JPH0417099A - Traffic flow measuring picture processor - Google Patents
Traffic flow measuring picture processorInfo
- Publication number
- JPH0417099A JPH0417099A JP12093190A JP12093190A JPH0417099A JP H0417099 A JPH0417099 A JP H0417099A JP 12093190 A JP12093190 A JP 12093190A JP 12093190 A JP12093190 A JP 12093190A JP H0417099 A JPH0417099 A JP H0417099A
- Authority
- JP
- Japan
- Prior art keywords
- feature point
- point
- image pickup
- feature
- traffic flow
- 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.)
- Pending
Links
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 238000001514 detection method Methods 0.000 claims abstract description 19
- 238000005259 measurement Methods 0.000 claims abstract description 17
- 239000000284 extract Substances 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 18
- 238000003384 imaging method Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 8
- 238000012423 maintenance Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 12
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
Description
【発明の詳細な説明】
し産業上の利用分野］
この発明は、画像処理を用いて、路上を走行する車両の
車速や車間距離を計測する交通流計測画像処理装置に関
するものである。DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a traffic flow measurement image processing device that uses image processing to measure vehicle speeds and inter-vehicle distances of vehicles traveling on a road.
［従来の技術］
近年、交通管制のための情報を得る手段として１”■カ
メラを用いた交通流計測装置が実用化されている。第８
図は、奥山他 「画像処理応用交通流計測システムのア
ーキテクチャ」、情報処理学会第３７回（昭和５３年後
期）全国大会、６Ｔ−６に記載された従来の交通流計測
画像処理装置の構成図である。図において、（］）は撮
（ｆ４器で例えばＴＶカメラ、（２）はＴＶカメラ（１
）を制御す６７Ｖ力メラ制御回路、（３）はデジタル化
された撮〔象信号を取り込むフレームメモリ、（４）は
車両抽出回路、（５）は奥行き情報管理テーブル５（６
）は重両情報計測回路である。[Prior Art] In recent years, traffic flow measurement devices using 1” cameras have been put into practical use as a means of obtaining information for traffic control.
The figure is a configuration diagram of a conventional traffic flow measurement image processing device described in Okuyama et al., "Architecture of a traffic flow measurement system using image processing," Information Processing Society of Japan's 37th National Conference (late 1978), 6T-6. It is. In the figure, (]) is a camera (F4 device, for example, a TV camera, and (2) is a TV camera (1
), (3) is a frame memory that captures digital imaging signals, (4) is a vehicle extraction circuit, and (5) is a depth information management table 5 (6).
) is a heavy information measurement circuit.
次に動作について説明する。′ｒＶカメラ（１）は′「
■カメラ制御回路（２）に従って任意のフレーム間隔で
走行車両を撮像する。フレームメモリ（３）は、Ｔ’
Ｖカメラ（１）の撮像信号をデジタルに変換したデータ
を取り込む。車両抽出回路（４）は、フレームメモリ（
３）のデータを処理し、屯両を抽出する。奥行き情報管
理テーブル（５）は、ｌ” Ｖカメラ（１）から道路上
のある地点までの３次元距離（奥行き）と、その地点の
画面上の座欄値との対応を示す管理テーブルで、第９図
（ａｌ で示すように１′Ｖカ、メラ（＋１を設置した
後、画面に写されている道路上一定高さのいくつかの点
（ｐｉ〜ｐ６）を選び、それぞれの点に対する実際の３
次元距離を人力し、それらを補間することによって画面
座標に対する奥行き情報を求めて格納する。第９図ｆｂ
）は第９図（ａ）の各座標を横軸とし、奥行（ｍ）を縦
軸に示すグラフであり、画面上の座標と、それぞれの点
に対する実際の３次元距離の関係を示すものである。車
両情報計測回路（６）では、複数フレームにわたって、
重両抽出回路（４）で得られた車両の画面」二の座標か
ら奥行き情報管理テーブル（５）を用いて実際の位置を
求める。この処理を複数フレームにわたって行なうこと
により５各車両の速度や車間距離を計測する。Next, the operation will be explained. 'rV camera (1) is '
■According to the camera control circuit (2), images of the traveling vehicle are taken at arbitrary frame intervals. Frame memory (3) is T'
