JP2019211402A - Vehicle measurement device and vehicle measurement program - Google Patents
Vehicle measurement device and vehicle measurement program Download PDFInfo
- Publication number
- JP2019211402A JP2019211402A JP2018109792A JP2018109792A JP2019211402A JP 2019211402 A JP2019211402 A JP 2019211402A JP 2018109792 A JP2018109792 A JP 2018109792A JP 2018109792 A JP2018109792 A JP 2018109792A JP 2019211402 A JP2019211402 A JP 2019211402A
- Authority
- JP
- Japan
- Prior art keywords
- vehicle
- image
- radar
- reflection point
- frustum
- 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
Links
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Traffic Control Systems (AREA)
- Radar Systems Or Details Thereof (AREA)
- Image Analysis (AREA)
Abstract
Description
本開示は、車両の三次元情報を計測する技術に関する。 The present disclosure relates to a technique for measuring three-dimensional information of a vehicle.
車両の三次元情報を計測する技術として、ステレオカメラ又はＬｉＤＡＲ（Ｌａｓｅｒ Ｉｍａｇｉｎｇ Ｄｅｔｅｃｔｉｏｎ ａｎｄ Ｒａｎｇｉｎｇ）を用いる技術が存在する。しかし、ステレオカメラを用いる技術では、ステレオカメラから遠距離になるほど、測距精度が低下する。そして、ステレオカメラ又はＬｉＤＡＲを用いる技術では、天候依存度が大きいため、霧及び雨等の悪天候時に対応することができない。 As a technique for measuring three-dimensional information of a vehicle, there is a technique using a stereo camera or LiDAR (Laser Imaging Detection and Ranging). However, in the technology using a stereo camera, the distance measurement accuracy decreases as the distance from the stereo camera increases. And since the technology using a stereo camera or LiDAR has a large degree of weather dependence, it cannot cope with bad weather such as fog and rain.
ところで、特許文献１に開示された技術として、ミリ波レーダ及び単眼カメラを併用する技術が存在する。よって、特許文献１に開示された技術では、レーダ装置から遠距離になっても、測距精度が向上する。そして、特許文献１に開示された技術では、天候依存度が小さいため、霧及び雨等の悪天候時に対応することができる。しかし、特許文献１に開示された技術では、レーダ装置から車両の正面までの距離及び車両の正面を縁取る矩形を検出するのみであり、車両の三次元情報を計測することができない。 By the way, as a technique disclosed in Patent Document 1, there is a technique using both a millimeter wave radar and a monocular camera. Therefore, with the technique disclosed in Patent Document 1, ranging accuracy is improved even when the distance from the radar apparatus is increased. And since the technique disclosed by patent document 1 has little weather dependence, it can respond at the time of bad weather, such as fog and rain. However, the technique disclosed in Patent Document 1 only detects the distance from the radar device to the front of the vehicle and the rectangle that borders the front of the vehicle, and cannot measure the three-dimensional information of the vehicle.
そこで、前記課題を解決するために、本開示は、ミリ波レーダ等及び単眼カメラ等を併用することにより、車両の三次元情報を計測することを目的とする。 Therefore, in order to solve the above-described problems, the present disclosure aims to measure three-dimensional information of a vehicle by using a millimeter wave radar or the like and a monocular camera or the like together.
ミリ波レーダ等では、車両の幅及び奥行といった二次元情報を計測するのみである。単眼カメラ等では、車両の幅及び高さといった二次元情報を計測するのみである。前記課題を解決するために、ミリ波レーダ等及び単眼カメラ等を併用して、各々に不足する次元の情報を互いに補完して、車両の幅、奥行及び高さといった三次元情報を計測する。 A millimeter wave radar or the like only measures two-dimensional information such as the width and depth of a vehicle. A monocular camera or the like only measures two-dimensional information such as the width and height of the vehicle. In order to solve the above-mentioned problems, three-dimensional information such as the width, depth, and height of a vehicle is measured by using a millimeter wave radar and a monocular camera together to supplement each other with insufficient dimension information.
具体的には、本開示は、車両からのレーダ反射点の情報及び前記車両の画像を取得する車両情報取得部と、前記画像上で前記レーダ反射点をプロットする反射点プロット部と、前記画像上で前記レーダ反射点の近傍でのエッジ解析を実行することにより、前記画像上で前記車両を縁取る四角錐台を検出する四角錐台検出部と、前記画像上の前記四角錐台の位置を実空間内の位置に変換する位置変換部と、を備えることを特徴とする車両計測装置である。 Specifically, the present disclosure relates to a vehicle information acquisition unit that acquires information on radar reflection points from a vehicle and an image of the vehicle, a reflection point plot unit that plots the radar reflection points on the image, and the image. By performing edge analysis in the vicinity of the radar reflection point above, a quadrangular frustum detector for detecting a quadrangular frustum that borders the vehicle on the image, and a position of the quadrangular frustum on the image And a position conversion unit that converts the position to a position in real space.
この構成によれば、ミリ波レーダ等及び単眼カメラ等を併用することにより、車両の幅、奥行及び高さといった三次元情報を計測することができる。なお、車両の幅及び奥行といった二次元情報を計測するのみであるレーダに代えて、車両の幅、奥行及び高さといった三次元情報を計測することができるレーダを用いてもよい。また、車両の幅及び高さといった二次元情報を計測するのみである単眼カメラに代えて、車両の幅、高さ及び奥行といった三次元情報を計測することができるステレオカメラを用いてもよい。 According to this configuration, three-dimensional information such as the width, depth, and height of the vehicle can be measured by using a millimeter wave radar or the like and a monocular camera or the like together. Instead of the radar that only measures two-dimensional information such as the width and depth of the vehicle, a radar that can measure three-dimensional information such as the width, depth, and height of the vehicle may be used. Further, instead of the monocular camera that only measures two-dimensional information such as the width and height of the vehicle, a stereo camera that can measure three-dimensional information such as the width, height, and depth of the vehicle may be used.
また、本開示は、前記反射点プロット部は、前記レーダ反射点が路面上にあると仮定し、前記四角錐台検出部は、前記反射点プロット部が前記路面上にあると仮定した前記レーダ反射点に基づいて、前記四角錐台の底面の辺を検出することを特徴とする車両計測装置である。 The present disclosure also assumes that the reflection point plot unit assumes that the radar reflection point is on the road surface, and the quadrangular frustum detection unit assumes that the reflection point plot unit is on the road surface. The vehicle measuring device is characterized in that the side of the bottom surface of the quadrangular frustum is detected based on a reflection point.
レーダ反射点の高さは、レーダ装置の設置高さの周りで不明である。この構成によれば、レーダ反射点の高さを路面の高さ（レーダ装置の設置高さに近い）に等しいと仮定することにより、車両を縁取る四角錐台の底面の辺を容易に検出することができる。なお、車両の下側のエッジ解析を実行しないのは、車両の下側のエッジと路面との間にタイヤの半径程度の高さの空間があり、車両の高さを精度高く計測することができないからである。 The height of the radar reflection point is unknown around the installation height of the radar device. According to this configuration, by assuming that the height of the radar reflection point is equal to the height of the road surface (close to the installation height of the radar device), the side of the bottom of the quadrangular frustum that borders the vehicle can be easily detected. can do. The reason why the lower edge analysis of the vehicle is not performed is that there is a space with a height about the radius of the tire between the lower edge of the vehicle and the road surface, so that the height of the vehicle can be measured with high accuracy. It is not possible.
