JP2001109885A5 - - Google Patents

Download PDF

Info

Publication number
JP2001109885A5
JP2001109885A5 JP1999284954A JP28495499A JP2001109885A5 JP 2001109885 A5 JP2001109885 A5 JP 2001109885A5 JP 1999284954 A JP1999284954 A JP 1999284954A JP 28495499 A JP28495499 A JP 28495499A JP 2001109885 A5 JP2001109885 A5 JP 2001109885A5
Authority
JP
Japan
Prior art keywords
matching
image
image processing
processing apparatus
feature points
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
JP1999284954A
Other languages
Japanese (ja)
Other versions
JP2001109885A (en
Filing date
Publication date
Application filed filed Critical
Priority to JP28495499A priority Critical patent/JP2001109885A/en
Priority claimed from JP28495499A external-priority patent/JP2001109885A/en
Publication of JP2001109885A publication Critical patent/JP2001109885A/en
Publication of JP2001109885A5 publication Critical patent/JP2001109885A5/ja
Abandoned legal-status Critical Current

Links

Description

【特許請求の範囲】
【請求項1】 2以上の画像のそれぞれについての特徴点を抽出する特徴抽出工程と、
上記2以上の画像のうち、一の画像と他の画像との特徴点を比較してマッチングを行うマッチング工程と、
上記マッチング工程の結果に基づいて、上記一の画像に含まれるオブジェクトを、上記他の画像に含まれるオブジェクトに対して同定する同定工程とを備えること
を特徴とする画像処理方法。
【請求項2】2以上の画像のそれぞれについての特徴点を抽出する特徴抽出手段と、
上記2以上の画像のうち、一の画像と他の画像との特徴点を比較してマッチングを行うマッチング手段と、
上記マッチング手段によるマッチングの結果に基づいて、上記一の画像に含まれるオブジェクトを、上記他の画像に含まれるオブジェクトに対して同定する同定手段とを備えること
を特徴とする画像処理装置。
【請求項3】 上記マッチング手段は、ハウスドルフマッチングを行うこと
を特徴とする請求項2記載の画像処理装置。
【請求項4】 上記マッチング手段は、ボトルネックマッチングを行うこと
を特徴とする請求項2記載の画像処理装置。
【請求項5】 上記マッチング手段は、上記一の画像についての全ての特徴点からなる特徴点群を示す第1の集合と、上記他の画像についての全ての特徴点からなる特徴点群を示す第2の集合との合同を検出すること
を特徴とする請求項2記載の画像処理装置。
【請求項6】 上記マッチング手段は、上記第1の集合及び上記第2の集合のそれぞれの全ての最近接点対を求め、これらの最近接点対が一致する距離を有するか否かを判別し、一致すると判別された上記第1の集合及び上記第2の集合のそれぞれの最近接点対が同じ基本性質を有するか否かを判別し、上記第1の集合及び上記第2の集合のそれぞれを正規化し、上記第1の集合のセントロイドと、上記第2の集合のセントロイドとを求めて重複させ、上記第1の集合を構成する全ての特徴点群と上記第1の集合のセントロイドとの距離と、上記第2の集合を構成する全ての特徴点群と上記第2の集合のセントロイドとの距離とを求め、上記第1の集合及び上記第2の集合のそれぞれの半径関数を計算し、上記第1の集合の半径関数の濃度が最小となる距離における上記第1の集合及び上記第2の集合のそれぞれの半径関数である第3の集合及び第4の集合を求め、上記第3の集合及び上記第4の集合の幾何学的グラフを計算し、これらの幾何学的グラフが同一の数の反復距離を有しているか否かを判別し、上記第3の集合及び上記第4の集合の幾何学的グラフが同一の数の反復距離を有していると判別された場合に、上記第3の集合の凸閉包の1つの面である稜線を算出し、算出された稜線を、上記第4の集合の凸閉包のいずれかの稜線に対してマッチさせ、このマッチさせた稜線対に基づいて与えられる全ての変換をテストすることによって、上記第1の集合と上記第2の集合との合同を検出すること
を特徴とする請求項5記載の画像処理装置。
【請求項7】 マッチさせた稜線対は、剛体運動を定義すること
を特徴とする請求項6記載の画像処理装置。
【請求項8】 マッチさせた稜線対に基づいて与えられる変換は、アフィン変換であること
を特徴とする請求項6記載の画像処理装置。
[Claims]
1. A feature extraction step of extracting feature points for each of two or more images, and
A matching step of comparing and matching the feature points of one image and another of the above two or more images, and
An image processing method comprising an identification step of identifying an object included in the one image with respect to an object included in the other image based on the result of the matching step.
2. A feature extraction means for extracting feature points for each of two or more images, and a feature extraction means.
Of the above two or more images, a matching means for comparing and matching the feature points of one image and the other image, and
An image processing apparatus comprising: an identification means for identifying an object included in the one image with respect to an object included in the other image based on the result of matching by the matching means.
3. The image processing apparatus according to claim 2, wherein the matching means performs Hausdorff matching.
4. The image processing apparatus according to claim 2, wherein the matching means performs bottleneck matching.
5. The matching means shows a first set showing a feature point group consisting of all the feature points of the one image, and a feature point group consisting of all the feature points of the other image. The image processing apparatus according to claim 2, wherein the congruence with the second set is detected.
6. The matching means obtains all the recent contact pairs of each of the first set and the second set, determines whether or not these recent contact pairs have a matching distance, and determines whether or not these recent contact pairs have a matching distance. It is determined whether or not the recent contact pairs of the first set and the second set that are determined to match have the same basic properties, and each of the first set and the second set is normal. The centroid of the first set and the centroid of the second set are obtained and overlapped, and all the feature point groups constituting the first set and the centroid of the first set are combined with each other. And the distance between all the feature point groups constituting the second set and the centroid of the second set, and the radius functions of the first set and the second set, respectively. The calculation is performed to obtain the third set and the fourth set, which are the radius functions of the first set and the second set, respectively, at the distance where the concentration of the radius function of the first set is minimized. The geometric graphs of the third set and the fourth set are calculated to determine whether these geometric graphs have the same number of repetition distances, and the third set and the fourth set are described. When it was determined that the geometric graph of the fourth set had the same number of repetition distances, the ridge line, which is one surface of the convex closure of the third set, was calculated and calculated. The first set and the first set by matching the ridges to any of the convex closures of the fourth set and testing all transformations given based on this matched ridge pair. The image processing apparatus according to claim 5, wherein the congruence with the set of 2 is detected.
7. The image processing apparatus according to claim 6, wherein the matched ridge line pair defines a rigid body motion.
8. The image processing apparatus according to claim 6, wherein the transformation given based on the matched ridge pair is an affine transformation.