Data obtained by converting the imaging signal of the V camera (1) into digital data is imported. The vehicle extraction circuit (4) has a frame memory (
3) Process the data and extract the ton. The depth information management table (5) is a management table that shows the correspondence between the three-dimensional distance (depth) from the l''V camera (1) to a certain point on the road and the seat column value on the screen at that point. After installing the 1'V camera and camera (+1) as shown in Figure 9 (al), select several points (pi to p6) at a certain height on the road shown on the screen, and actual 3
Depth information for screen coordinates is calculated and stored by manually calculating dimensional distances and interpolating them. Figure 9 fb
) is a graph in which the horizontal axis is each coordinate in Figure 9 (a) and the vertical axis is depth (m), and it shows the relationship between the coordinates on the screen and the actual three-dimensional distance to each point. be. In the vehicle information measurement circuit (6), over multiple frames,
Using the depth information management table (5), the actual position is determined from the coordinates of the vehicle screen obtained by the overlapped vehicle extraction circuit (4). By performing this process over a plurality of frames, the speed and inter-vehicle distance of each of the five vehicles is measured.
［発明が解決しようとする課題］
従来の交通流計測画像処理装置は以上のように構成され
ているので、振動やメインデナンス作業の際に′ｒ■カ
メラの角度が変化した場合、奥行き情報管理テーブルを
新たに設定しなおさなＣＪればならないといつｌｊ１題
があった。[Problem to be solved by the invention] Since the conventional traffic flow measurement image processing device is configured as described above, if the angle of the camera changes due to vibration or maintenance work, the depth information management table There was one issue when I had to set up a new CJ.
この発明は一１二記のような問題を解消するためになさ
れたもので、振動やメーインデナンスによ−）で１”
Ｖカメラの角度が変化した場合でも自動的に奥行き情報
を求め、速やかに副側が続行できる交通流計測画像処理
装置を得ることを目的とする。This invention was made in order to solve the problem like the one in 112, and it was made by vibration and main indenance.
To provide a traffic flow measurement image processing device that automatically obtains depth information even when the angle of a V camera changes and allows a secondary side to quickly continue.
［課題を解決するためのＬ段］
この発明による交通流計２１１１１画像処理装置は、走
行車両を任、αのフレーム間隔で撮像する撮像器、走行
車両の特徴点を複数個設定し、撮像信号から特徴点を抽
出する特徴点抽出回路、１Ｍ数フレームにわたって同一
特徴点を抽出して特徴点列とし、特徴点列の画面座標を
結ぶことによ−）て出来る軌跡を特徴点の各々に対して
求め、それらの軌跡群から撮像器の消失点を求める消失
点検出回路、及び消失点を用いて撮像器の振れ角を求め
る撮像パラメータ検出回路を備えたものである。[L stage for solving the problem] The traffic flow meter 21111 image processing device according to the present invention includes a moving vehicle, an image pickup device that captures an image at a frame interval of α, a plurality of feature points of the moving vehicle, and an imaging signal. A feature point extraction circuit that extracts feature points from 1M frames, extracts the same feature points as a feature point string, and connects the screen coordinates of the feature point string to create a trajectory for each feature point. A vanishing point detection circuit that determines the vanishing point of the imager from a group of trajectories, and an imaging parameter detection circuit that uses the vanishing point to determine the deflection angle of the imager.
