates the image registration unit 33 to store the distant view image and the close-up image. Are stored in the memory areas 32a and 32b, and are rearranged by the rearranging unit 40. Then, various parameters and data of a rearrangement file (not shown) are transmitted (d1, d2).
[0105]
Next, the image registration unit 33 is activated, a nearby image is selected, and transmitted using the transmission / reception unit 53 (d3).
[0106]
On the other hand, the server 55 of the service site receives the rearranged data (distant view) and the close-up image by the transmission / reception unit 57, and performs the same normalization processing of the distant view image and the close-up image as described above to perform the memory area (58a, 58b), and a mosaic process is performed using each normalized distant view image (d4).
[0107]
Then, the mosaic image (distant view image) and the normalized proximity image are transmitted to the mobile personal computer 50 using the transmission / reception unit 57 (d5).
[0108]
On the other hand, the mobile personal computer 50 stores the mosaic image from the server 55 in the memory area 32d and the normalized proximity image in the memory area 32e.
[0109]
After displaying the mosaic image on the screen, the worker on the mobile side displays the normalized proximity image, and allocates which area of the mosaic image is the proximity image (processing of the allocation unit).
[0110]
The worker transmits this allocation data to the server 55 (d6). The server 55 stores the data in the image allocation data file 56.
[0111]
In other words, the server 55 knows which area of the mosaic image has been assigned with which proximity image.
[0112]
On the other hand, the worker of the mobile personal computer 50 activates the analysis instructing unit 52 and gives an instruction (for example, width) for measuring the crack (d6). This command is transmitted to the server 55 via the wireless Internet, and the analysis unit 37 of the server measures the width of the crack in the specified proximity image, and transmits it along with the scale to the mobile personal computer (d7).
[0113]
Therefore, in a place where the wireless Internet is connected, the result of the crack can be immediately judged on the site side, and the necessary treatment can be instructed or dealt with more quickly on the site side.
[0114]
In the above embodiment, the dam is described as an example, but a concrete structure such as a building or a bridge base may be used.
[0115]
【The invention's effect】
As described above, according to the image processing method using an image with a laser light spot of the present invention, an image with a laser light spot obtained by a photographic device including the first laser projectile and the second laser projectile is obtained. The center of the camera at the time of shooting is obtained based on the pixel coordinates of the four points obtained by the second laser projectile within the camera, and the horizontal and vertical angles can be obtained from the distance between the center and the four points. Since the scale factor is obtained from the pixel coordinates of the two points obtained by the first laser projectile (the distance between the two points does not change much), the effect that the image taken automatically from the front can be easily converted can be obtained. Have been.
[0116]
Further, according to the mosaic processing method using the image with laser light spots of the present invention, the first distant images are rearranged so as to correspond to the survey plane when the operator took the images, and the rearranged second distant images are rearranged. Since one image synthesized by normalization is obtained for each image, an effect that a mosaic image of the survey surface can be obtained accurately is obtained.
[0117]
Further, according to the normalized service system using the image with laser light spots of the present invention, the server can store the image with the laser light spot from the terminal and the pixel coordinates of four points of the second laser projectile on the screen of the terminal. The left and right angles, the up and down angles, and the scale are calculated and normalized based on the pixel coordinates of the two points of the first laser projectile and sent to the terminal. Images can be easily obtained. In particular, when a large number of distant images are photographed on the terminal side, a normalized mosaic image is transmitted from the server side, and when a close-up image is allocated by the terminal, a numerical value of a crack in the close-up image is transmitted. Therefore, the effect is obtained that the overall image of the investigation surface and the state of the cracked portion can be immediately determined on the site side.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a mosaic processing system using an image with laser light spots according to a first embodiment.
FIG. 2 is a schematic perspective view of a damaged photographing camera 2 for photographing a surface 1 to be inspected such as a concrete structure such as a building, a bridge, a road, or a dam.
FIG. 3 is an explanatory diagram illustrating a situation where a camera body 7 is started from directly in front.
FIG. 4 is an explanatory diagram illustrating a situation where a camera body 7 is started from directly in front.
FIG. 5 is an explanatory diagram illustrating a method of photographing.
FIG. 6 is an explanatory diagram of a menu screen.
FIG. 7 is an explanatory diagram of camera parameters.
FIG. 8 is an explanatory diagram of image registration.
FIG. 9 is an explanatory diagram of an initial screen for normalizing an image.
FIG. 10 is an explanatory diagram of rearrangement.
FIG. 11 is a diagram illustrating the principle of converting a photographic image with points Pi (P1, P2,...) Into an image of a dam taken from the front.
FIG. 12 is a principle explanation that a photographic image with points Pi (P1, P2,...) Can be converted into an image of a dam taken from the front.
FIG. 13 is a flowchart illustrating a normalization process according to the present invention.
FIG. 14 is an explanatory diagram of a photographic image with points Pi (P1, P2,...).
