JP2000123300A - Method and device for detecting white line accompanied with correction based on disappearing point - Google Patents

Method and device for detecting white line accompanied with correction based on disappearing point

Info

Publication number
JP2000123300A
JP2000123300A JP10291239A JP29123998A JP2000123300A JP 2000123300 A JP2000123300 A JP 2000123300A JP 10291239 A JP10291239 A JP 10291239A JP 29123998 A JP29123998 A JP 29123998A JP 2000123300 A JP2000123300 A JP 2000123300A
Authority
JP
Japan
Prior art keywords
line
intersections
lines
estimated
vanishing point
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.)
Granted
Application number
JP10291239A
Other languages
Japanese (ja)
Other versions
JP3047895B2 (en
Inventor
Kenichi Matsumura
謙一 松村
Kazuyuki Sakurai
和之 櫻井
Yuji Katsuno
有二 勝野
Hirofumi Matsuura
洋文 松浦
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.)
NEC Corp
Original Assignee
NEC Corp
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 NEC Corp filed Critical NEC Corp
Priority to JP10291239A priority Critical patent/JP3047895B2/en
Publication of JP2000123300A publication Critical patent/JP2000123300A/en
Application granted granted Critical
Publication of JP3047895B2 publication Critical patent/JP3047895B2/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Landscapes

  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Image Processing (AREA)

Abstract

PROBLEM TO BE SOLVED: To make compatible robustness and high accuracy by estimating a disappearing point while using only lines, in which a specified number of lines alternately cross near, out of the combination of detected white lines or road end lines and other lines found from substitutive feature points. SOLUTION: A picked-up image is provided by an image pick means 801 and lines such as white lines or road end lines more than four are extracted by a means 802 for estimating the white line/road end line from the feature points of white lines/road ends with respect to this image. Next, three lines are selected out of detected lines and three intersections are found by applying a means for finding the mutual intersections of three estimated lines. Afterwards, the distance of three intersections is evaluated by a means 804 for evaluating it as an index, three intersections to most shorten the sum of mutual distance is found and the other intersections are excluded. In respect to three intersections to most shorten the sum of mutual distances, the disappearing point is estimated by a means 805 for estimating the disappearing point from recent three intersections.

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、車載カメラで撮像
した道路画像情報から白線を検出する方法及び装置に関
し、特に、道路端線等の直線の交差する消失点算出方法
及び装置と消失点による補正を伴う白線検出方法に関す
る。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for detecting a white line from road image information captured by an on-vehicle camera, and more particularly to a method and an apparatus for calculating a vanishing point where straight lines such as road edge lines intersect. The present invention relates to a white line detection method with correction.

【0002】[0002]

【従来の技術】車載カメラで撮像された画像情報から道
路領域等を検出するに際して、白線等の道路端線が交叉
する消失点を求める処理が行われる。この消失点の算出
として、例えば特開平7−78255号公報には、消失
点を画面内の直線部分の交点とし求めるに際して、画面
上の点を所定方法に沿って順に処理したとき、色調など
の変化が小さい安定状態と色調などの変化が大きい不安
定状態との切り換わる点を直線候補点として抽出するこ
とで、白線のみならず、中央分離帯、路側帯に対応した
直線部分を抽出し、道路消失点の算出精度を高めるよう
にした方法が提案されている。
2. Description of the Related Art When detecting a road area or the like from image information captured by an on-vehicle camera, processing is performed to find a vanishing point where a road edge line such as a white line crosses. As the calculation of the vanishing point, for example, in Japanese Patent Application Laid-Open No. 7-78255, when the vanishing point is determined as an intersection of a straight line portion in the screen, when the points on the screen are sequentially processed according to a predetermined method, the color tone and the like are determined. By extracting the point at which the stable state where the change is small and the unstable state where the change of the color tone is large is extracted as a line candidate point, not only the white line, but also the straight line portion corresponding to the median strip and the roadside zone, A method has been proposed in which the calculation accuracy of a road vanishing point is improved.

【0003】ところで、画像情報から推定した道路端線
について、本来、道路上の線の平行性から、その消失点
上の1点で交わるはずであるが、実際に推定した道路端
線では、例えば図1に示すように、1点で交差していな
い。
By the way, the road edge estimated from the image information should originally intersect at one point on the vanishing point due to the parallelism of the line on the road. As shown in FIG. 1, they do not intersect at one point.

【0004】[0004]

【発明が解決しようとする課題】すなわち、従来の消失
点算出方法は下記記載の問題点を有している。
That is, the conventional vanishing point calculation method has the following problems.

【0005】第1の問題点は、道路白線を検出して得ら
れる消失点の精度が低い、ということである。また、こ
の消失点を基準に、その後の、画像処理、車の制御を行
っている為、消失点の算出精度が低いことは、車の制御
にまで影響を及ぼす。
The first problem is that the vanishing point obtained by detecting a road white line has low accuracy. Further, since the subsequent image processing and vehicle control are performed on the basis of this vanishing point, the low calculation accuracy of the vanishing point affects the control of the vehicle.