［作用］
この発明における特徴点抽出回路は、走行車両の複数個
の特徴点を複数フレームにわたって抽出し、消失点検出
回路は同一特徴点の複数フレームにわたる画面座標のデ
ータを結ぶことによってできる軌跡を各特徴点に対して
求め５それらの軌跡群から撮像器の消失点を求める。こ
の消失点を用いて撮像器の振れ角を計算し、振れ角に基
いて奥行座標を補正する。[Operation] The feature point extraction circuit of the present invention extracts a plurality of feature points of a running vehicle over a plurality of frames, and the vanishing point detection circuit extracts a trajectory created by connecting screen coordinate data of the same feature point over a plurality of frames. 5. Find the vanishing point of the imager from a group of trajectories for each feature point. Using this vanishing point, the deflection angle of the imager is calculated, and the depth coordinate is corrected based on the deflection angle.
［実施例］
以Ｆ、この発明の一実施例を図について説明する。第１
図はこの発明の一実施例による交通流工１測画像処理装
置を示す構成図であり、（７）は特徴点抽出回路、（８
）は消失点検出回路、（９）はＴＶカメラ（１）の振れ
角などを検出する撮像パラメタ検出回路、（１０）はス
クリーン座標から３次元座標を求める３次元座標変換回
路である。[Embodiment] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1st
The figure is a block diagram showing a traffic flow measurement image processing device according to an embodiment of the present invention, in which (7) is a feature point extraction circuit, (8) is a feature point extraction circuit;
) is a vanishing point detection circuit, (9) is an imaging parameter detection circuit that detects the deflection angle of the TV camera (1), and (10) is a three-dimensional coordinate conversion circuit that obtains three-dimensional coordinates from screen coordinates.
次に動作について説明する。ＴＶカメラ（１）は１”■
カメラ制御回路（２）に従って任意のフレーム間隔で走
１１車両を撮像する。フレームメモリ（３）は、Ｔ Ｖ
カメラ（１）の撮像信号をデジタルに変換したデータを
取り込む。重両抽出回路（４）は、フレ−ムメモリ（３
）のデータを処理し、車両を抽出する。例えば、第２図
は直線路を走行する車の様子を１”■カメラ（１）で撮
像したものである。（２１１ａ） 、 （２０ｂ）は
道路両端、（２１ａ）、 ｆ２１ｂｌは特徴点の一例と
して、市の後部ホンネットの最下部両端であり、（２２
）は消失点−である。第３図は、道路（２３）と平行に
２軸を取った３次元座標系を表した図で各軸は右手系で
取られている（ Ｏｗ −Ｘ ｗ Ｙ ｗ７、 ｗ ）
。第４図は、スクリーン（２４）を含むスクリン座標系
を表した図で各軸は左下系で取られている（ ＯＳ＝
Ｘ ｓ Ｙ ｓ Ｚ ｓ） 、第５図は゛ｒｖカメラの
設定条件によるスクリーン座標系と３次元座標系との関
係をｉトシた図で、（５０）はＹｗ回りの、Ｘｗに対す
るＸｓの角度で横振れ角と呼び、（５１）はＸｓ回りの
、Ｙｗに対するＹｓの角度で縦振れ角と呼ぶことに１−
る。第２図のように３次元の実１１↓界で互いに平行な
直線群を′ｒ■カメラ（１）で撮像すると、これらの直
線群は、１点で交わることが知られており、この交点は
消失点（２２）と呼ばれている。この消失点（２２）の
画面位置座標は、撮像型自身のパラメータ（素子サイズ
、焦点距離）と横振れ角（５ｆｌ ｌ、縦振れ角（５１
）及び画面サイズによって一意に決まる。逆に消失点（
２２）の画面位置座ケ５．撮像系パラメータ、及び画面
サイズが決まれば、撮像器の横振れ角（５０）、及び縦
振れ角（５１）は計算することができる。このとき撮像
系パラメータ、及び画面サイズは固定と考えられるので
、消失点（２２）の画面位置座標がわかれば、撮像器（
１）の横振れ角（５０）、及び縦振れ角（５りはただち
に求まる事になる。Next, the operation will be explained. TV camera (1) is 1”■
According to the camera control circuit (2), images of the 11 vehicles are taken at arbitrary frame intervals. Frame memory (3) is TV
Data obtained by converting the image signal of the camera (1) into digital data is imported. The overlap extraction circuit (4) is connected to the frame memory (3).
