ROAD SURFACE DETECTION DEVICE

• Main Authors: MITA SEIICHI, HOSSEIN TEHRANI NIKNEJAD, ISHIMARU KAZUHISA, LONG QIAN, SHIRAI TAKAMASA
• Format: Patent
• Language: English; Japanese
• Published: 21.12.2017
• Subjects: CALCULATING; COMPUTING; COUNTING; GYROSCOPIC INSTRUMENTS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; MEASURING; MEASURING DISTANCES, LEVELS OR BEARINGS; NAVIGATION; PHOTOGRAMMETRY OR VIDEOGRAMMETRY; PHYSICS; SURVEYING; TESTING

PROBLEM TO BE SOLVED: To make it possible to more accurately detect a position of a road surface in a road surface detection device detecting the position of the road surface.SOLUTION: A road surface detection device is configured to: divide a parallax image into a plurality of sub-parallax images according to a vertical coordinate value representing a coordinate value in a vertical direction in the parallax image; use an integrated value integrating a plurality of pixels included in the divided zone for each parallax size as an integration value, and acquire parallax information in which an element having the parallax size associated with a vertical coordinate value and the integration value is collected for each parallax size and vertical coordinate value (S140); set the number of retrieval of the vertical coordinate value for each vertical coordinate value, and in this instance, set the number of retrieval of the vertical coordinate value to a smaller value as the vertical coordinate value indicates that a distance becomes greater (S150); set a range of the vertical coordinate value according to the number of retrieval for each parallax size; create a cost map in which a cost is determined by the parallax size and the vertical coordinate value for the element within the range of the vertical coordinate value (S210); and extract a road surface parallax on the basis of the cost map (S220 and S230).SELECTED DRAWING: Figure 2 【課題】路面の位置を検出する路面検出装置において、路面の位置をより正確に推定できるようにする。【解決手段】路面検出装置は、視差画像において、縦方向の座標値を表す縦座標値に従って視差画像を複数に分割し、該分割した領域内に含まれる複数の画素を視差の大きさ毎に積算した値を積算値として、視差の大きさと縦座標値と積算値とを対応付けた要素を、視差の大きさおよび縦座標値毎に集めた視差情報を取得し（Ｓ１４０）、縦座標値毎に縦座標値の探索数を設定し、この際、縦座標値が、距離が大きくなることを示すにつれて、縦座標値の探索数を小さな値に設定する（Ｓ１５０）。そして、視差の大きさ毎に縦座標値の範囲を探索数に従って設定し、縦座標値の範囲内の要素について、視差の大きさおよび縦座標値によってコストが特定されるコストマップを生成し（Ｓ２１０）、コストマップに基づいて路面視差を抽出する（Ｓ２２０、Ｓ２３０）。【選択図】図２