また、本開示は、前記四角錐台検出部は、レーダ装置から遠距離の前記レーダ反射点に基づかず、前記レーダ装置から近距離の前記レーダ反射点に基づいて、前記四角錐台の底面の辺を検出することを特徴とする車両計測装置である。 Further, according to the present disclosure, the square frustum detection unit is not based on the radar reflection point at a long distance from the radar device, but based on the radar reflection point at a short distance from the radar device. A vehicle measurement device that detects a side.
レーダ反射点の高さは、レーダ装置の設置高さの周りで不明である。ここで、レーダ装置から遠距離のレーダ反射点の高さは、レーダ装置の設置高さの周りで大きな誤差を有する。一方で、レーダ装置から近距離のレーダ反射点の高さは、レーダ装置の設置高さの周りで大きな誤差を有さない。この構成によれば、レーダ装置から遠距離のレーダ反射点に基づかず、レーダ装置から近距離のレーダ反射点に基づいて、車両を縁取る四角錐台の底面の辺を精度高く検出することができる。そして、車両の奥行方向の辺の傾きを精度高く計測することにより、車両の奥行方向の長さを精度高く計測することができる。 The height of the radar reflection point is unknown around the installation height of the radar device. Here, the height of the radar reflection point far from the radar apparatus has a large error around the installation height of the radar apparatus. On the other hand, the height of the radar reflection point at a short distance from the radar apparatus does not have a large error around the installation height of the radar apparatus. According to this configuration, it is possible to accurately detect the side of the bottom of the quadrangular frustum that borders the vehicle based on the radar reflection point at a short distance from the radar device, based on the radar reflection point at a short distance from the radar device. it can. The length of the vehicle in the depth direction can be measured with high accuracy by measuring the inclination of the side in the depth direction of the vehicle with high accuracy.
また、本開示は、前記四角錐台検出部は、前記四角錐台の底面の辺から前記画像上の高さ方向の所定距離でのエッジ解析を実行することにより、前記四角錐台の上面の辺を検出することを特徴とする車両計測装置である。 Further, according to the present disclosure, the quadrangular frustum detection unit performs an edge analysis at a predetermined distance in a height direction on the image from a side of a bottom surface of the quadrangular frustum, so that an upper surface of the square frustum is detected. A vehicle measurement device that detects a side.
この構成によれば、車両を縁取る四角錐台の上面の辺を検出することができる。 According to this configuration, it is possible to detect the side of the upper surface of the truncated pyramid that borders the vehicle.
また、本開示は、前記四角錐台検出部は、前記四角錐台の底面の二辺の交点から前記画像上の高さ方向に延伸する直線に基づいて、前記四角錐台の正面と側面との境界の辺を検出することを特徴とする車両計測装置である。 Further, according to the present disclosure, the quadrangular frustum detection unit includes a front surface and a side surface of the quadrangular frustum based on a straight line extending in the height direction on the image from the intersection of two sides of the bottom surface of the quadrangular frustum. It is a vehicle measuring device characterized by detecting a side of the boundary.
この構成によれば、車両を縁取る四角錐台の正面と側面との境界の辺を検出することができる。 According to this configuration, it is possible to detect the side of the boundary between the front surface and the side surface of the quadrangular frustum that borders the vehicle.
また、本開示は、前記四角錐台検出部は、前記四角錐台の正面と側面との境界の辺から前記画像上の幅方向の所定距離でのエッジ解析を実行することにより、前記四角錐台の正面の四辺のうち、前記四角錐台の正面と側面との境界の辺と向かい合う辺を検出することを特徴とする車両計測装置である。 Further, according to the present disclosure, the quadrangular pyramid detection unit performs the edge analysis at a predetermined distance in the width direction on the image from a side of a boundary between the front surface and the side surface of the quadrangular pyramid. Of the four sides of the front of the base, the vehicle measuring device is characterized in that the side facing the side of the boundary between the front and side of the quadrangular pyramid is detected.
この構成によれば、車両を縁取る四角錐台の正面の四辺のうち、四角錐台の正面と側面との境界の辺と向かい合う辺を検出することができる。 According to this configuration, of the four sides on the front of the quadrangular frustum that borders the vehicle, it is possible to detect a side that faces the side of the boundary between the front and side of the quadrangular frustum.
また、本開示は、前記四角錐台検出部は、前記車両の最前方の前記レーダ反射点から前記画像上の奥行方向の所定距離でのエッジ解析を実行することにより、前記四角錐台の側面の四辺のうち、前記四角錐台の正面と側面との境界の辺と向かい合う辺を検出することを特徴とする車両計測装置である。 Further, according to the present disclosure, the quadrangular frustum detection unit performs an edge analysis at a predetermined distance in the depth direction on the image from the radar reflection point at the forefront of the vehicle. Among the four sides, a vehicle measuring device that detects a side facing a boundary side between a front surface and a side surface of the quadrangular frustum.
この構成によれば、車両を縁取る四角錐台の側面の四辺のうち、四角錐台の正面と側面との境界の辺と向かい合う辺を検出することができる。 According to this configuration, of the four sides on the side of the quadrangular frustum that borders the vehicle, it is possible to detect the side that faces the side of the boundary between the front and side of the quadrangular frustum.
また、本開示は、前記四角錐台検出部は、レーダ装置から前記車両までの距離が近いほど、エッジ解析を実行する前記画像上の前記所定距離の範囲を広くし、前記レーダ装置から前記車両までの距離が遠いほど、エッジ解析を実行する前記画像上の前記所定距離の範囲を狭くすることを特徴とする車両計測装置である。 Further, according to the present disclosure, as the distance from the radar device to the vehicle is shorter, the square pyramid detection unit widens the range of the predetermined distance on the image on which the edge analysis is performed, and from the radar device to the vehicle. The range of the predetermined distance on the image on which the edge analysis is performed is narrowed as the distance to the vehicle increases.
レーダ装置から近距離の車両のサイズは、画像上で大きい。一方で、レーダ装置から遠距離の車両のサイズは、画像上で小さい。この構成によれば、レーダ装置から車両までの距離に応じて、エッジ解析を実行する画像上の範囲を調整し、車両を縁取る四角錐台の上面、正面及び／又は側面の辺を容易に検出することができる。 The size of the vehicle at a short distance from the radar device is large on the image. On the other hand, the size of the vehicle at a long distance from the radar device is small on the image. According to this configuration, the range on the image on which the edge analysis is performed is adjusted according to the distance from the radar device to the vehicle, and the top, front, and / or side edges of the quadrangular pyramid that borders the vehicle can be easily obtained. Can be detected.
また、本開示は、車両からのレーダ反射点の情報及び前記車両の画像を取得する車両情報取得ステップと、前記画像上で前記レーダ反射点をプロットする反射点プロットステップと、前記画像上で前記レーダ反射点の近傍でのエッジ解析を実行することにより、前記画像上で前記車両を縁取る四角錐台を検出する四角錐台検出ステップと、前記画像上の前記四角錐台の位置を実空間内の位置に変換する位置変換ステップと、を順にコンピュータに実行させるための車両計測プログラムである。 The present disclosure also includes a vehicle information acquisition step for acquiring radar reflection point information from a vehicle and an image of the vehicle, a reflection point plotting step for plotting the radar reflection point on the image, and the image on the image. By performing edge analysis in the vicinity of a radar reflection point, a quadrangular frustum detection step for detecting a quadrangular frustum that borders the vehicle on the image, and a position of the quadrangular frustum on the image in real space It is a vehicle measurement program for making a computer perform the position conversion step which converts into the position in the inside in order.