) data and extract vehicles. For example, Figure 2 shows a car traveling on a straight road, captured by a 1" camera (1). (211a) and (20b) are both ends of the road, and (21a) and f21bl are examples of feature points. , the lowermost ends of the rear honnet of the city, (22
) is the vanishing point -. Figure 3 is a diagram showing a three-dimensional coordinate system with two axes parallel to the road (23), and each axis is taken in a right-handed system (Ow -X w Y w7, w)
. Figure 4 is a diagram showing the screen coordinate system including the screen (24), and each axis is taken in the lower left system (OS=
X s Y s Z s), Figure 5 is a diagram showing the relationship between the screen coordinate system and the three-dimensional coordinate system according to the setting conditions of the rv camera, and (50) is the angle of Xs with respect to Xw around Yw. It is called the lateral deflection angle, and (51) is the angle of Ys with respect to Yw around Xs, which is called the vertical deflection angle.
Ru. It is known that when a group of straight lines that are parallel to each other in a three-dimensional real 11↓ field are imaged by camera (1) as shown in Figure 2, these straight lines intersect at one point, and this intersection point is called the vanishing point (22). The screen position coordinates of this vanishing point (22) are determined by the parameters of the imaging type itself (element size, focal length), horizontal shake angle (5fl l, vertical shake angle (51
) and the screen size. On the other hand, the vanishing point (
22) Screen position 5. Once the imaging system parameters and screen size are determined, the horizontal shake angle (50) and vertical shake angle (51) of the image pickup device can be calculated. At this time, the imaging system parameters and screen size are considered to be fixed, so if the screen position coordinates of the vanishing point (22) are known, the imager (
The horizontal deflection angle (50) and vertical deflection angle (5) in 1) can be found immediately.
第７図はこの実施例に係る特徴点抽出回路（７）と消失
点検出回路（８）の処理を示すフローチャトであり、ブ
ロック（３０）は特徴点抽出回路（７）における処理を
示し、ブロック（３りは消失点検出回路（８）における
処理を示す。特徴点抽出回路（７）において、ステップ
（３２）で第２図に示すようにｌフレームの走行車両を
撮像する。次に、ステップ（３３）で、最低２点の特徴
点のスクリーンにおける座標を求める。即ち、第２図の
スクリーンで走行車両の後部ボンネットの最下部両端（
２０ａ） 、 （２０ｂ）の座標を求める。ステップ
（３４）で必要シーン分あるかどうか判定し、必要シー
ン分以下の時は（モ意のフレーム間隔だけ時間遅れの後
（ステップ（３５））、ステップ（３２）で次のフレー
ムを撮像する。FIG. 7 is a flowchart showing the processing of the feature point extraction circuit (7) and the vanishing point detection circuit (8) according to this embodiment. Block (30) shows the processing in the feature point extraction circuit (7); (3 indicates the processing in the vanishing point detection circuit (8). In the feature point extraction circuit (7), in step (32), l-frame images of the traveling vehicle are taken as shown in FIG. 2. Next, step In (33), find the coordinates of at least two feature points on the screen. That is, on the screen of FIG.
20a) Find the coordinates of (20b). In step (34), it is determined whether there are enough scenes, and if it is less than the required number of scenes (after a time delay of the desired frame interval (step (35)), the next frame is captured in step (32). .
この実施例ではフレーム間隔を１７３０秒程度１撮像す
るが、特徴点を得るのは、これを間引いて６フレ一ム間
隔程度で行なっている。In this embodiment, one image is captured at a frame interval of about 1730 seconds, but feature points are obtained at intervals of about 6 frames by thinning out this.
