IMAGE PROCESSING DEVICE, IMAGE PROCESSING METHOD, AND IMAGE PROCESSING PROGRAM

To specify a shape error of an object using the shape information of the object and the feature line detected from an image of the object.SOLUTION: A storage unit 211 stores shape information 221 that includes a plurality of line segments representing the shape of an object. A detection unit 212 det...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: MOGI ATSUNORI, ISHIKAWA MASARU, YOSHITAKE TOSHIYUKI
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To specify a shape error of an object using the shape information of the object and the feature line detected from an image of the object.SOLUTION: A storage unit 211 stores shape information 221 that includes a plurality of line segments representing the shape of an object. A detection unit 212 detects a plurality of feature lines from an image of the object, and a projection unit 213 projects some of the plurality of line segments included in the shape information 221 to the object image and thereby generates a target projection line. A selection unit 214 extracts a feature line that overlaps a prescribed region based on the target projection line in the object image from among the plurality of feature lines, generates a set of feature lines and selects a target feature line from the set of feature lines on the basis of gradient information that indicates the gradient of pixel values in the object image. A calculation unit 215 calculates the amount of shift between the line segment that corresponds to the target projection line and the target feature line.SELECTED DRAWING: Figure 16 【課題】物体の形状情報と、その物体の画像から検出された特徴線とを用いて、物体の形状の誤差を特定する。【解決手段】記憶部211は、物体の形状を表す複数の線分を含む形状情報221を記憶する。検出部212は、物体の画像から複数の特徴線を検出し、投影部213は、形状情報221に含まれる複数の線分のうち、いずれかの線分を物体の画像上に投影することで、対象投影線を生成する。選択部214は、物体の画像上で対象投影線に基づく所定領域と重なる特徴線を、複数の特徴線の中から抽出して、特徴線の集合を生成し、物体の画像上における画素値の勾配を示す勾配情報に基づいて、特徴線の集合から対象特徴線を選択する。計算部215は、対象投影線に対応する線分と対象特徴線との間のずれ量を計算する。【選択図】図16