この構成によれば、ミリ波レーダ等及び単眼カメラ等を併用することにより、車両の幅、奥行及び高さといった三次元情報を計測することができる。なお、車両の幅及び奥行といった二次元情報を計測するのみであるレーダに代えて、車両の幅、奥行及び高さといった三次元情報を計測することができるレーダを用いてもよい。また、車両の幅及び高さといった二次元情報を計測するのみである単眼カメラに代えて、車両の幅、高さ及び奥行といった三次元情報を計測することができるステレオカメラを用いてもよい。 According to this configuration, three-dimensional information such as the width, depth, and height of the vehicle can be measured by using a millimeter wave radar or the like and a monocular camera or the like together. Instead of the radar that only measures two-dimensional information such as the width and depth of the vehicle, a radar that can measure three-dimensional information such as the width, depth, and height of the vehicle may be used. Further, instead of the monocular camera that only measures two-dimensional information such as the width and height of the vehicle, a stereo camera that can measure three-dimensional information such as the width, height, and depth of the vehicle may be used.
このように、本開示は、ミリ波レーダ等及び単眼カメラ等を併用することにより、車両の三次元情報を計測することができる。 As described above, the present disclosure can measure three-dimensional information of a vehicle by using a millimeter wave radar or the like and a monocular camera or the like together.
添付の図面を参照して本開示の実施形態を説明する。以下に説明する実施形態は本開示の実施の例であり、本開示は以下の実施形態に制限されるものではない。 Embodiments of the present disclosure will be described with reference to the accompanying drawings. The embodiments described below are examples of the present disclosure, and the present disclosure is not limited to the following embodiments.
本開示の車両計測システムの構成を示すブロック図を図１に示す。本開示の車両計測装置の処理手順を示すフローチャートを図２に示す。車両計測システムＣは、レーダ装置１、撮像装置２及び車両計測装置３から構成される。車両計測装置３は、図２に示した車両計測プログラムをインストールされたコンピュータであり、車両情報取得部３１、反射点プロット部３２、四角錐台検出部３３及び位置変換部３４から構成される。 A block diagram showing the configuration of the vehicle measurement system of the present disclosure is shown in FIG. A flowchart showing a processing procedure of the vehicle measurement device of the present disclosure is shown in FIG. The vehicle measurement system C includes a radar device 1, an imaging device 2, and a vehicle measurement device 3. The vehicle measurement device 3 is a computer installed with the vehicle measurement program shown in FIG.
レーダ装置１は、自車のバンパーの内部等に設置される、ＭＩＭＯ（Ｍｕｌｔｉｐｌｅ Ｉｎｐｕｔ−Ｍｕｌｔｉｐｌｅ Ｏｕｔｐｕｔ）ミリ波レーダ等である。撮像装置２は、自車のフロントガラスの上部又は自車のバンパーの内部等に設置される、単眼カメラ等である。車両計測装置３は、レーダ装置１及び撮像装置２を併用することにより、車両の三次元情報を計測する。 The radar apparatus 1 is a MIMO (Multiple Input-Multiple Output) millimeter wave radar or the like installed inside a bumper of the own vehicle. The imaging device 2 is a monocular camera or the like installed on the upper part of the windshield of the own vehicle or inside the bumper of the own vehicle. The vehicle measurement device 3 measures the three-dimensional information of the vehicle by using the radar device 1 and the imaging device 2 together.
本開示の車両情報取得の処理内容を図３に示す。車両情報取得部３１は、車両からのレーダ反射点の情報及び車両の画像を取得する（ステップＳ１）。 The details of the vehicle information acquisition process of the present disclosure are shown in FIG. The vehicle information acquisition unit 31 acquires radar reflection point information from the vehicle and a vehicle image (step S1).
車両からのレーダ反射点の情報は、ビームフォーマ法又はＭＵＳＩＣ（Ｍｕｌｔｉｐｌｅ Ｓｉｇｎａｌ Ｃｌａｓｓｉｆｉｃａｔｉｏｎ）法等により取得される。車両情報取得部３１は、複数の白線Ｌの情報を取得している。レーダ装置１のアンテナＡから見て、左側の隣接車線には、複数のレーダ反射点のクラスタとして車両Ｃ１が検出され、同一の車線には、複数のレーダ反射点のクラスタとして車両Ｃ２が検出され、右側の隣接車線には、複数のレーダ反射点のクラスタとして車両Ｃ３が検出される。 Information on the radar reflection point from the vehicle is acquired by a beamformer method, a MUSIC (Multiple Signal Classification) method, or the like. The vehicle information acquisition unit 31 acquires information on a plurality of white lines L. When viewed from the antenna A of the radar apparatus 1, the vehicle C1 is detected as a cluster of a plurality of radar reflection points in the left adjacent lane, and the vehicle C2 is detected as a cluster of a plurality of radar reflection points in the same lane. In the adjacent lane on the right side, the vehicle C3 is detected as a cluster of a plurality of radar reflection points.
車両の画像は、単眼カメラ等により取得される。撮像装置２のセンサから見て、前方の方向には、複数の白線Ｌが検出され、左側の隣接車線には、車両Ｃ１が検出され、同一の車線には、車両Ｃ２が検出され、右側の隣接車線には、車両Ｃ３が検出される。 The vehicle image is acquired by a monocular camera or the like. When viewed from the sensor of the imaging device 2, a plurality of white lines L are detected in the forward direction, the vehicle C1 is detected in the left adjacent lane, the vehicle C2 is detected in the same lane, and the right side is detected. The vehicle C3 is detected in the adjacent lane.
レーダ装置１は、複数のアンテナＡを高さ方向に離して設置する余裕がないため、車両の幅Ｘ［ｍ］及び奥行Ｚ［ｍ］といった二次元情報を計測するのみである。撮像装置２は、複数のセンサを幅方向に離して設置する余裕がないため、車両の幅ｘ［ｐｉｘ］及び高さｙ［ｐｉｘ］といった二次元情報を計測するのみである。そこで、車両計測装置３は、レーダ装置１及び撮像装置２を併用して、各々に不足する次元の情報を互いに補完して、車両の幅Ｘ［ｍ］、奥行Ｚ［ｍ］及び高さＹ［ｍ］といった三次元情報を計測する。 Since the radar apparatus 1 cannot afford to install a plurality of antennas A apart in the height direction, the radar apparatus 1 only measures two-dimensional information such as the width X [m] and the depth Z [m] of the vehicle. Since the imaging device 2 cannot afford to install a plurality of sensors apart in the width direction, it only measures two-dimensional information such as the vehicle width x [pix] and the height y [pix]. Therefore, the vehicle measurement device 3 uses the radar device 1 and the imaging device 2 together to supplement each other with insufficient dimension information, so that the vehicle width X [m], depth Z [m], and height Y Three-dimensional information such as [m] is measured.
本開示の反射点プロット及び底面の辺検出の処理内容を図４及び図５に示す。反射点プロット部３２は、画像上でレーダ反射点をプロットする（ステップＳ２）。四角錐台検出部３３は、画像上で車両を縁取る四角錐台の底面の辺を検出する（ステップＳ３）。 The processing contents of the reflection point plot and the bottom edge detection of the present disclosure are shown in FIGS. The reflection point plot unit 32 plots radar reflection points on the image (step S2). The quadrangular frustum detection unit 33 detects the side of the bottom of the quadrangular frustum that borders the vehicle on the image (step S3).