ステップ（３４）の判定で必要シーン分あると判断した
時、後部ボンネットの最Ｆ部両端（２１１ａ） 、
（２０ｂｌ に関しての２組の特徴点列が得られること
になる。この特徴点列の画面座標を結ぶことによって出
来る軌跡は第６図における軌跡（６０） （ｉ’ｌ 、
（６１）（ｉ）［ｉ＝１．２． ・・ｎ］のよう
に直線（６２ａ） 、 （６２ｂ）となる。従って消
失点検出回路（８）では、ステップ（３５）で２組の特
徴点列に対して直線近似を行なう。さらにステップ（３
７）で２直線（５２ａＬ、 （５２ｂ）の交点を求め
て消失点（２２）とする。When it is determined in step (34) that there are enough scenes, both ends of the F-most part of the rear bonnet (211a),
(Two sets of feature point sequences regarding 20bl will be obtained.The trajectory created by connecting the screen coordinates of this feature point string is the trajectory (60) (i'l,
(61) (i) [i=1.2. ...n], resulting in straight lines (62a) and (62b). Therefore, the vanishing point detection circuit (8) performs linear approximation to the two sets of feature point sequences in step (35). Further steps (3
In step 7), find the intersection of the two straight lines (52aL, (52b)) and set it as the vanishing point (22).
道路上を走行する車は、狭い領域内では道路に平行にほ
ぼ直線的に移動すると考えられる。従って車両自身のあ
る特徴点に着目し、複数シーンにまたがってその画面上
での座標１直を求め、それらの点を直線近似することに
よって一つの直線が得られる。第６図で示すように、２
つの特徴点の軌跡（６０ｉ） 、 ｆ６１ｉ１ ［ｉ
＝１．２． ・・＋、 ｎ ］が描く２つの直線（
６２ａ） 、 （６２ｂ）の交点を計算すれば消失点
（２２）の画面座標が求まることになる。A car traveling on a road is considered to move in a substantially straight line parallel to the road within a narrow area. Therefore, one straight line can be obtained by focusing on a certain feature point of the vehicle itself, finding its coordinates on the screen across multiple scenes, and linearly approximating those points. As shown in Figure 6, 2
The locus of one feature point (60i), f61i1 [i
=1.2. ...+, n ] drawn by two straight lines (
By calculating the intersection of 62a) and (62b), the screen coordinates of the vanishing point (22) can be found.
次に、撮像パラメータ検出回路（９）は消失点座標から
ＴＶ右カメラ１）の槓振れ角（５ａ）、縦振れ角（５１
）を求める。３次元座標変換回路（１０）は、］゛■カ
■カメラの焦点距離、素子サイズ、３次元位置座標、横
振れ角、縦振れ角を用いて、画面上の任意の座標に対し
て、その座標の３次元座標での高さを仮定して（例えば
、路面上ならＯｍとする）、その座標の３次元の奥行き
座標を計算する。特徴屯抽出から撮像パラメータ検出ま
での一連の処理は、一定の周期で繰り返される。車両情
報計測回路（６） は、複数シーンにわたって、車両抽
出回路（４）で得られた車両の実際の位置を、３次元座
標変換回路（１０）を用いて求め、それらの情報より各
小雨の速度や車間距離を計測する。Next, the imaging parameter detection circuit (9) calculates the oscillation angle (5a) and the vertical vibration angle (51) of the TV right camera 1) from the vanishing point coordinates.
). The three-dimensional coordinate conversion circuit (10) uses the camera focal length, element size, three-dimensional position coordinates, horizontal shake angle, and vertical shake angle to convert arbitrary coordinates on the screen. Assuming the height of the coordinate in three-dimensional coordinates (for example, Om if it is on the road surface), calculate the three-dimensional depth coordinate of that coordinate. A series of processes from feature extraction to imaging parameter detection are repeated at regular intervals. The vehicle information measurement circuit (6) calculates the actual position of the vehicle obtained by the vehicle extraction circuit (4) over multiple scenes using the three-dimensional coordinate conversion circuit (10), and uses the information to determine the actual position of the vehicle obtained by the vehicle extraction circuit (4). Measures speed and distance between vehicles.