【0006】その理由は、消失点は補外(外挿)して、
交点を求めることで、算出している為、算出した白線の
傾き変動の影響を非常に受け、交点位置が大きく変わる
ためである。
[0006] The reason is that the vanishing point is extrapolated (extrapolated),
This is because the calculation is performed by obtaining the intersection, and thus the calculation is greatly affected by the calculated variation in the inclination of the white line, and the position of the intersection greatly changes.

【0007】第2の問題点は、線推定単独の精度を確保
することが困難である、ということである。
[0007] The second problem is that it is difficult to secure the accuracy of line estimation alone.

【0008】その理由は、ある影などの系統的な誤差要
因が、屋外画像の場合、多く発生している、からであ
る。
The reason is that systematic error factors such as certain shadows occur frequently in outdoor images.

【0009】したがって本発明は、上記問題点に鑑みて
なされたものであって、その目的は、ロバスト性と高精
度の両立を可能とした消失点算出方法及び装置を提供す
ることにある。
Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to provide a vanishing point calculation method and apparatus capable of achieving both robustness and high accuracy.

【0010】[0010]

【課題を解決するための手段】前記目的を達成する本発
明は、その概略を述べれば、検出した白線もしくは道路
端線、その他の代替特徴点から求めた線の組み合わせの
中から、最も3本の線が近くで交互に交わる線のみを用
いて消失点を推定することで、ロバスト性、及び高精度
化を達成するものである。
According to the present invention, which achieves the above objects, the present invention can be summarized as follows: a white line or a road edge detected; By estimating the vanishing point by using only the line where the line intersects alternately, robustness and high accuracy are achieved.

【0011】本発明は、撮像画像から、白線及び道路端
線、あるいは白線、道路端線、及び代替特徴点からの線
を推定し、前記推定された線のうち3本以上の線の相互
の交点から、前記交点相互の距離の和を最小とする3つ
線の3交点(「最近3交点」という)を取り出し、前記
3交点の平均位置を、推定消失点とする。
According to the present invention, a white line and a road edge line or a line from a white line, a road edge line, and an alternative feature point are estimated from a captured image, and three or more of the estimated lines are mutually reciprocated. From the intersections, three intersections of three lines that minimize the sum of the distances between the intersections (referred to as “most recent three intersections”) are taken out, and an average position of the three intersections is set as an estimated vanishing point.

【0012】そして、レーン(白線)検出において、検
出した白線もしくは道路端線、その他の代替特徴点から
求めた線についての消失点を利用して、誤検出した白線
もしくは道路端線、その他の代替特徴点から求めた線を
除外することで、整合性の低い推定結果の線を使用しな
い、ようにしたものである。
In the detection of a lane (white line), a vanishing point of a detected white line or a road edge line or a line obtained from another alternative feature point is used to detect a falsely detected white line or a road edge line or another alternative line. By excluding the line obtained from the feature points, the line of the estimation result having low consistency is not used.

【0013】[0013]

【発明の実施の形態】本発明の実施の形態について説明
する。図1において、計4本の白線、道路端線の交点計
6箇所から、3本に対応する3点について、相互距離の
和が最短になる3交点を取出し、その平均位置を、図
2、及び図3に示すように、推定消失点とする。消失点
が高信頼度・高精度で算出でき、その推定消失点を用い
ることで、白線検出のロバスト性や精度向上を実現する
ものである。
Embodiments of the present invention will be described. In FIG. 1, three intersections where the sum of the mutual distances is the shortest are taken out of a total of six intersections of four white lines and road end lines, for three points corresponding to the three lines, and the average position is shown in FIG. And an estimated vanishing point as shown in FIG. The vanishing point can be calculated with high reliability and high accuracy, and by using the estimated vanishing point, robustness and improvement in accuracy of white line detection are realized.

【0014】なお、道路がカーブの場合、左右の白線と
道路端線から推定した直線に代えて、接線を用いること
で、上記効果を奏する。
When the road is a curve, the above effect is obtained by using a tangent line instead of a straight line estimated from the left and right white lines and the road end line.

【0015】[0015]

【実施例】本発明の実施例について図面を参照して以下
に説明する。
DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings.

【0016】図9は、本発明の一実施例の構成を示す図
である。図1、図2、図3は、本発明の一実施例を説明
するための図である。
FIG. 9 is a diagram showing the configuration of one embodiment of the present invention. FIGS. 1, 2 and 3 are diagrams for explaining an embodiment of the present invention.

【0017】図9を参照すると、本発明の一実施例の装
置は、画像撮像手段801で取得された撮像画像情報か
ら、白線、道路端などの特徴点から白線・道路端線を推
定する手段802と、推定された線の3本の相互の交点
を求める手段803と、3交点の距離を指標に評価する
手段804と、最も近い3交点から消失点を推定する手
段805と、を有する。
Referring to FIG. 9, an apparatus according to an embodiment of the present invention estimates a white line and a road edge line from characteristic points such as a white line and a road edge from image information acquired by an image imaging unit 801. 802, means 803 for obtaining three mutual intersections of the estimated line, means 804 for evaluating the distance between the three intersections as an index, and means 805 for estimating a vanishing point from the closest three intersections.