車両Ｃ１からのレーダ反射点の情報では、車両Ｃ１の正面には、複数のレーダ反射点Ｐ１、Ｐ２、Ｐ３が検出され、車両Ｃ１の側面には、複数のレーダ反射点Ｐ４、Ｐ５、Ｐ６、Ｐ７が検出される。車両Ｃ１の画像では、車両Ｃ１の近傍に、複数のレーダ反射点Ｐ１〜Ｐ７がプロットされる。ここで、図１０に示した画像上の位置を実空間内の位置に変換する数式に対して、逆処理である実空間内の位置を画像上の位置に変換する数式を用いて、複数のレーダ反射点Ｐ１〜Ｐ７が車両Ｃ１の画像上にプロットされる。 In the radar reflection point information from the vehicle C1, a plurality of radar reflection points P1, P2, and P3 are detected in front of the vehicle C1, and a plurality of radar reflection points P4, P5, P6, P7 is detected. In the image of the vehicle C1, a plurality of radar reflection points P1 to P7 are plotted in the vicinity of the vehicle C1. Here, in contrast to the mathematical expression for converting the position on the image shown in FIG. 10 to the position on the real space, a mathematical expression for converting the position on the real space to the position on the image, which is the reverse process, is used. Radar reflection points P1 to P7 are plotted on the image of the vehicle C1.
図４に示した車両Ｃ１の画像では、反射点プロット部３２は、複数のレーダ反射点Ｐ１〜Ｐ７が路面上にあると仮定する。というのは、複数のレーダ反射点Ｐ１〜Ｐ７の高さは、レーダ装置１の設置高さの周りで不明である。そこで、複数のレーダ反射点Ｐ１〜Ｐ７の高さは、路面の高さ（レーダ装置１の設置高さに近い）に等しいと仮定する。 In the image of the vehicle C1 illustrated in FIG. 4, the reflection point plot unit 32 assumes that a plurality of radar reflection points P1 to P7 are on the road surface. This is because the heights of the plurality of radar reflection points P <b> 1 to P <b> 7 are unknown around the installation height of the radar apparatus 1. Therefore, it is assumed that the heights of the plurality of radar reflection points P1 to P7 are equal to the height of the road surface (close to the installation height of the radar device 1).
そして、四角錐台検出部３３は、反射点プロット部３２が路面上にあると仮定した複数のレーダ反射点Ｐ１〜Ｐ７に基づいて、画像上で車両Ｃ１を縁取る四角錐台Ｒの底面の辺Ｒ１、Ｒ２を検出する。具体的には、四角錐台検出部３３は、複数のレーダ反射点Ｐ１〜Ｐ３の近似直線を最小二乗法等により算出し、複数のレーダ反射点Ｐ１〜Ｐ３の近似直線を四角錐台Ｒの底面の辺Ｒ１として検出する。そして、四角錐台検出部３３は、複数のレーダ反射点Ｐ４〜Ｐ７の近似直線を最小二乗法等により算出し、複数のレーダ反射点Ｐ４〜Ｐ７の近似直線を四角錐台Ｒの底面の辺Ｒ２として検出する。 Then, the quadrangular frustum detection unit 33 is based on a plurality of radar reflection points P1 to P7 on the assumption that the reflection point plot unit 32 is on the road surface, and the bottom surface of the quadrangular frustum R that borders the vehicle C1 on the image. Edges R1 and R2 are detected. Specifically, the quadrangular frustum detection unit 33 calculates an approximate line of the plurality of radar reflection points P1 to P3 by the least square method or the like, and calculates the approximate line of the plurality of radar reflection points P1 to P3 of the square frustum R. Detected as side R1 on the bottom. Then, the quadrangular frustum detection unit 33 calculates an approximate line of the plurality of radar reflection points P4 to P7 by the least square method or the like, and calculates the approximate line of the plurality of radar reflection points P4 to P7 to the side of the bottom surface of the quadrangular pyramid R. Detect as R2.
このように、図４に示した車両Ｃ１の画像では、複数のレーダ反射点Ｐ１〜Ｐ７の高さを路面の高さ（レーダ装置１の設置高さに近い）に等しいと仮定することにより、車両Ｃ１を縁取る四角錐台Ｒの底面の辺Ｒ１、Ｒ２を容易に検出することができる。なお、図７から図９までと異なり、図４において、車両Ｃ１の下側のエッジ解析を実行しないのは、車両Ｃ１の下側のエッジと路面との間にタイヤの半径程度の高さの空間があり、車両Ｃ１の高さを精度高く計測することができないからである（図１０の第３式を参照）。 Thus, in the image of the vehicle C1 shown in FIG. 4, by assuming that the height of the plurality of radar reflection points P1 to P7 is equal to the height of the road surface (close to the installation height of the radar device 1), The sides R1 and R2 of the bottom surface of the quadrangular frustum R that borders the vehicle C1 can be easily detected. Unlike FIG. 7 to FIG. 9, in FIG. 4, the lower edge analysis of the vehicle C1 is not executed because the height of the tire radius is between the lower edge of the vehicle C1 and the road surface. This is because there is a space and the height of the vehicle C1 cannot be measured with high accuracy (see the third formula in FIG. 10).
図５に示した車両Ｃ１の画像では、四角錐台検出部３３は、レーダ装置１から遠距離のレーダ反射点Ｐ６、Ｐ７に基づかず、レーダ装置１から近距離のレーダ反射点Ｐ４、Ｐ５に基づいて、車両Ｃ１を縁取る四角錐台Ｒの底面の辺Ｒ２を検出する。というのは、複数のレーダ反射点Ｐ４〜Ｐ７の高さは、レーダ装置１の設置高さの周りで不明である。ここで、レーダ装置１から遠距離のレーダ反射点Ｐ６、Ｐ７の高さは、レーダ装置１の設置高さの周りで大きな誤差を有する。一方で、レーダ装置１から近距離のレーダ反射点Ｐ４、Ｐ５の高さは、レーダ装置１の設置高さの周りで大きな誤差を有さない。 In the image of the vehicle C1 shown in FIG. 5, the quadrangular frustum detection unit 33 is not based on the radar reflection points P6 and P7 at a long distance from the radar apparatus 1, but is applied to the radar reflection points P4 and P5 at a short distance from the radar apparatus 1. Based on this, the bottom side R2 of the quadrangular frustum R that borders the vehicle C1 is detected. This is because the heights of the plurality of radar reflection points P4 to P7 are unknown around the installation height of the radar apparatus 1. Here, the heights of the radar reflection points P6 and P7 at a long distance from the radar apparatus 1 have a large error around the installation height of the radar apparatus 1. On the other hand, the heights of the radar reflection points P4 and P5 at a short distance from the radar apparatus 1 do not have a large error around the installation height of the radar apparatus 1.