FIG. 15 is an explanatory diagram illustrating coordinate recognition with a point Pi (P1, P2,...).
FIG. 16 is an explanatory diagram illustrating calculation of a scale and an objective.
FIG. 17 is an explanatory diagram illustrating detection of a left-right angle and calculation of a shake angle.
FIG. 18 is an explanatory diagram illustrating detection of a vertical angle.
FIG. 19 is an explanatory diagram illustrating correction of left and right images.
FIG. 20 is an explanatory diagram illustrating correction of upper and lower images.
FIG. 21 is an explanatory diagram of image normalization.
FIG. 22 is an explanatory diagram of image normalization.
FIG. 23 is an explanatory diagram of image normalization.
FIG. 24 is an explanatory diagram illustrating image composition.
FIG. 25 is an explanatory diagram illustrating allocation of a proximity image.
FIG. 26 is an explanatory diagram of crack measurement.
FIG. 27 is a system configuration diagram of the second embodiment.
FIG. 28 is a sequence diagram of the second embodiment.
[Explanation of symbols]
30 Environment setting section
33 Image registration section
34 Distant image normalization unit
35 Mosaic Department
36 Proximity image normalization unit
37 Image layout section
38 Analysis unit
32a distant view memory area
32c Memory area for sorting
32d distant view normalization memory area
32e Memory area for proximity normalization
32f Analysis result memory area

Claims (8)

The support frame supports the periphery of the digital camera, and emits parallel light rays to an upper corner of the left frame and a lower corner of the right frame of the support frame with respect to an optical axis passing through the lens center of the digital camera. A first laser projecting body is provided, and the upper and lower frames of the support frame are provided with a second laser projecting body for emitting a light beam having a predetermined angle outward with respect to the optical axis. An image processing method using an image with a laser light spot to photograph an investigation surface and normalize using the image with a laser light spot,
On the computer,
Capture the image with laser light spots taken by the photographic device, and display on the screen,
Recognizing pixel coordinates on the screen each time there is a designation of the position of the laser light spot in the image with the laser light spot displayed on the screen;
A center is obtained based on the pixel coordinates of the four points obtained by the second laser projectile, distances on the x-axis and y-axis from the center to the pixel coordinates of the four points are obtained, and a ratio of these distances Calculating the shake angle of the digital camera up and down, left and right,
Based on the pixel coordinates of the two points obtained by the first laser projectile and the camera parameters input in advance, the object distance on the film of the digital camera and the surface to be inspected is obtained, and from this object distance Determining the scale of the image on the film;
Normalizing the image with laser light spots on the screen to an image taken from the front using the vertical and horizontal deflection angles and the scale factor, the image having laser light spots. Image processing method using The support frame supports the periphery of the digital camera, and emits parallel light rays to an upper corner of the left frame and a lower corner of the right frame of the support frame with respect to an optical axis passing through the lens center of the digital camera. A first laser projecting body is provided, and the upper and lower frames of the support frame are provided with a second laser projecting body for emitting a light beam having a predetermined angle outward with respect to the optical axis. A mosaic processing method using an image with a laser light spot to photograph a survey surface and generate a mosaic image using the image with a laser light spot,
On the computer,
A large number of first distant images obtained by sequentially photographing predetermined adjacent areas of the inspected surface at an arbitrary inclination while keeping the focal length of the digital camera constant with respect to the inspected surface are read and displayed on the screen. Steps and
A step of displaying a photographing place piece table in which the surveyed surface is divided by a mesh at a fixed interval and the number of the surveyed surface is assigned;
When a frame in the shooting location frame table is designated and a distant view image among a number of first distant view images displayed on the screen is designated, the designated second distant view image is moved to the designated frame. Letting it know where the distant view image of the survey surface is,
Displaying one distant view image from the plurality of first distant view images on a screen, specifying all the laser light points in the distant view image in a predetermined direction, and recognizing pixel coordinates of the specified position; ,
A center is determined based on the vertical and horizontal coordinates of four points of the second laser projectile, vertical and horizontal distances of the four points are determined from the center, and a vertical angle and a horizontal angle of the image capturing apparatus are determined by a ratio of these distances. Steps and
The object distance between the film of the digital camera and the surface to be inspected is obtained based on the pixel coordinates of two points obtained by the first laser projectile and the camera parameters input in advance. Calculating the scale of the image on the film from the ratio;
Correcting the aspect ratio of one distant image on the screen based on the scale and the object distance and the left and right angles and the up and down angles to obtain one distant image that assumes that the object to be investigated is photographed from the front; ,
Successively displaying adjacent distant images, performing bundle orientation by specifying the same location on these images for each display, and obtaining a mosaic image that is an image of the entire surface to be inspected. Mosaic processing method using an image with laser light spots. Determining a center from the pre-input camera parameters of the digital camera and the vertical and horizontal coordinates of the four points to determine whether or not the center satisfies the condition of the digital camera center. 3. A mosaic processing method using an image with laser light spots according to claim 2. Reading a close-up image obtained by photographing a desired region of the surface to be inspected at an arbitrary inclination with a focal length longer than the focal length of the digital camera, and displaying the image on a screen,