【0018】画像撮像手段801により、撮像画像10
1を得、これに対し、白線・道路端などの特徴点から白
線・道路端線を推定する手段802により、白線10
2、道路端線103等の4本以上の線を抽出する。
The image pickup means 801 causes the picked-up image 10
1, and the white line 10 is estimated by the means 802 for estimating the white line / road edge from the characteristic points such as the white line / road edge.
2. Extract four or more lines such as the road edge line 103.

【0019】次に、検出された線から3本の線を選ん
で、推定された線の3本の相互の交点を求める手段80
3を適用して3交点を求め、その後、3交点の距離を指
標に評価する手段804にて評価を行い、相互距離の和
が最短になる3交点107を求め、それ以外の交点は除
外する。
Next, three lines are selected from the detected lines, and a means 80 for determining the three intersections of the estimated lines is provided.
3 is applied to obtain three intersections, and then evaluated by means 804 for evaluating the distance between the three intersections as an index, the three intersections 107 having the shortest sum of mutual distances are obtained, and other intersections are excluded. .

【0020】3交点107に対し、最近3交点から消失
点を推定する手段805において、消失点104を推定
する。
For the three intersections 107, the vanishing point 104 is estimated by means 805 for estimating the vanishing point from the three most recent intersections.

【0021】なお、これらの手段802〜805は、コ
ンピュータ上で実行されるプログラム制御によってその
機能を実現するようにしてもよい。この場合、該プログ
ラムを格納した記録媒体からコンピュータが該プログラ
ムを読み出し実行することで本発明を実施することがで
きる。
The functions of these means 802 to 805 may be realized by program control executed on a computer. In this case, the present invention can be implemented by a computer reading and executing the program from a recording medium storing the program.

【0022】次に、図1、図2、図3を参照して、本実
施例の全体の動作について詳細に説明する。
Next, the overall operation of the present embodiment will be described in detail with reference to FIGS.

【0023】車両から撮像された道路の撮像画像101
から白線102と道路端線103の計4本の線を推定す
る。
A captured image 101 of a road captured from a vehicle
, A total of four lines of a white line 102 and a road end line 103 are estimated.

【0024】その相互の交点は、図1の106の部分拡
大図である図2に示すように、必ずしも1箇所で交わっ
てはいない。この為、3本の線の相互交点間の距離を基
準に、どの程度の一致度で、3本の線が交わっているか
評価を行う。
As shown in FIG. 2, which is a partially enlarged view of 106 in FIG. 1, the intersections do not always intersect at one point. For this reason, the degree of coincidence between the three lines is evaluated based on the distance between the intersections of the three lines.

【0025】この結果、最も近くで交互に交わっている
3本の線と、最近の3交点107を求めることが出来
る。図2の109の部分拡大図である図3に示すよう
に、最近の3交点107からなる三角形の平均位置11
0を、消失点の推定位置とする。
As a result, it is possible to find the three lines that intersect at the closest point and the three most recent intersections 107. As shown in FIG. 3 which is a partially enlarged view of 109 in FIG.
0 is the estimated position of the vanishing point.

【0026】なお、平均位置としては、例えば、消失点
を通る線を再算出する場合には、(1)単純平均、もし
くは交点をなした線の信頼度に応じた荷重からの重心、
(2)推定した線の傾きを保存して推定線を平行移動さ
せる場合の内心、(3)線の傾きの補正を抑える場合は
外心と、その後の処理に応じた処理にて、平均位置とし
て消失点を推定する。
As the average position, for example, when recalculating a line passing through the vanishing point, (1) a simple average or the center of gravity from a load corresponding to the reliability of the line forming the intersection,
(2) Inner center when the estimated line inclination is stored and the estimated line is translated and (3) Outer center when the line inclination correction is suppressed, and the average position in the processing according to the subsequent processing. To estimate the vanishing point.

【0027】次に本発明の第2の実施例について説明す
る。図4乃至図8は、本発明の第2の実施例を説明する
ための図である。
Next, a second embodiment of the present invention will be described. FIGS. 4 to 8 are diagrams for explaining a second embodiment of the present invention.

【0028】図4を参照すると、検出した白線102、
道路端線103以外に、代替特徴点から推定した線11
1も用いて、より多くの推定線について、3線の組み合
わせについて3交点を算出している。
Referring to FIG. 4, the detected white line 102,
Line 11 estimated from alternative feature points other than road edge line 103
1 is also used to calculate three intersections for a combination of three lines for more estimated lines.

【0029】この場合、図5に示すように、検出した線
の対象種類ごとに、ポイントを与えたり、検出した線の
信頼度を用いて、判定を優先度順に行う。判定基準に合
致するものがその範囲内であれば、その先の判定を行わ
ない。図5に示す例では、線種別毎にポイント付して、
3本の線の優先度を求めており、検出された白線102
である線、をポイント1、道路端線103である線
、をポイント2、代表特徴点からの線111である
線をポイント3とし、3本の組合せについて、白線
、白線、道路端線又は道路端線の組合せの場
合、ポイントは1+1+2=4と優先度は最も大とな
り、道路単線、、代替特徴点から線の組合せの場
合、ポイントは2+2+3=7となり、優先度は最も低
くなる。
In this case, as shown in FIG. 5, a determination is made in order of priority by giving a point or using the reliability of the detected line for each target type of the detected line. If an item that meets the judgment criteria is within the range, no further judgment is made. In the example shown in FIG. 5, a point is attached for each line type,
The priority of the three lines is obtained, and the detected white line 102
Is a point 1, a line which is a road edge line 103, a point 2, a line which is a line 111 from a representative feature point is a point 3, and a white line, a white line, a road edge line or a road is determined for three combinations. In the case of the combination of the end lines, the point has the highest priority of 1 + 1 + 2 = 4, and in the case of the combination of the road single line and the line from the alternative feature point, the point has the priority of 2 + 2 + 3 = 7 and the lowest priority.