第１の方法として、四角錐台検出部３３は、複数のレーダ反射点Ｐ６、Ｐ７を利用せず、複数のレーダ反射点Ｐ４、Ｐ５の近似直線を最小二乗法等により算出し、複数のレーダ反射点Ｐ４、Ｐ５の近似直線を四角錐台Ｒの底面の辺Ｒ２として検出する。第２の方法として、四角錐台検出部３３は、複数のレーダ反射点Ｐ６、Ｐ７を複数のレーダ反射点Ｐ４、Ｐ５とほぼ同一の直線上に乗るように補正し、複数のレーダ反射点Ｐ４〜Ｐ７の近似直線を最小二乗法等により算出し、複数のレーダ反射点Ｐ４〜Ｐ７の近似直線を四角錐台Ｒの底面の辺Ｒ２として検出する。第３の方法として、四角錐台検出部３３は、図５に示した方法を、四角錐台Ｒの底面の辺Ｒ２の検出と同様に、四角錐台Ｒの底面の辺Ｒ１の検出で適用する（レーダ装置１から近距離のレーダ反射点Ｐ２、Ｐ３を利用する。）。 As a first method, the quadrangular frustum detection unit 33 does not use the plurality of radar reflection points P6 and P7, calculates approximate lines of the plurality of radar reflection points P4 and P5 by the least square method or the like, An approximate straight line of the reflection points P4 and P5 is detected as the side R2 of the bottom surface of the quadrangular frustum R. As a second method, the quadrangular pyramid detector 33 corrects the plurality of radar reflection points P6 and P7 so that they are on substantially the same straight line as the plurality of radar reflection points P4 and P5, and the plurality of radar reflection points P4. The approximate straight line of ~ P7 is calculated by the least square method or the like, and the approximate straight lines of the plurality of radar reflection points P4 to P7 are detected as the side R2 of the bottom surface of the square frustum R. As a third method, the quadrangular frustum detection unit 33 applies the method illustrated in FIG. 5 in the detection of the side R1 of the bottom surface of the square frustum R, similarly to the detection of the side R2 of the bottom surface of the square frustum R. (Radar reflection points P2 and P3 at a short distance from the radar apparatus 1 are used).
このように、レーダ装置１から遠距離のレーダ反射点Ｐ６、Ｐ７に基づかず、レーダ装置１から近距離のレーダ反射点Ｐ４、Ｐ５に基づいて、車両Ｃ１を縁取る四角錐台Ｒの底面の辺Ｒ２を精度高く検出することができる。そして、レーダ装置１から遠距離のレーダ反射点Ｐ１に基づかず、レーダ装置１から近距離のレーダ反射点Ｐ２、Ｐ３に基づいて、車両Ｃ１を縁取る四角錐台Ｒの底面の辺Ｒ１を精度高く検出することができる。さらに、車両Ｃ１の奥行方向の辺Ｒ２の傾きを精度高く計測することにより、車両Ｃ１の奥行方向の長さを精度高く計測することができる（図１０に示した第１式を参照）。 Thus, based on the radar reflection points P4 and P5 at a short distance from the radar device 1, not based on the radar reflection points P6 and P7 at a long distance from the radar device 1, the bottom surface of the square frustum R that borders the vehicle C1. The side R2 can be detected with high accuracy. Then, based on the radar reflection points P2 and P3 at a short distance from the radar device 1 without being based on the radar reflection point P1 at a long distance from the radar device 1, the side R1 on the bottom surface of the quadrangular frustum R that borders the vehicle C1 is accurately determined. Highly detectable. Furthermore, the length of the vehicle C1 in the depth direction can be measured with high accuracy by measuring the inclination of the side R2 in the depth direction of the vehicle C1 (see the first equation shown in FIG. 10).
本開示の正面と側面との境界の辺検出の処理内容を図６に示す。四角錐台検出部３３は、四角錐台Ｒの底面の二辺Ｒ１、Ｒ２の交点から画像上の高さ方向に延伸する直線に基づいて、四角錐台Ｒの正面と側面との境界の辺Ｒ３を検出する（ステップＳ３）。 The processing content of the edge detection at the boundary between the front and the side according to the present disclosure is shown in FIG. The quadrangular frustum detector 33 is based on a straight line extending in the height direction on the image from the intersection of the two sides R1 and R2 of the bottom surface of the quadrangular frustum R, and the side of the boundary between the front and side surfaces of the quadrangular frustum R R3 is detected (step S3).
本開示の上面の辺検出の処理内容を図７に示す。四角錐台検出部３３は、四角錐台Ｒの底面の辺Ｒ１、Ｒ２から画像上の高さ方向の所定距離でのエッジ解析を実行することにより、四角錐台Ｒの上面の辺Ｒ４、Ｒ５を検出する（ステップＳ３）。 FIG. 7 shows the processing content of the upper side detection according to the present disclosure. The quadrangular frustum detection unit 33 performs edge analysis at a predetermined distance in the height direction on the image from the sides R1 and R2 of the bottom surface of the quadrangular frustum R, so that the sides R4 and R5 on the top surface of the quadrangular frustum R are obtained. Is detected (step S3).
具体的には、四角錐台検出部３３は、レーダ装置１から車両Ｃ１までの距離が近いほど、エッジ解析を実行する画像上の高さ方向の所定距離の範囲ｙＥ［ｐｉｘ］を広くし、レーダ装置１から車両Ｃ１までの距離が遠いほど、エッジ解析を実行する画像上の高さ方向の所定距離の範囲ｙＥ［ｐｉｘ］を狭くする。ここで、辺Ｒ４、Ｒ５を検出するためのエッジ解析範囲Ｅ４、Ｅ５の最小高さは、例えば現行軽自動車の最小高さ及び図１０に示した第３の座標変換式に基づいて設定され、エッジ解析範囲Ｅ４、Ｅ５の最大高さは、法律で定められた最大高さ規制及び図１０に示した第３の座標変換式に基づいて設定される。 Specifically, the quadrangular pyramid detector 33 increases the predetermined range y E [pix] in the height direction on the image on which the edge analysis is performed as the distance from the radar apparatus 1 to the vehicle C1 is shorter. As the distance from the radar apparatus 1 to the vehicle C1 increases, the predetermined range y E [pix] in the height direction on the image on which the edge analysis is performed is narrowed. Here, the minimum height of the edge analysis ranges E4 and E5 for detecting the sides R4 and R5 is set based on, for example, the minimum height of the current light vehicle and the third coordinate conversion formula shown in FIG. The maximum heights of the edge analysis ranges E4 and E5 are set based on the maximum height regulation defined by law and the third coordinate conversion formula shown in FIG.
そして、四角錐台検出部３３は、エッジ解析範囲Ｅ４、Ｅ５のみにおいて、幅方向に長いエッジを検出するＣａｎｎｙ法等を適用することにより、四角錐台Ｒの上面の辺Ｒ４、Ｒ５を容易に検出することができる。なお、四角錐台検出部３３は、エッジ解析範囲Ｅ４、Ｅ５のみにおいて、様々な幅座標位置で高さ方向に輝度勾配を計測することにより、四角錐台Ｒの上面の辺Ｒ４、Ｒ５を検出することもできる。 Then, the quadrangular frustum detector 33 easily applies the Canny method or the like for detecting edges that are long in the width direction only in the edge analysis ranges E4 and E5, so that the sides R4 and R5 on the upper surface of the quadrangular frustum R can be easily obtained. Can be detected. Note that the square pyramid detector 33 detects the edges R4 and R5 on the upper surface of the square pyramid R by measuring the luminance gradient in the height direction at various width coordinate positions only in the edge analysis ranges E4 and E5. You can also
本開示の正面の他辺検出の処理内容を図８に示す。四角錐台検出部３３は、四角錐台Ｒの正面と側面との境界の辺Ｒ３から画像上の幅方向の所定距離でのエッジ解析を実行することにより、四角錐台Ｒの正面の四辺のうち、四角錐台Ｒの正面と側面との境界の辺Ｒ３と向かい合う辺Ｒ６を検出する（ステップＳ３）。 The processing content of the other side detection of the front of this indication is shown in FIG. The quadrangular frustum detection unit 33 performs edge analysis at a predetermined distance in the width direction on the image from the side R3 of the boundary between the front and side surfaces of the quadrangular frustum R, so that the four sides of the front of the quadrangular frustum R are detected. Among these, the side R6 that faces the side R3 at the boundary between the front surface and the side surface of the quadrangular frustum R is detected (step S3).
