A robust method of feature extraction from noised endoscopic images(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))

スポンサーリンク

概要

• 論文の詳細を見る

• One of the main problems in multi-view geometry is determining an effective method of feature extraction to locate corresponding points between successive frames. Finding the corresponding points is important in order to satisfy the epipolar geometry between the successive frames. Although the ROBPCA-SIFT method can be used in noised endoscopic images, it cannot be applied to endoscopic images with excessive noise because the method introduces unacceptable error which exceeds the criterion. An ICA-based denoising method such as preprocessing should therefore be applied. Experimental results show that the ROBPCA-SIFT method with the ICA-based denoising algorithm applied as a preprocessing step is reliable and has good performance for feature extraction from endoscopic images with excessive noise.

• 社団法人電子情報通信学会の論文
• 2009-01-12

著者

• Sun Ming

  Department Of Electronics & Information Engineering Korea University

• Sun Ming

  Department Of Electrical Engineering National Taiwan University

• Kim Mingi

  Department Of Electronics & Information Engineering Korea University

• Oh Jangseok

  Department Of Electronics & Information Engineering Korea University

• Choi Seokyoon

  Department Of Electronics & Information Engineering Korea University

• KIM Hochul

  Biomedical Engineering, Biomedical Science of Brain Korea 21, Korea University

• CHOI Kwanghee

  Department of Electronics & Information Engineering, Korea University

• HA Seunghan

  Research Institute for Skin Image, Korea University College of Medicine

• LEE Onseok

  Research Institute for Skin Image, Korea University College of Medicine

• Kim Hochul

  Biomedical Engineering Biomedical Science Of Brain Korea 21 Korea University:korea Artificial Organ

• Ha Seunghan

  Research Institute For Skin Image Korea University College Of Medicine

• Choi Kwanghee

  Department Of Electronics & Information Engineering Korea University

• Lee Onseok

  Research Institute For Skin Image Korea University College Of Medicine

関連論文

• A robust Human Face Recognition algorithm for flexible situations using SIFT(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))
• L1 minimization based EM algorithm for PET Reconstruction(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))
• A One-Step Input Matching Method for Cascode CMOS Low-Noise Amplifiers(Electronic Circuits)
• Improvement of snake model for ventricle segmentation in MRI(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))
• A robust method of feature extraction from noised endoscopic images(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))

スポンサーリンク