【0030】なお、線〜の検出信頼度(例えば0.
8,0.5,0.4,0,6,0.3)を用いて、判定
を優先度を求めるようにしてもよい。
It should be noted that the detection reliability of the line (for example, 0.
8, 0.5, 0.4, 0, 6, 0.3) may be used to determine the priority.

【0031】図6は、判定の様子を説明する模式図であ
る。図6に示すように、3交点の単純な距離の合計では
なく、推定した線種に従い、加重を設けて、加重付きで
の距離評価を行っている。すなわち、 3点距離としてL=L1×1+L2×L3×3 L1:推定した白線上の交点間距離 荷重=1 L2:推定した道路端線上の交点間距離 荷重=2 L3:推定した代替特徴量からの線線上の交点間距離
荷重=3
FIG. 6 is a schematic diagram for explaining the state of the judgment. As shown in FIG. 6, weights are provided according to the estimated line type, not the sum of the simple distances at the three intersections, and the weighted distance evaluation is performed. That is, L = L1 × 1 + L2 × L3 × 3 as three-point distance L1: Estimated distance between intersections on white line Load = 1 L2: Estimated distance between intersections on road edge line Load = 2 L3: From estimated alternative feature amount Distance between intersections on the line
Load = 3

【0032】同時に、更に、図7に示すように、同じサ
イド(側)にある線(同じ傾き方向の線)の組み合わせ
は、交点の位置算出精度が低いことから、判定対象外と
している。
At the same time, as shown in FIG. 7, a combination of lines on the same side (lines in the same inclination direction) is excluded from the judgment because the position calculation accuracy of the intersection is low.

【0033】最近3交点からの消失点の推定及び、推定
線の再計算としては、図8に示すように、内心を利用し
て、推定線を同時に内心上を通るように、平行移動(平
行のシフトさせる方法もある。
As shown in FIG. 8, the estimation of the vanishing point from the three most recent intersections and the recalculation of the estimated line are performed by parallel movement (parallel movement) using the inner center so that the estimated line passes on the inner center at the same time. There is also a method of shifting.

【0034】これは、レーン走行では、傾き情報が重要
であるため、この情報を保存する場合に適用する。
This is applied to the case where the inclination information is important in lane running, and this information is stored.

【0035】なお、標準的には、3交点の重心を推定消
失点とし、重心を通ることを拘束条件に、線を再推定す
ることも行っている。
Note that, as a standard, the line of gravity is re-estimated with the center of gravity of the three intersections as the estimated vanishing point, and passing through the center of gravity as a constraint.

【0036】[0036]

【発明の効果】以上説明したように、本発明によれば、
下記記載の効果を奏する。
As described above, according to the present invention,
The following effects are obtained.

【0037】本発明の第1の効果は、画像処理の基準と
なる消失点をロバストに且つ高精度に求めることができ
る、ということである。
A first effect of the present invention is that a vanishing point serving as a reference for image processing can be obtained robustly and with high accuracy.

【0038】その理由は、本発明においては、推定した
線の信頼度を加味して、選択した3線の3交点のまとま
り具合を指標にして、消失点の推定に使用する3線を選
別しており、例外データに引きずられた誤差の多い推定
線を、消失点推定から除外している、ためである。
The reason is that, in the present invention, considering the reliability of the estimated line, the three lines used for estimating the vanishing point are selected based on the degree of unity of the three intersections of the selected three lines. This is because the estimation line with a large amount of error, which is dragged by the exception data, is excluded from the vanishing point estimation.

【0039】本発明の第2の効果は、線の再推定を行う
ことができる、ということである。
A second advantage of the present invention is that the line can be re-estimated.

【0040】その理由は、本発明においては、求まった
消失点の精度が高いため、この消失点を通って傾きを変
化させながら線としての再評価を行う際に、消失点精度
が低い場合に発生する線評価値のピークの平坦化を抑止
することができる、ためである。
The reason is that, in the present invention, since the accuracy of the vanishing point obtained is high, when re-evaluating a line while changing the slope through this vanishing point, if the vanishing point precision is low, This is because flattening of the peak of the generated line evaluation value can be suppressed.

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

【図1】本発明の一実施例を説明するための図である。FIG. 1 is a diagram for explaining an embodiment of the present invention.

【図2】本発明の一実施例を説明するための図であり、
図1の領域106の部分拡大図である。
FIG. 2 is a diagram for explaining one embodiment of the present invention;
FIG. 2 is a partially enlarged view of a region 106 in FIG. 1.

【図3】本発明の一実施例を説明するための図であり、
図2の領域109の部分拡大図である。
FIG. 3 is a diagram for explaining one embodiment of the present invention;
FIG. 3 is a partially enlarged view of a region 109 in FIG. 2.

