[工学]图像特征提取.ppt

3/31/2005 CSE 576: Computer Vision * Descriptors Invariant to Scale Use the scale determined by detector to compute descriptor in a normalized frame For example: moments integrated over an adapted window derivatives adapted to scale: sIx 3/31/2005 CSE 576: Computer Vision * Affine Invariant Descriptors Affine invariant color moments F.Mindru et.al. “Recognizing Color Patterns Irrespective of Viewpoint and Illumination”. CVPR99 Different combinations of these moments are fully affine invariant Also invariant to affine transformation of intensity I ? a I + b 3/31/2005 CSE 576: Computer Vision * Affine Invariant Descriptors Find affine normalized frame J.Matas et.al. “Rotational Invariants for Wide-baseline Stereo”. Research Report of CMP, 2003 A A1 A2 rotation Compute rotational invariant descriptor in this normalized frame 3/31/2005 CSE 576: Computer Vision * SIFT – Scale Invariant Feature Transform Descriptor overview: Determine scale (by maximizing DoG in scale and in space), local orientation as the dominant gradient direction.Use this scale and orientation to make all further computations invariant to scale and rotation. Compute gradient orientation histograms of several small windows (128 values for each point) Normalize the descriptor to make it invariant to intensity change D.Lowe. “Distinctive Image Features from Scale-Invariant Keypoints”. IJCV 2004 3/31/2005 CSE 576: Computer Vision * Select canonical orientation Create histogram of local gradient directions computed at selected scale Assign canonical orientation at peak of smoothed histogram Each key specifies stable 2D coordinates (x, y, scale, orientation) 3/31/2005 CSE 576: Computer Vision * Example of keypoint detection Threshold on value at DOG peak and on ratio of principle curvatures (Harris approach) (a) 233x189 image (b) 832 DOG extrema (c) 729 left after peak value threshold (d) 536 left after testing ratio of principle curvatures 3/31/2005 CSE 576: Computer Vision * SIFT