具体的には、四角錐台検出部３３は、レーダ装置１から車両Ｃ１までの距離が近いほど、エッジ解析を実行する画像上の幅方向の所定距離の範囲ｘＥ［ｐｉｘ］を広くし、レーダ装置１から車両Ｃ１までの距離が遠いほど、エッジ解析を実行する画像上の幅方向の所定距離の範囲ｘＥ［ｐｉｘ］を狭くする。ここで、辺Ｒ６を検出するためのエッジ解析範囲Ｅ６の最小幅座標は、例えば現行軽自動車の最小幅及び図１０に示した第２の座標変換式に基づいて設定され、エッジ解析範囲Ｅ６の最大幅座標は、法律で定められた最大幅規制及び図１０に示した第２の座標変換式に基づいて設定される。 Specifically, the quadrangular pyramid detector 33 increases the range x E [pix] of the predetermined distance in the width direction on the image on which the edge analysis is performed, as the distance from the radar device 1 to the vehicle C1 is shorter, As the distance from the radar apparatus 1 to the vehicle C1 increases, the range x E [pix] of the predetermined distance in the width direction on the image on which the edge analysis is performed is narrowed. Here, the minimum width coordinate of the edge analysis range E6 for detecting the side R6 is set based on, for example, the minimum width of the current light vehicle and the second coordinate conversion formula shown in FIG. The maximum width coordinate is set based on the maximum width restriction defined by law and the second coordinate conversion formula shown in FIG.
そして、四角錐台検出部３３は、エッジ解析範囲Ｅ６のみにおいて、高さ方向に長いエッジを検出するＣａｎｎｙ法等を適用することにより、四角錐台Ｒの正面と側面との境界の辺Ｒ３と向かい合う辺Ｒ６を容易に検出することができる。なお、四角錐台検出部３３は、エッジ解析範囲Ｅ６のみにおいて、様々な高さ座標位置で幅方向に輝度勾配を計測することにより、四角錐台Ｒの正面の他辺Ｒ６を検出することもできる。 Then, the quadrangular frustum detection unit 33 applies the Canny method or the like that detects an edge that is long in the height direction only in the edge analysis range E6, so that the side R3 at the boundary between the front surface and the side surface of the quadrangular frustum R The facing side R6 can be easily detected. Note that the square frustum detection unit 33 may also detect the other side R6 in front of the square frustum R by measuring the luminance gradient in the width direction at various height coordinate positions only in the edge analysis range E6. it can.
本開示の側面の他辺検出の処理内容を図９に示す。四角錐台検出部３３は、車両Ｃ１の最前方のレーダ反射点Ｐ７から画像上の奥行方向の所定距離でのエッジ解析を実行することにより、四角錐台Ｒの側面の四辺のうち、四角錐台Ｒの正面と側面との境界の辺Ｒ３と向かい合う辺Ｒ７を検出する（ステップＳ３）。 The processing content of the other side detection of the side surface of the present disclosure is shown in FIG. The quadrangular frustum detection unit 33 performs the edge analysis at a predetermined distance in the depth direction on the image from the foremost radar reflection point P7 of the vehicle C1, and thus among the four sides of the quadrangular frustum R, the quadrangular pyramid. A side R7 facing the side R3 at the boundary between the front and side surfaces of the table R is detected (step S3).
具体的には、四角錐台検出部３３は、レーダ装置１から車両Ｃ１までの距離が近いほど、エッジ解析を実行する画像上の奥行方向の所定距離の範囲ｚＥ［ｐｉｘ］を広くし、レーダ装置１から車両Ｃ１までの距離が遠いほど、エッジ解析を実行する画像上の奥行方向の所定距離の範囲ｚＥ［ｐｉｘ］を狭くする。ここで、辺Ｒ７を検出するためのエッジ解析範囲Ｅ７の最小奥行及び最大奥行は、辺Ｒ７を確実に検出できるように設定される。 Specifically, the quadrangular pyramid detector 33 widens the predetermined distance range z E [pix] in the depth direction on the image on which the edge analysis is performed, as the distance from the radar device 1 to the vehicle C1 is shorter, As the distance from the radar device 1 to the vehicle C1 increases, the predetermined distance range z E [pix] in the depth direction on the image on which the edge analysis is performed is narrowed. Here, the minimum depth and the maximum depth of the edge analysis range E7 for detecting the side R7 are set so that the side R7 can be reliably detected.
そして、四角錐台検出部３３は、エッジ解析範囲Ｅ７のみにおいて、様々な高さ座標位置で奥行方向に輝度勾配を計測することにより、四角錐台Ｒの正面と側面との境界の辺Ｒ３と向かい合う辺Ｒ７を確実に検出することができる。なお、四角錐台検出部３３は、エッジ解析範囲Ｅ７のみにおいて、高さ方向に長いエッジを検出するＣａｎｎｙ法等を適用することにより、四角錐台Ｒの側面の他辺Ｒ７を検出することもできる。 Then, the quadrangular frustum detector 33 measures the luminance gradient in the depth direction at various height coordinate positions only in the edge analysis range E7, so that the side R3 at the boundary between the front surface and the side surface of the quadrangular frustum R The opposite side R7 can be reliably detected. Note that the quadrangular frustum detection unit 33 may also detect the other side R7 of the side surface of the quadrangular frustum R by applying the Canny method or the like that detects edges that are long in the height direction only in the edge analysis range E7. it can.
本開示の位置変換の処理内容を図１０に示す。位置変換部３４は、画像上の四角錐台Ｒの位置を実空間内の位置に変換する（ステップＳ４）。 FIG. 10 shows the processing contents of the position conversion of the present disclosure. The position conversion unit 34 converts the position of the square frustum R on the image into a position in the real space (step S4).
辺Ｒ１、Ｒ６の交点の座標を、画像上で（ｘ１［ｐｉｘ］、ｙ１［ｐｉｘ］）とし、実空間内で（Ｘ１［ｍ］、Ｙ１［ｍ］、Ｚ１［ｍ］）とする。辺Ｒ２、Ｒ７の交点の座標を、画像上で（ｘ２［ｐｉｘ］、ｙ２［ｐｉｘ］）とし、実空間内で（Ｘ２［ｍ］、Ｙ２［ｍ］、Ｚ２［ｍ］）とする。辺Ｒ１、Ｒ２、Ｒ３の交点の座標を、画像上で（ｘ３［ｐｉｘ］、ｙ３［ｐｉｘ］）とし、実空間内で（Ｘ３［ｍ］、Ｙ３［ｍ］、Ｚ３［ｍ］）とする。辺Ｒ３、Ｒ４、Ｒ５の交点の座標を、画像上で（ｘ４［ｐｉｘ］、ｙ４［ｐｉｘ］）とし、実空間内で（Ｘ４［ｍ］、Ｙ４［ｍ］、Ｚ４［ｍ］）とする。辺Ｒ４、Ｒ６の交点の座標を、画像上で（ｘ５［ｐｉｘ］、ｙ５［ｐｉｘ］）とし、実空間内で（Ｘ５［ｍ］、Ｙ５［ｍ］、Ｚ５［ｍ］）とする。辺Ｒ５、Ｒ７の交点の座標を、画像上で（ｘ６［ｐｉｘ］、ｙ６［ｐｉｘ］）とし、実空間内で（Ｘ６［ｍ］、Ｙ６［ｍ］、Ｚ６［ｍ］）とする。画像の消失点Ｖ（例えば、複数の白線Ｌの延長線の交点）の座標を、画像上で（ｘＶ［ｐｉｘ］、ｙＶ［ｐｉｘ］）とする。 The coordinates of the intersection of the sides R1 and R6 are (x 1 [pix], y 1 [pix]) on the image, and (X 1 [m], Y 1 [m], Z 1 [m] in real space. ). The coordinates of the intersection of the sides R2 and R7 are (x 2 [pix], y 2 [pix]) on the image, and (X 2 [m], Y 2 [m], Z 2 [m] in real space. ). The coordinates of the intersections of the sides R1, R2, and R3 are (x 3 [pix], y 3 [pix]) on the image, and (X 3 [m], Y 3 [m], Z 3 [ m]). The coordinates of the intersections of the sides R3, R4, and R5 are (x 4 [pix], y 4 [pix]) on the image, and (X 4 [m], Y 4 [m], Z 4 [ m]). The coordinates of the intersection of the sides R4 and R6 are (x 5 [pix], y 5 [pix]) on the image, and (X 5 [m], Y 5 [m], Z 5 [m] in real space. ). The coordinates of the intersection of the sides R5 and R7 are (x 6 [pix], y 6 [pix]) on the image, and (X 6 [m], Y 6 [m], Z 6 [m] in real space. ). The coordinates of the vanishing point V of the image (for example, the intersection of the extended lines of a plurality of white lines L) are set to (x V [pix], y V [pix]) on the image.