【図4】本発明の第2の実施例を説明するための図であ
る。
FIG. 4 is a diagram for explaining a second embodiment of the present invention.

【図5】本発明の第2の実施例を説明するための図であ
る。
FIG. 5 is a diagram for explaining a second embodiment of the present invention.

【図6】本発明の第2の実施例を説明するための図であ
る。
FIG. 6 is a diagram for explaining a second embodiment of the present invention.

【図7】本発明の第2の実施例を説明するための図であ
る。
FIG. 7 is a diagram for explaining a second embodiment of the present invention.

【図8】本発明の第2の実施例を説明するための図であ
る。
FIG. 8 is a diagram for explaining a second embodiment of the present invention.

【図9】本発明の一実施例の構成を示す図である。FIG. 9 is a diagram showing a configuration of one embodiment of the present invention.

【符号の説明】[Explanation of symbols]

101 撮像画像 102 検出した白線 103 検出した道路端線 104 推定した消失点 105 推定から除外された道路端線 106 撮像画像の部分領域 107 最近の3交点(評価指標が最小となる3交点) 108 除外する3交点 109 部分領域 110 最近3交点の平均位置 111 代替特徴量から検出した線 112 撮像画像の部分図 113 距離算出の例での3線の3交点 801 画像撮像手段 802 白線道路端推定手段 803 推定された3本の線の交点を求める手段 804 3交点の距離の指標を評価する手段 805 3交点から消失点を推定する手段 DESCRIPTION OF SYMBOLS 101 Captured image 102 Detected white line 103 Detected road edge line 104 Estimated vanishing point 105 Road edge line excluded from estimation 106 Partial region of captured image 107 Recent three intersections (three intersections where evaluation index becomes minimum) 108 Exclusion 3 intersections 109 to be performed 109 Partial area 110 Average position of the latest 3 intersections 111 Lines detected from alternative feature amounts 112 Partial view of captured image 113 Three intersections of three lines in an example of distance calculation 801 Image capturing means 802 White line road end estimating means 803 Means for finding the intersection of the estimated three lines 804 Means for evaluating the index of the distance between the three intersections 805 Means for estimating the vanishing point from the three intersections

─────────────────────────────────────────────────────
────────────────────────────────────────────────── ───

【手続補正書】[Procedure amendment]

【提出日】平成11年9月27日(1999.9.2
7)
[Submission date] September 27, 1999 (September 9, 1999
7)

【手続補正1】[Procedure amendment 1]

【補正対象書類名】明細書[Document name to be amended] Statement

【補正対象項目名】特許請求の範囲[Correction target item name] Claims

【補正方法】変更[Correction method] Change

【補正内容】[Correction contents]

【特許請求の範囲】[Claims]

【手続補正2】[Procedure amendment 2]

【補正対象書類名】明細書[Document name to be amended] Statement

【補正対象項目名】0011[Correction target item name] 0011

【補正方法】変更[Correction method] Change

【補正内容】[Correction contents]

【0011】本発明は、撮像画像から、白線及び道路端
線、あるいは白線、道路端線、及び代替特徴点からの線
を推定し、前記推定された線のうち3本以上の線の相互
の交点から、前記交点相互の距離の和を最小とする3つ
線の3交点(「最近3交点」という)を取り出し、前記
3交点の平均位置を、推定消失点とする。本発明の白線
検出方法においては、前記最近3交点からの推定位置と
して内心を用い、線の再推定では、傾きを保存し上で、
線の推定位置の補正を最小にする。また、前記最近3交
点からの推定位置として重心を用い、線の再推定では、
前記重心を通る線として再推定する。 ─────────────────────────────────────────────────────
According to the present invention, a white line and a road edge line or a line from a white line, a road edge line, and an alternative feature point are estimated from a captured image, and three or more of the estimated lines are mutually reciprocated. From the intersections, three intersections of three lines that minimize the sum of the distances between the intersections (referred to as “most recent three intersections”) are taken out, and an average position of the three intersections is set as an estimated vanishing point. White line of the present invention
In the detection method, the estimated position from the three most recent intersections is
Using the inner center, and re-estimating the line, save the slope,
Minimize the correction of the estimated line position. In addition, the recent three
Using the center of gravity as the estimated position from the point, and re-estimating the line,
It is re-estimated as a line passing through the center of gravity. ────────────────────────────────────────────────── ───

【手続補正書】[Procedure amendment]

【提出日】平成11年12月20日(1999.12.
20)
[Submission date] December 20, 1999 (1999.12.
20)

【手続補正1】[Procedure amendment 1]

【補正対象書類名】明細書[Document name to be amended] Statement

【補正対象項目名】特許請求の範囲[Correction target item name] Claims

【補正方法】変更[Correction method] Change

【補正内容】[Correction contents]

【特許請求の範囲】[Claims]