消失点を用いてＴ’ Ｖカメラ（Ｉｌｌの横振れ角、縦
振れ角を求める方法、ｊｌＤびに画面上の任意の座標に
対して、その座標の３次元ＵＫ標での高さを仮定して、
その座標の３次元の奥行き座標を計算する方法について
は、例えば、山［Ｊ冨士夫著「コンピュータデイスプレ
ィによる図形処理工学」、日刊工業斬１１ｆ１社（１９
８１Ｑｌに詳しく紹介されている。ここではその説明に
ついでは省略する。A method for finding the horizontal and vertical deflection angles of a T'V camera (Ill) using the vanishing point. ,
Regarding the method of calculating the three-dimensional depth coordinates of the coordinates, for example, see Yama [J.
It is introduced in detail in 81Ql. The explanation thereof will be omitted here.
」二記実施例では、撮像器のパラメータを自動的に求め
ることを可能にしたので、１゛Ｖカメラの角度変化が生
じても奥行き情報を逐次求めることができ、これを用い
て車速なとの４測精度を向上できると共に、メインテナ
ンスを容易にできる。In the second embodiment, it is possible to automatically obtain the parameters of the imager, so even if the angle of the camera changes by 1゛V, depth information can be obtained sequentially, and this can be used to determine the vehicle speed. 4 measurement accuracy can be improved and maintenance can be facilitated.
なお、上記実施例では、特徴点に車両の後部ボンネット
の最下部両端を選んだが、テールランプやヘッドランプ
など、特徴点として抽出されるものならなんでもよく、
また道路との羽行性があるものなら、車両の持つ特徴点
以外の特徴点を用いてよい。この場合、例えば街路灯の
列などを用いる場合は、複数フレームから軌跡を求める
必要はなく、ｌフレームで実現できる。In the above embodiment, the lowermost ends of the rear hood of the vehicle were selected as the feature points, but any feature point that can be extracted as a feature point may be used, such as a tail lamp or a headlamp.
Further, as long as there is a relationship with the road, feature points other than the feature points of the vehicle may be used. In this case, for example, when using a row of street lights, it is not necessary to obtain the trajectory from multiple frames, and it can be realized using one frame.
また、消失点を求めるにあたって、２木の直線の交点と
したが５２個以上の特徴点を設定し、複数組の交点を用
いて求めてもよい。Further, in determining the vanishing point, the intersection of two straight lines is used, but 52 or more feature points may be set and a plurality of sets of intersection points may be used to determine the vanishing point.
また５、に記実施例では特徴点抽出回路（７）と消失点
検出回路（８）をソフトウェアによって構成したが、ハ
ードウェアで実現してもよい。Further, in the embodiment described in 5., the feature point extraction circuit (7) and the vanishing point detection circuit (8) are configured by software, but they may also be implemented by hardware.
［発明の効果］
以上のように、この発明によれば、車両を撮像し、その
撮像信号を処理して交通流をａｔ ｌｉｌ’ｌする交通
流計測画像処理装置において、走行車両を任意のフレー
ム間隔で撮像する撮像器、走行車両の特徴点を複数個設
定し、撮像信号から特徴点を抽出する特徴点抽出回路、
複数フレームにわたって同一・特徴点を抽出して特徴点
列とし、この特徴点列の画面座標を結ぶことによって出
来る軌跡を特徴点の各々に対して求め、それらの軌跡群
から撮像器の消失°点を求める消失点検出回路、及び消
失点を用いて撮像器の振れ角を求める撮像パラメータ検
出回路を備えたことにより、撮像器のパラメタを自動的
に求めることを司能にしたので、ゴＶカメラの角度変化
が生しても奥行き情報を逐次求、めることができ、それ
により重速なとの、ｅｌ” １ｌｌｌｌ精度を上げたり
、メインテナンスを容易にできる交通流計測画像処理装
置が得られる効果がある。[Effects of the Invention] As described above, according to the present invention, in the traffic flow measurement image processing device that images a vehicle and processes the imaged signal to determine the traffic flow, a traveling vehicle can be captured in an arbitrary frame. An imager that captures images at intervals, a feature point extraction circuit that sets multiple feature points of a running vehicle and extracts feature points from the image signal;
Extract the same feature points over multiple frames to create a feature point sequence, find a trajectory for each feature point by connecting the screen coordinates of this feature point sequence, and calculate the vanishing point of the imager from these trajectory groups. Equipped with a vanishing point detection circuit that calculates the image sensor and an imaging parameter detection circuit that uses the vanishing point to calculate the deflection angle of the imager, it is possible to automatically calculate the imager parameters. Depth information can be obtained sequentially even when angle changes occur, thereby providing a traffic flow measurement image processing device that can increase precision even at high speeds and facilitate maintenance. effective.