ｘ１〜ｘ６をｘで代表し、ｙ１〜ｙ６をｙで代表し、Ｘ１〜Ｘ６をＸで代表し、Ｙ１〜Ｙ６をＹで代表し、Ｚ１〜Ｚ６をＺで代表する。画像上での座標（ｘ［ｐｉｘ］、ｙ［ｐｉｘ］）は、実空間内での座標（Ｘ［ｍ］、Ｙ［ｍ］、Ｚ［ｍ］）に、以下の数式により変換される。
Ｚ［ｍ］＝（焦点距離［ｍｍ］×高さ方向の画素数［ｐｉｘ］×センサの設置高さ［ｃｍ］）／（（ｙ［ｐｉｘ］−ｙＶ［ｐｉｘ］）×センササイズ［ｍｍ］×１００）
Ｘ［ｍ］＝（ｘ［ｐｉｘ］−ｘＶ［ｐｉｘ］）×Ｚ［ｍ］での解像度［ｍ／ｐｉｘ］
Ｙ［ｍ］＝（ｙ［ｐｉｘ］−ｙＶ［ｐｉｘ］）×Ｚ［ｍ］での解像度［ｍ／ｐｉｘ］
x 1 to x 6 are represented by x, y 1 to y 6 are represented by y, X 1 to X 6 are represented by X, Y 1 to Y 6 are represented by Y, and Z 1 to Z 6 are represented by Y. Represented by Z. The coordinates (x [pix], y [pix]) on the image are converted into the coordinates (X [m], Y [m], Z [m]) in the real space by the following mathematical formula.
Z [m] = (focal length [mm] × number of pixels in the height direction [pix] × sensor installation height [cm]) / ((y [pix] −y V [pix]) × sensor size [mm ] X 100)
X [m] = (x [ pix] -x V [pix]) × resolution in the Z [m] [m / pix ]
Y [m] = (y [pix] −y V [pix]) × resolution [m / pix] at Z [m]
以上に説明したように、レーダ装置１及び撮像装置２を併用することにより、車両Ｃ１の幅Ｘ［ｍ］、奥行Ｚ［ｍ］及び高さＹ［ｍ］といった三次元情報を計測することができる。そして、自動運転に応用可能な周辺環境認識を行うことができる。 As described above, by using the radar device 1 and the imaging device 2 together, three-dimensional information such as the width X [m], the depth Z [m], and the height Y [m] of the vehicle C1 can be measured. it can. And the surrounding environment recognition applicable to an automatic driving | operation can be performed.
なお、車両Ｃ１の幅Ｘ［ｍ］及び奥行Ｚ［ｍ］といった二次元情報を計測するのみであるレーダに代えて、車両Ｃ１の幅Ｘ［ｍ］、奥行Ｚ［ｍ］及び高さＹ［ｍ］といった三次元情報を計測することができるレーダを用いてもよい。 It should be noted that instead of a radar that only measures two-dimensional information such as the width X [m] and the depth Z [m] of the vehicle C1, the width X [m], the depth Z [m], and the height Y [of the vehicle C1 are measured. A radar capable of measuring three-dimensional information such as m] may be used.
また、車両Ｃ１の幅ｘ［ｐｉｘ］及び高さｙ［ｐｉｘ］といった二次元情報を計測するのみである単眼カメラに代えて、車両Ｃ１の幅ｘ［ｐｉｘ］、高さｙ［ｐｉｘ］及び奥行ｚ［ｐｉｘ］といった三次元情報を計測することができるステレオカメラを用いてもよい。 Further, instead of a monocular camera that only measures two-dimensional information such as the width x [pix] and the height y [pix] of the vehicle C1, the width x [pix], the height y [pix], and the depth of the vehicle C1 A stereo camera that can measure three-dimensional information such as z [pix] may be used.
本開示の車両計測装置及び車両計測プログラムは、ミリ波レーダ等及び単眼カメラ等を併用することにより、車両の三次元情報を計測することができる。 The vehicle measurement device and the vehicle measurement program of the present disclosure can measure three-dimensional information of a vehicle by using a millimeter wave radar or the like and a monocular camera or the like together.
Ｃ：車両計測システム
Ｃ１、Ｃ２、Ｃ３：車両
Ｌ：白線
Ａ：アンテナ
Ｐ１、Ｐ２、Ｐ３、Ｐ４、Ｐ５、Ｐ６、Ｐ７：レーダ反射点
Ｒ：四角錐台
Ｒ１、Ｒ２、Ｒ３、Ｒ４、Ｒ５、Ｒ６、Ｒ７：辺
Ｅ４、Ｅ５、Ｅ６、Ｅ７：エッジ解析範囲
Ｖ：消失点
１：レーダ装置
２：撮像装置
３：車両計測装置
３１：車両情報取得部
３２：反射点プロット部
３３：四角錐台検出部
３４：位置変換部
C: Vehicle measurement systems C1, C2, C3: Vehicle L: White line A: Antennas P1, P2, P3, P4, P5, P6, P7: Radar reflection point R: Square frustum R1, R2, R3, R4, R5, R6, R7: Sides E4, E5, E6, E7: Edge analysis range V: Vanishing point 1: Radar device 2: Imaging device 3: Vehicle measuring device 31: Vehicle information acquisition unit 32: Reflection point plotting unit 33: Square frustum Detection unit 34: position conversion unit
Claims (9)
前記画像上で前記レーダ反射点をプロットする反射点プロット部と、
前記画像上で前記レーダ反射点の近傍でのエッジ解析を実行することにより、前記画像上で前記車両を縁取る四角錐台を検出する四角錐台検出部と、
前記画像上の前記四角錐台の位置を実空間内の位置に変換する位置変換部と、
を備えることを特徴とする車両計測装置。 A vehicle information acquisition unit for acquiring information of a radar reflection point from the vehicle and an image of the vehicle;
A reflection point plotting part for plotting the radar reflection point on the image;
A quadrangular frustum detector for detecting a quadrangular frustum that borders the vehicle on the image by performing an edge analysis in the vicinity of the radar reflection point on the image;
A position converter that converts the position of the truncated pyramid on the image into a position in real space;
A vehicle measurement device comprising:
前記四角錐台検出部は、前記反射点プロット部が前記路面上にあると仮定した前記レーダ反射点に基づいて、前記四角錐台の底面の辺を検出する
ことを特徴とする、請求項１に記載の車両計測装置。 The reflection point plot unit assumes that the radar reflection point is on the road surface,
2. The quadrangular frustum detection unit detects a side of a bottom surface of the quadrangular frustum based on the radar reflection point on the assumption that the reflection point plot unit is on the road surface. The vehicle measurement device described in 1.