───────────────────────────────────────────────────── フロントページの続き (72)発明者 勝野 有二 神奈川県横浜市神奈川区新浦島町1丁目1 番地25 日本電気ロボットエンジニアリン グ株式会社内 (72)発明者 松浦 洋文 神奈川県横浜市神奈川区新浦島町1丁目1 番地25 日本電気ロボットエンジニアリン グ株式会社内 Fターム(参考) 2F065 AA03 AA32 BB02 BB27 CC40 DD03 FF01 FF26 JJ03 JJ26 MM06 QQ21 QQ32 QQ42 UU05 5B057 AA16 DA07 DB02 DC16 5H180 AA01 CC04 CC24 LL01 LL02 LL09  ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Yuji Katsuno 1-1-1 Shinurashima-cho, Kanagawa-ku, Yokohama-shi, Kanagawa-ken Within Nippon Electric Robot Engineering Co., Ltd. (72) Inventor Hirofumi Matsuura Kanagawa-ku, Yokohama-shi, Kanagawa 1-1-1 Shinurashimacho 25-term F-term (reference) within NEC Robot Engineering Co., Ltd.

Claims (14)

【特許請求の範囲】[Claims] 【請求項1】撮像画像から、白線及び道路端線、あるい
は白線、道路端線、及び代替特徴点からの線を推定し、 前記推定された線のうち3本以上の線の相互の交点か
ら、前記交点相互の距離の和を最小とする3つ線の3交
点(「最近3交点」という)を取り出し、前記3交点の
平均位置を、推定消失点とする、ことを特徴とする消失
点算出方法。
1. A white line and a road edge line, or a line from a white line, a road edge line, and an alternative feature point are estimated from a captured image, and a line from three or more intersections among the estimated lines is determined. A vanishing point wherein three intersections of three lines that minimize the sum of the distances between the intersections (referred to as “most recent three intersections”) are taken out, and an average position of the three intersections is set as an estimated vanishing point. Calculation method.
【請求項2】撮像画像から、白線、道路端線、及び代替
特徴点からの線を推定し、 前記推定された線の交点から相互の距離の和を計算する
際に、前記線の検出信頼性に対応するペナルティとし
て、白線<道路端線<代替特徴点から、推定した線の順
に、その交点間を結ぶ線種に応じて、予め定められた荷
重を掛けた相互距離の和を最小にする3交点(「最近3
交点」という)を取り出し、前記3交点の平均位置を推
定消失点とする、、ことを特徴とする消失点算出方法。
2. Estimating a line from a white line, a road edge line, and an alternative feature point from a captured image, and calculating a sum of mutual distances from an intersection of the estimated line. As a penalty corresponding to the gender, minimize the sum of the mutual distances with a predetermined load applied according to the line type connecting the intersections in the order of the estimated line from the white line <road end line <alternate feature point 3 intersections ("Recent 3
A point of intersection)), and taking the average position of the three intersections as an estimated vanishing point.
【請求項3】白線の検出信頼性評価の評価値が低い場合
に、請求項1又は2記載の方法を用いて消失点を推定す
る、ことを特徴とする消失点算出方法。
3. A vanishing point calculation method characterized by estimating a vanishing point by using the method according to claim 1 or 2 when the evaluation value of the detection reliability evaluation of the white line is low.
【請求項4】白線のみ3本ある場合、 白線2本と道路端線1本、 白線1本と道路端線2本、 白線2本と代替特徴点からの線1本、 白線1本と道路端線1本と代替特徴点からの線1本 の順番でどの程度の一致度で3本の線が交わっているか
の評価を行い、 評価基準に最初合致したものが検出された時点で、消失
点の推定処理を打ち切る、ことを特徴とする請求項1又
は2記載の消失点算出方法。
4. When there are only three white lines, two white lines and one road end line, one white line and two road end lines, two white lines and one line from an alternative feature point, one white line and a road An evaluation is made to determine the degree of coincidence between the three lines in the order of one edge line and one line from the alternative feature point. When the first line that meets the evaluation criteria is detected, it disappears. 3. The vanishing point calculation method according to claim 1, wherein the point estimation processing is terminated.
【請求項5】請求項1乃至4のいずれか一に記載の消失
点算出方法により消失点を求め、誤検出した白線もしく
は道路端線、その他の代替特徴点から求めた線を推定か
ら除外することを特徴とする白線検出方法。
5. A vanishing point is calculated by the vanishing point calculation method according to any one of claims 1 to 4, and a white line or a road edge line detected erroneously, and a line obtained from another alternative feature point are excluded from the estimation. A method for detecting white lines, characterized in that:
【請求項6】請求項1乃至4のいずれか一に記載の消失
点算出方法による前記最近3交点からの推定位置として
内心を用い、線の再推定では、傾きを保存し上で、線の
推定位置の補正を最小にする、ことを特徴とする白線検
出方法。
6. An inner center is used as an estimated position from the three most recent intersections according to the vanishing point calculation method according to any one of claims 1 to 4, and in re-estimation of a line, a gradient is preserved, and A white line detection method characterized by minimizing correction of an estimated position.
【請求項7】請求項1乃至4のいずれか一に記載の消失
点算出方法による前記最近3交点からの推定位置として
重心を用い、線の再推定では、前記重心を通る線として
再推定する、ことを特徴とする白線検出方法。
7. A center of gravity is used as an estimated position from the three most recent intersections according to the vanishing point calculation method according to any one of claims 1 to 4, and in line re-estimation, a line passing through the center of gravity is re-estimated. And a white line detecting method.
【請求項8】3交点算出のための3線として、画面上の
垂直を基準として、同じ傾きの線3本については、指標
算出対象としない、ことを特徴とする請求項1乃至4の