第１図はこの発明の一実施例による交通流計？■１画像
処理装置を示す構成図、第２図は道路上を走行才る車を
′］゛Ｖカメラで撮像した画像を示す説明図５第３図は
３次元座標系を示す説明図、第４図はスクリーン座標系
を示す説明図、第５１￥１はＩ”Ｖカメラの横振れ角、
縦振れ角を示す説明図、第６図は消失点を求める方法を
示す説明図、第７図は実施例に係る特徴点抽出回路と消
失点検出回路における処理を示すフローチャート、第８
図は従来の交通流計測画像処理装置を示す構成図、第９
図（ａｌ 、 ｆｂｌ は従来の奥行き情報の作成方
法を示す説明図である。
（１）・・パ１”■カメラ、（７）・・・特徴点抽出回
路、（８） ・・・消失点検出回路、（９）・・・撮
像バラ・メータ検出回路。
なお、図中、同一符号は同一、または相当部分を示す。Figure 1 shows a traffic flow meter according to an embodiment of this invention. ■1 A configuration diagram showing the image processing device, Figure 2 is an explanatory diagram showing an image taken by a V camera, and Figure 3 is an explanatory diagram showing a three-dimensional coordinate system. Figure 4 is an explanatory diagram showing the screen coordinate system, No. 51 ¥1 is the horizontal deflection angle of the I''V camera,
FIG. 6 is an explanatory diagram showing the method of determining the vanishing point; FIG. 7 is a flowchart showing processing in the feature point extraction circuit and the vanishing point detection circuit according to the embodiment; FIG.
Figure 9 is a configuration diagram showing a conventional traffic flow measurement image processing device.
Figures (al and fbl) are explanatory diagrams showing the conventional method of creating depth information. (1)...Pa1"■ camera, (7)...Feature point extraction circuit, (8)...Disappearance inspection Output circuit, (9)...imaging variable meter detection circuit. In the drawings, the same reference numerals indicate the same or equivalent parts.
Claims (1)
る交通流計測画像処理装置において、走行車両を任意の
フレーム間隔で撮像する撮像器、上記走行車両の特徴点
を複数個設定し、上記撮像信号から上記特徴点を抽出す
る特徴点抽出回路、複数フレームにわたって同一特徴点
を抽出して特徴点列とし、この特徴点列の画面座標を結
ぶことによって出来る軌跡を上記特徴点の各々に対して
求め、それらの軌跡群から上記撮像器の消失点を求める
消失点検出回路、及び上記消失点を用いて上記撮像器の
振れ角を求める撮像パラメータ検出回路を備えた交通流
計測画像処理装置。A traffic flow measurement image processing device that captures an image of a vehicle and processes the captured image signal to measure traffic flow includes an imager that captures an image of a traveling vehicle at arbitrary frame intervals, a plurality of characteristic points of the traveling vehicle, and A feature point extraction circuit extracts the feature points from the imaging signal, extracts the same feature points over multiple frames to form a feature point sequence, and creates a trajectory for each of the feature points by connecting the screen coordinates of this feature point sequence. A traffic flow measurement image processing device comprising: a vanishing point detection circuit for determining the vanishing point of the image pickup device from a group of trajectories; and an imaging parameter detection circuit for determining the deflection angle of the image pickup device using the vanishing point. .