ことを特徴とする、請求項２に記載の車両計測装置。 The quadrangular frustum detection unit detects a side of the bottom surface of the quadrangular frustum based on the radar reflection point at a short distance from the radar device, not based on the radar reflection point at a long distance from the radar device. The vehicle measurement device according to claim 2, wherein the vehicle measurement device is characterized.
ことを特徴とする、請求項２又は３に記載の車両計測装置。 The quadrangular frustum detection unit detects an upper side of the square frustum by performing an edge analysis at a predetermined distance in the height direction on the image from a side of the bottom surface of the square frustum. The vehicle measurement device according to claim 2, wherein the vehicle measurement device is characterized.
ことを特徴とする、請求項２から４のいずれかに記載の車両計測装置。 The quadrangular frustum detection unit detects a side of the boundary between the front and side surfaces of the square frustum based on a straight line extending in the height direction on the image from an intersection of two sides of the bottom surface of the square frustum. The vehicle measuring device according to any one of claims 2 to 4, wherein the vehicle measuring device is characterized in that:
ことを特徴とする、請求項５に記載の車両計測装置。 The square frustum detection unit performs edge analysis at a predetermined distance in the width direction on the image from a side of a boundary between a front surface and a side surface of the quadrangular frustum, so that the four sides of the front surface of the square frustum The vehicle measuring device according to claim 5, wherein a side facing a side of a boundary between a front surface and a side surface of the quadrangular frustum is detected.
ことを特徴とする、請求項５又は６に記載の車両計測装置。 The quadrangular frustum detection unit performs edge analysis at a predetermined distance in the depth direction on the image from the radar reflection point at the forefront of the vehicle, and among the four sides of the side surface of the quadrangular frustum, The vehicle measuring device according to claim 5 or 6, wherein a side facing a side of a boundary between a front surface and a side surface of the quadrangular pyramid is detected.
ことを特徴とする、請求項４、６又は７に記載の車両計測装置。 As the distance from the radar device to the vehicle decreases, the square pyramid detector detects a larger range of the predetermined distance on the image for performing edge analysis, and as the distance from the radar device to the vehicle increases. The vehicle measurement device according to claim 4, wherein the range of the predetermined distance on the image for performing edge analysis is narrowed.
前記画像上で前記レーダ反射点をプロットする反射点プロットステップと、
前記画像上で前記レーダ反射点の近傍でのエッジ解析を実行することにより、前記画像上で前記車両を縁取る四角錐台を検出する四角錐台検出ステップと、
前記画像上の前記四角錐台の位置を実空間内の位置に変換する位置変換ステップと、
を順にコンピュータに実行させるための車両計測プログラム。 Vehicle information acquisition step for acquiring radar reflection information from the vehicle and an image of the vehicle;
A reflection point plotting step for plotting the radar reflection point on the image;
A quadrangular frustum detection step for detecting a quadrangular frustum that borders the vehicle on the image by performing an edge analysis in the vicinity of the radar reflection point on the image;
A position converting step of converting the position of the square frustum on the image into a position in real space;
Is a vehicle measurement program for causing a computer to execute in order.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018109792A JP7064389B2 (en) | 2018-06-07 | 2018-06-07 | Vehicle measuring device and vehicle measuring program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018109792A JP7064389B2 (en) | 2018-06-07 | 2018-06-07 | Vehicle measuring device and vehicle measuring program |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP2019211402A true JP2019211402A (en) | 2019-12-12 |
| JP7064389B2 JP7064389B2 (en) | 2022-05-10 |
Family
ID=68846655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2018109792A Active JP7064389B2 (en) | 2018-06-07 | 2018-06-07 | Vehicle measuring device and vehicle measuring program |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP7064389B2 (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009223504A (en) * | 2008-03-14 | 2009-10-01 | Omron Corp | Vehicle detection device, method, and program |
| JP2010256040A (en) * | 2009-04-21 | 2010-11-11 | Toyota Motor Corp | Vehicle detecting device |
| JP2011204054A (en) * | 2010-03-26 | 2011-10-13 | Mazda Motor Corp | Drive support device for vehicle |
| JP2015185045A (en) * | 2014-03-25 | 2015-10-22 | 本田技研工業株式会社 | Object detection processing apparatus |
-
2018
- 2018-06-07 JP JP2018109792A patent/JP7064389B2/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009223504A (en) * | 2008-03-14 | 2009-10-01 | Omron Corp | Vehicle detection device, method, and program |
| JP2010256040A (en) * | 2009-04-21 | 2010-11-11 | Toyota Motor Corp | Vehicle detecting device |
| JP2011204054A (en) * | 2010-03-26 | 2011-10-13 | Mazda Motor Corp | Drive support device for vehicle |
| JP2015185045A (en) * | 2014-03-25 | 2015-10-22 | 本田技研工業株式会社 | Object detection processing apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| JP7064389B2 (en) | 2022-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107272021B (en) | Object detection using radar and visually defined image detection areas | |
| US9470548B2 (en) | Device, system and method for calibration of camera and laser sensor | |
| CN102016921B (en) | Image processing device | |
| US11762071B2 (en) | Multi-mode multi-sensor calibration | |
| CN110782465B (en) | Ground segmentation method and device based on laser radar and storage medium | |
| EP1909064A1 (en) | Object detection device | |
| CN109410264B (en) | Front vehicle distance measuring method based on laser point cloud and image fusion | |
| JP5316572B2 (en) | Object recognition device | |
| JP2006151125A (en) | On-vehicle image processing device | |
| CN105699985A (en) | Single-line laser radar device | |
| KR101565900B1 (en) | Device, method for calibration of camera and laser range finder | |
| US11062153B2 (en) | Apparatus and method for converting image | |
| US11030761B2 (en) | Information processing device, imaging device, apparatus control system, movable body, information processing method, and computer program product | |
| JP6552448B2 (en) | Vehicle position detection device, vehicle position detection method, and computer program for vehicle position detection | |
| JP4052291B2 (en) | Image processing apparatus for vehicle | |
| JP2016200557A (en) | Calibration device, distance measurement apparatus and calibration method | |
| JP2019211403A (en) | Object position measurement device and object position measurement program | |
| US8031908B2 (en) | Object recognizing apparatus including profile shape determining section | |
| KR20160125803A (en) | Apparatus for defining an area in interest, apparatus for detecting object in an area in interest and method for defining an area in interest | |
| JP7064389B2 (en) | Vehicle measuring device and vehicle measuring program | |
| JP3690260B2 (en) | Vehicle distance measurement method | |
| JP7043787B2 (en) | Object detection system | |
| KR102568111B1 (en) | Apparatus and method for detecting road edge | |
| WO2018100971A1 (en) | Information processing device, imaging device, apparatus control system, movable body, information processing method, and computer program product | |
| JP4571204B2 (en) | Object recognition device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20210601 |
|
| A625 | Written request for application examination (by other person) |
Free format text: JAPANESE INTERMEDIATE CODE: A625 Effective date: 20210601 |
|
| TRDD | Decision of grant or rejection written | ||
| A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20220413 |
|
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20220419 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20220422 |
|
| R150 | Certificate of patent or registration of utility model |
Ref document number: 7064389 Country of ref document: JP Free format text: JAPANESE INTERMEDIATE CODE: R150 |