いずれか一に記載の消失点算出方法。
8. The method according to claim 1, wherein three lines having the same inclination with respect to a vertical line on the screen as three lines for calculating three intersections are not included in the index calculation. The vanishing point calculation method according to claim 1.
【請求項9】画像撮像手段と、 前記画像撮像手段で取得された撮像画像情報から、白線
及び道路端線、もしくは、白線、道路端線、及び代替特
徴点からの線、を含む4本以上の線を推定する手段と、 前記推定された線のうち3本の線の相互の交点を求める
手段と、 前記3交点の距離を指標にどの程度の一致度で3本の線
が交わっているかを評価する手段と、 前記評価の結果、最も近い3交点から、その平均位置を
求めることで消失点を推定する手段と、 を備えたことを特徴とする消失点算出装置。
9. At least four lines including a white line and a road edge line, or a white line, a road edge line, and a line from an alternative feature point, based on the image capturing means and the captured image information acquired by the image capturing means. Means for estimating the three lines of the estimated lines, means for determining the intersection of three lines among the estimated lines, and how much the three lines intersect with the distance between the three intersections as an index And a means for estimating a vanishing point by obtaining an average position of the three intersections as a result of the evaluation, and estimating a vanishing point.
【請求項10】前記推定された3本の線の交点から相互
の距離の和を計算する際に、前記線の検出信頼性に対応
するペナルティとして、白線<道路端線<代替特徴点か
ら、推定した線の順に、その交点間を結ぶ線種に応じ
て、予め定められた荷重を掛けた相互距離の和を最小と
する3交点を取り出し、前記3交点の平均位置を推定消
失点とする、ことを特徴とする請求項9記載の消失点算
出装置。
10. A penalty corresponding to the reliability of detection of the line when calculating the sum of mutual distances from the intersections of the three estimated lines is as follows: In the order of the estimated lines, three intersections that minimize the sum of the mutual distances applied with a predetermined load are extracted according to the type of line connecting the intersections, and the average position of the three intersections is set as an estimated vanishing point. 10. The vanishing point calculation device according to claim 9, wherein:
【請求項11】請求項8又は9記載の消失点算出装置に
より消失点を求め、誤検出した白線もしくは道路端線、
その他の代替特徴点から求めた線を、線の推定から除外
することを特徴とする白線検出装置。
11. A vanishing point is obtained by the vanishing point calculating device according to claim 8 or 9, and a white line or a road edge line detected erroneously is detected.
A white line detection device, wherein a line obtained from another alternative feature point is excluded from line estimation.
【請求項12】請求項8又は9記載の消失点算出装置に
よる前記最近3交点からの推定位置として内心を用い、
線の再推定では、傾きを保存し上で、線の推定位置の補
正を最小にする、ことを特徴とする白線検出装置。
12. An inner center is used as an estimated position from the three most recent intersections by the vanishing point calculation device according to claim 8 or 9,
In the line re-estimation, the inclination is preserved and the correction of the estimated position of the line is minimized.
【請求項13】請求項8又は9記載の消失点算出装置に
よる前記最近3交点からの推定位置として重心を用い、
線の再推定では、前記重心を通る線として再推定する、
ことを特徴とする白線検出装置。
13. A center of gravity used as an estimated position from the three most recent intersections by the vanishing point calculation device according to claim 8 or 9,
In the line re-estimation, re-estimate as a line passing through the center of gravity,
A white line detection device, characterized in that:
【請求項14】(a)画像撮像手段で取得された撮像画
像情報から、白線及び道路端線、もしくは白線、道路端
線、代替特徴点からの線、を含む4本以上の線を推定す
る処理と (b)前記推定された線のうち3本の線の相互の交点を
求める処理と、 (c)前記3交点の距離を指標にどの程度の一致度で3
本の線が交わっているかを評価する処理と、 (d)前記評価の結果、最も近い3交点から、その平均
位置を求めて消失点を推定する処理と、 の上記(a)〜(d)の各処理をコンピュータで実行す
るためのプログラムを記録した記録媒体。
(A) Estimating four or more lines including a white line and a road edge line, or a white line, a road edge line, and a line from an alternative feature point, from the captured image information obtained by the image capturing means. And (b) a process of obtaining mutual intersections of three lines among the estimated lines; and (c) a process of determining the degree of coincidence using the distance between the three intersections as an index.
And (d) estimating the vanishing point by calculating the average position from the three nearest intersections as a result of the evaluation, and (d) estimating the vanishing point. A recording medium on which a program for executing each of the above processes by a computer is recorded.
JP10291239A 1998-10-14 1998-10-14 White line detection method and apparatus with correction based on vanishing point Expired - Fee Related JP3047895B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP10291239A JP3047895B2 (en) 1998-10-14 1998-10-14 White line detection method and apparatus with correction based on vanishing point

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP10291239A JP3047895B2 (en) 1998-10-14 1998-10-14 White line detection method and apparatus with correction based on vanishing point

Publications (2)