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP12093190A JPH0417099A (en) | 1990-05-10 | 1990-05-10 | Traffic flow measuring picture processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP12093190A JPH0417099A (en) | 1990-05-10 | 1990-05-10 | Traffic flow measuring picture processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH0417099A true JPH0417099A (en) | 1992-01-21 |
Family
ID=14798519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP12093190A Pending JPH0417099A (en) | 1990-05-10 | 1990-05-10 | Traffic flow measuring picture processor |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0417099A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003050107A (en) * | 2001-08-07 | 2003-02-21 | Matsushita Electric Ind Co Ltd | Camera calibration device |
| JP2010041322A (en) * | 2008-08-04 | 2010-02-18 | Sumitomo Electric Ind Ltd | Mobile object identification device, image processing apparatus, computer program and method of specifying optical axis direction |
| JP2013191073A (en) * | 2012-03-14 | 2013-09-26 | Mitsubishi Heavy Ind Ltd | Vehicle size measurement device, vehicle size measurement method and program |
| JP2014115793A (en) * | 2012-12-10 | 2014-06-26 | Mitsubishi Electric Corp | Image processing apparatus and image processing method |
-
1990
- 1990-05-10 JP JP12093190A patent/JPH0417099A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003050107A (en) * | 2001-08-07 | 2003-02-21 | Matsushita Electric Ind Co Ltd | Camera calibration device |
| JP2010041322A (en) * | 2008-08-04 | 2010-02-18 | Sumitomo Electric Ind Ltd | Mobile object identification device, image processing apparatus, computer program and method of specifying optical axis direction |
| JP2013191073A (en) * | 2012-03-14 | 2013-09-26 | Mitsubishi Heavy Ind Ltd | Vehicle size measurement device, vehicle size measurement method and program |
| JP2014115793A (en) * | 2012-12-10 | 2014-06-26 | Mitsubishi Electric Corp | Image processing apparatus and image processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5615441B2 (en) | Image processing apparatus and image processing method | |
| JP3167752B2 (en) | Vehicle distance detection device | |
| JP2000011133A (en) | Device and method for detecting moving object | |
| JP2007078358A (en) | Measurement device for mobile body, signal control system, measurement method of mobile body, and computer program | |
| JP3709879B2 (en) | Stereo image processing device | |
| JPH10269362A (en) | Object recognition method and device therefor | |
| JP3562250B2 (en) | Leading vehicle detection device | |
| JPH08136237A (en) | Device for calculating gradient of road and car speed controller | |
| JP2014130429A (en) | Photographing device and three-dimensional object area detection program | |
| JPH08172620A (en) | Image input means for vehicle | |
| JP3516118B2 (en) | Object recognition method and object recognition device | |
| JP3629935B2 (en) | Speed measurement method for moving body and speed measurement device using the method | |
| JPH0948298A (en) | Object position measuring method in running passage | |
| JPH0417099A (en) | Traffic flow measuring picture processor | |
| JP3465531B2 (en) | Object recognition method and apparatus | |
| JP3501841B2 (en) | Three-dimensional object region detection device, distance measuring device to three-dimensional object region, and their detection and measurement method | |
| JPH10512953A (en) | Optical range and speed detection system | |
| JPH1141521A (en) | Image pickup device, instrument and method for measuring distance between vehicles | |
| WO2018034304A1 (en) | Image processing system | |
| JPH09145368A (en) | Moving and tracing method for object by stereoscopic image | |
| JPH0757200A (en) | Method and device for recognizing travel course | |
| JPH10283478A (en) | Method for extracting feature and and device for recognizing object using the same method | |
| JP3419919B2 (en) | Moving object detection device | |
| JP3487045B2 (en) | Automatic tracking device | |
| JPH11190611A (en) | Three-dimensional measuring method and three-dimensional measuring processor using this method |