METHOD FOR ACCURATELY DETERMINING POSITION AND ORIENTATION OF EACH OF PLURAL IDENTICAL RECOGNITION TARGET OBJECTS IN SEARCH TARGET IMAGE

PROBLEM TO BE SOLVED: To detect, even when a plurality of recognition target objects are present in a search target image, the number, positions and inclinations of the objects.SOLUTION: Dictionary image data including a recognition target pattern, includes a size (Rm) and a direction ( m) of a feat...

Full description

Saved in:
Bibliographic Details
Main Authors NAKANO HIROTAKE, OISHI CHIAKI, MORI YUMI
Format Patent
LanguageEnglish
Japanese
Published 10.03.2016
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:PROBLEM TO BE SOLVED: To detect, even when a plurality of recognition target objects are present in a search target image, the number, positions and inclinations of the objects.SOLUTION: Dictionary image data including a recognition target pattern, includes a size (Rm) and a direction ( m) of a feature vector for a plurality of feature points on a pattern and also includes an offset (Ox, Oy) from the coordinates of a center of an image to the coordinates of the feature point. Sizes (Rt) and directions ( t) of feature vectors are provided for coordinates (Tx, Ty) of a plurality of feature points on a target image. Coordinates (Fx, Fy) of a virtual center point derived from Tx, Ty, Ox, Oy, Rm, Rt, m, and t are calculated. The number of other virtual center points within a predetermined radius (r) with the coordinates (Fx, Fy) as a center is counted. When the number of counted virtual center points is equal to or greater than a predetermined threshold (Nc), the virtual center points are assumed to be candidates which might include recognition patterns.SELECTED DRAWING: Figure 8 【課題】検索対象画像中に複数の認識対象物体が存在する場合であっても、物体の個数、位置、傾きを検出する。【解決手段】認識対象パターンを含む辞書画像データであって、パターン上の複数の特徴点について、特徴ベクトルの大きさ(Rm)および方向(θm)を含んでおり、かつ、画像中心の座標からその特徴点の座標へのオフセット(Ox、Oy)を含んでいる。対象画像上の複数の特徴点の座標(Tx、Ty)について、特徴ベクトルの大きさ(Rt)および方向(θt)を提供する。Tx、Ty、Ox、Oy、Rm、Rt、θm、θtから導かれるところの、仮想中心点の座標(Fx、Fy)を計算する。Fx、Fyを中心とした所定の半径(r)内に含まれる、他の仮想中心点の数をカウントする。カウントされた仮想中心点の数が所定の閾値(Nc)以上のとき、認証パターンが含まれるであろう候補とする。【選択図】図8
Bibliography:Application Number: JP20140156930