Publication Number Publication Date
JP2000123300A true JP2000123300A (en) 2000-04-28
JP3047895B2 JP3047895B2 (en) 2000-06-05

Family

ID=17766287

Family Applications (1)

Application Number Title Priority Date Filing Date
JP10291239A Expired - Fee Related JP3047895B2 (en) 1998-10-14 1998-10-14 White line detection method and apparatus with correction based on vanishing point

Country Status (1)

Country Link
JP (1) JP3047895B2 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006109385A (en) * 2004-10-08 2006-04-20 Sony Corp Image processing apparatus and method therefor, and program
WO2008056660A1 (en) * 2006-11-08 2008-05-15 Nec Corporation Vanishing point detecting system, vanishing point detecting method, and vanishing point detecting program
JP2008293172A (en) * 2007-05-23 2008-12-04 Toyota Central R&D Labs Inc Disappearing dot estimation device and program
JP2018074431A (en) * 2016-10-31 2018-05-10 キヤノンマーケティングジャパン株式会社 Information processing device, information processing method, and program
JP2018148520A (en) * 2017-03-09 2018-09-20 国立研究開発法人農業・食品産業技術総合研究機構 Attachment direction parameter calculation device for on-vehicle camera and attachment direction parameter calculation method
JP2019174243A (en) * 2018-03-28 2019-10-10 アイシン・エィ・ダブリュ株式会社 Direction detection system, direction detection method, and direction detection program
CN112525147A (en) * 2020-12-08 2021-03-19 北京嘀嘀无限科技发展有限公司 Distance measurement method for automatic driving equipment and related device

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006109385A (en) * 2004-10-08 2006-04-20 Sony Corp Image processing apparatus and method therefor, and program
JP4556114B2 (en) * 2004-10-08 2010-10-06 ソニー株式会社 Image processing apparatus and method, and program
WO2008056660A1 (en) * 2006-11-08 2008-05-15 Nec Corporation Vanishing point detecting system, vanishing point detecting method, and vanishing point detecting program
JP2008123036A (en) * 2006-11-08 2008-05-29 Nec Corp Vanishing point detection system, vanishing point detection method and program for vanishing point detection
US8396299B2 (en) 2006-11-08 2013-03-12 Nec Corporation Vanishing point detecting system, vanishing point detecting method, and vanishing point detecting program
JP2008293172A (en) * 2007-05-23 2008-12-04 Toyota Central R&D Labs Inc Disappearing dot estimation device and program
JP2018074431A (en) * 2016-10-31 2018-05-10 キヤノンマーケティングジャパン株式会社 Information processing device, information processing method, and program
JP2018148520A (en) * 2017-03-09 2018-09-20 国立研究開発法人農業・食品産業技術総合研究機構 Attachment direction parameter calculation device for on-vehicle camera and attachment direction parameter calculation method
JP2019174243A (en) * 2018-03-28 2019-10-10 アイシン・エィ・ダブリュ株式会社 Direction detection system, direction detection method, and direction detection program
JP7069943B2 (en) 2018-03-28 2022-05-18 株式会社アイシン Direction detection system, direction detection method, and direction detection program
CN112525147A (en) * 2020-12-08 2021-03-19 北京嘀嘀无限科技发展有限公司 Distance measurement method for automatic driving equipment and related device
CN112525147B (en) * 2020-12-08 2022-11-08 北京嘀嘀无限科技发展有限公司 Distance measurement method for automatic driving equipment and related device

Also Published As

Publication number Publication date
JP3047895B2 (en) 2000-06-05

Similar Documents

Publication Publication Date Title
CN110147774B (en) Table format picture layout analysis method and computer storage medium
WO2016127545A1 (en) Character segmentation and recognition method
US20070258647A1 (en) Surveying apparatus and method of analyzing measuring data
JP2007042072A (en) Tracking apparatus
CN109241345B (en) Video positioning method and device based on face recognition
US20150204663A1 (en) 3d scanner using structured lighting
US6519358B1 (en) Parallax calculating apparatus, distance calculating apparatus, methods of the same, and information providing media
JP3047895B2 (en) White line detection method and apparatus with correction based on vanishing point
JP3812770B2 (en) Camera shake parameter detection method and information recording medium
JP3807651B2 (en) White line recognition device
CN107808165B (en) Infrared image matching method based on SUSAN corner detection
JP2007316950A (en) Method, apparatus and program for processing image
JP3760426B2 (en) Traveling lane detection method and apparatus
JPH08190690A (en) Method for determining number plate
JP3333224B2 (en) Traveling path detection device for mobile vehicles
JP2001209804A (en) Dynamic image processor and dynamic image processing method
JPH1151644A (en) Distance measuring instrument for vehicle
JP3118484B2 (en) Image segmentation method
JP4479174B2 (en) Object detection device
JPH05250475A (en) Pattern matching method
JP3466899B2 (en) Character recognition device and method, and program storage medium
JP2897439B2 (en) Corner position detection method
JPH08178622A (en) Edge line measuring method
CN110349183B (en) Tracking method and device based on KCF, electronic equipment and storage medium
JP3464463B2 (en) Image processing device

Legal Events

Date Code Title Description
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20000222

LAPS Cancellation because of no payment of annual fees