One proximity image is displayed on the screen from the plurality of first proximity images, all the laser light points in the second proximity image are designated in a predetermined direction, and the pixel coordinates of the designated position are recognized. Steps to
In the second proximity image, a center is obtained based on the vertical and horizontal coordinates of the four points of the second laser projectile, the vertical and horizontal distances of the four points are obtained from the center, and the imaging device is calculated by the ratio of these distances. Determining the vertical and horizontal angles of the
Based on the pixel coordinates of two points obtained by the first laser projectile in the second proximity image, and camera parameters input in advance, a position on the film of the digital camera and the surface to be inspected is determined. Obtaining an object distance, obtaining the scale of the image on the film from the object distance and the ratio,
Correcting the aspect ratio of the second proximity image on the screen based on the scale, the object distance, the left-right angle, and the up-down angle to obtain a third proximity image in which the subject is photographed from the front; ,
Displaying the third proximity image and the mosaic image on a screen, and allocating the third proximity image displayed on the screen to a range specified on the mosaic image. 4. A mosaic processing method using the image with laser light spots according to 2 or 3. Obtaining a locus of a crack from the difference in density and luminance on the third proximity image, obtaining and displaying various analysis results of the locus of the crack,
5. The mosaic processing method using an image with laser light spots according to claim 2, further comprising the step of displaying a scale at a fixed interval on the trajectory of the crack. The support frame supports the periphery of the digital camera, and emits parallel light rays to an upper corner of the left frame and a lower corner of the right frame of the support frame with respect to an optical axis passing through the lens center of the digital camera. A first laser projecting body is provided, and the upper and lower frames of the support frame are provided with a second laser projecting body for emitting a light beam having a predetermined angle outward with respect to the optical axis. The image with the laser light spot obtained by photographing the survey surface is stored in the terminal, and the normalized distance image and the near image are provided from the server by communicating with the terminal via the computer network system. A service system using
The server comprises:
For the terminal, the image with the laser light spot photographed by the photographing device, the pixel coordinates of four points of the second laser projectile on the screen of the terminal, and the pixel coordinates of two points of the first laser projectile Means for transmitting
Receiving the image with the laser light spot and the pixel coordinates, obtaining a center based on the vertical and horizontal coordinates of the four points on the image, obtaining the vertical and horizontal distances of the four points from the center, and determining the ratio of these distances Means for determining the vertical angle, left and right angles of the imaging device,
The object distance between the film of the digital camera and the surface to be inspected is obtained based on the pixel coordinates of two points obtained by the first laser projectile and the camera parameters input in advance. Means for determining the scale of the image on the film from the ratio,
Correcting the aspect ratio of one distant image on the screen based on the scale and the object distance, and the right and left angles and the up and down angles, and normalizing one distant image assuming that the subject is photographed from the front. Means to obtain,
Means for providing the normalized one distant view image to the terminal. A normalized service system using an image with laser light spots. The server comprises:
When a group of images obtained by rearranging the images with the laser light spots according to the surface to be inspected is transmitted from the terminal, these images are normalized and transmitted to the terminal, and adjacent images are transmitted to the terminal. Means for specifying and transmitting the reference point of the
Means for performing bundle orientation using the designated reference points and providing a mosaic image obtained by synthesizing the respective images to the terminal, the image with laser light spots according to claim 6, The normalization service system used. The terminal is
A large number of first distant images obtained by sequentially photographing predetermined adjacent areas of the inspected surface at an arbitrary inclination while keeping the focal length of the digital camera constant with respect to the inspected surface are read and displayed on the screen. Means,
Means for displaying a shooting place piece table in which the surveyed surface is divided by a mesh at a fixed interval and the number of the surveyed surface is assigned;
When a frame in the shooting location frame table is designated and a distant view image among a number of first distant view images displayed on the screen is designated, the designated second distant view image is moved to the designated frame. Means for obtaining a rearranged image group that shows a distant view image of a place on the survey surface,
Means for displaying one distant view image from the plurality of first distant view images on a screen, specifying all the laser light points in the distant view image in a predetermined direction, and recognizing pixel coordinates of the specified position; ,
Means for designating four pixel coordinates of the second laser projectile and two pixel coordinates of the first laser projectile, and transmitting the pixel coordinates and the rearranged image group to the server; ,
Receiving the normalized distant view images from the server, sequentially displaying adjacent distant view images, specifying the same reference point on these images for each of the displays, and Means for sending to
8. The system according to claim 6, further comprising: means for receiving a mosaic image from the server and displaying the mosaic image on a screen.

Images