JP28495499A 1999-10-05 1999-10-05 Method and device for image processing Abandoned JP2001109885A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP28495499A JP2001109885A (en) 1999-10-05 1999-10-05 Method and device for image processing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP28495499A JP2001109885A (en) 1999-10-05 1999-10-05 Method and device for image processing

Publications (2)

Publication Number Publication Date
JP2001109885A JP2001109885A (en) 2001-04-20
JP2001109885A5 true JP2001109885A5 (en) 2006-05-25

Family

ID=17685238

Family Applications (1)

Application Number Title Priority Date Filing Date
JP28495499A Abandoned JP2001109885A (en) 1999-10-05 1999-10-05 Method and device for image processing

Country Status (1)

Country Link
JP (1) JP2001109885A (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3982817B2 (en) 2003-03-07 2007-09-26 株式会社東芝 Image processing apparatus and image processing method
US8155451B2 (en) 2004-11-12 2012-04-10 Kitakyushu Foundation For The Advancement Of Industry, Science And Technology Matching apparatus, image search system, and histogram approximate restoring unit, and matching method, image search method, and histogram approximate restoring method
JP4525526B2 (en) * 2005-08-26 2010-08-18 パナソニック電工株式会社 Pattern matching method and apparatus
US7983482B2 (en) 2005-11-08 2011-07-19 Kitakyushu Foundation For The Advancement Of Industry, Science And Technology Matching apparatus, image search system, and histogram approximate restoring unit, and matching method, image search method, and histogram approximate restoring method
CN117990204B (en) * 2024-04-03 2024-06-25 深圳市美格信测控技术有限公司 Motor noise evaluation method and system under multiple working conditions of electric automobile

Similar Documents

Publication Publication Date Title
Kumar et al. Personal authentication using hand images
KR20170081176A (en) Fingerprint authentication using stitch and cut
KR910010353A (en) Fingerprint Identification Method
JP2000215317A5 (en)
Patil et al. Yoga tutor visualization and analysis using SURF algorithm
CN105550630B (en) A kind of fingerprint authentication method and device
Vretos et al. 3D facial expression recognition using Zernike moments on depth images
JP6900664B2 (en) Image processing equipment, image processing method and image processing program
JP2005222419A (en) Biometrics information collation system
Julasayvake et al. An algorithm for fingerprint core point detection
Ouyang et al. Fingerprint pose estimation based on faster R-CNN
Trung et al. Blind inpainting forgery detection
Tukur Fingerprint recognition and matching using Matlab
CN107563292A (en) A kind of face emotion identification method based on LDDMM Curve Matchings
JP2001109885A5 (en)
Gao et al. Feature-level fusion in personal identification
Farajzadeh et al. Vehicle logo recognition using image matching and textural features
KR20050094228A (en) Fingerprint recognition method
CN107516088B (en) Multi-knuckle grain identification method
JP2734245B2 (en) Image feature extraction method and apparatus
JP2002117405A (en) Palm shape authentication method
JP3859347B2 (en) Object recognition apparatus and method
Khanyile et al. Similarity score computation for minutiae-based fingerprint recognition
EP1394726A3 (en) Image processing for pattern recognition using edge detection and projections along predetermined directions
Mehrotra et al. An efficient dual stage approach for iris feature extraction using interest point pairing