JP2001109886A5 - - Google Patents

Download PDF

Info

Publication number
JP2001109886A5
JP2001109886A5 JP1999284956A JP28495699A JP2001109886A5 JP 2001109886 A5 JP2001109886 A5 JP 2001109886A5 JP 1999284956 A JP1999284956 A JP 1999284956A JP 28495699 A JP28495699 A JP 28495699A JP 2001109886 A5 JP2001109886 A5 JP 2001109886A5
Authority
JP
Japan
Prior art keywords
radius
matching
image processing
processing apparatus
feature point
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
JP1999284956A
Other languages
Japanese (ja)
Other versions
JP2001109886A (en
Filing date
Publication date
Application filed filed Critical
Priority to JP28495699A priority Critical patent/JP2001109886A/en
Priority claimed from JP28495699A external-priority patent/JP2001109886A/en
Publication of JP2001109886A publication Critical patent/JP2001109886A/en
Publication of JP2001109886A5 publication Critical patent/JP2001109886A5/ja
Abandoned legal-status Critical Current

Links

Description

【特許請求の範囲】
【請求項1】 画像の特徴点を抽出する特徴抽出工程と、
上記画像の特徴点を用い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、上記画像を圧縮する圧縮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】 画像の特徴点を抽出する特徴抽出手段と、
上記画像の特徴点を用い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、上記画像を圧縮する圧縮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、上記画像についての全ての特徴点からなる特徴点群を示す集合の対称性を検出すること
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、2次元の点集合である上記集合のセントロイドを算出し、上記集合を構成する全ての特徴点群と上記セントロイドとの距離を求め、上記集合の半径関数を計算し、計算された上記半径関数により定義される任意の半径上に位置する特徴点群の濃度が1以上であるか否かを判別し、上記半径上に位置する特徴点群の濃度が1以上であると判別された場合に、上記半径上に位置する任意の特徴点を選択し、選択された特徴点における軸が対称であるか否かを判別し、上記軸が対称であると判別された場合には、上記集合が上記軸を対称とする恒等写像となる変換を報告し、上記軸が対称でないと判別された場合には、上記集合が恒等写像であることを報告することによって、上記集合の対称性を検出すること
を特徴とする請求項5記載の画像処理装置。
【請求項7】 上記マッチング手段は、上記半径上に位置する特徴点群の濃度が1以上でないと判別された場合に、上記集合の特徴点の凸閉包を計算し、上記凸閉包を用いて全ての変換を列挙し、列挙された全ての変換集合の論理積を計算し、計算された上記論理積を報告すること
を特徴とする請求項6記載の画像処理装置。
【請求項8】 上記マッチング手段は、次数dが3以上の任意次元の点集合である上記集合のセントロイドを算出し、上記集合を構成する全ての特徴点群と上記セントロイドとの距離を求め、上記集合の半径関数を計算し、計算された上記半径関数により定義される任意の半径上に位置する特徴点群の濃度がd−1以上であるか否かを判別し、上記半径上に位置する特徴点群の濃度がd−1以上であると判別された場合に、少なくともd−1個且つ多くとも2d−3個の点を有する半径の最小値を選択することによって、上記集合の対称性を検出すること
を特徴とする請求項5記載の画像処理装置。
【請求項9】 上記マッチング手段は、上記半径上に位置する特徴点群の濃度がd−1以上でないと判別された場合に、上記集合の特徴点の凸閉包を計算し、上記凸閉包を用いて全ての変換を列挙し、列挙された全ての変換集合の論理積を計算すること
を特徴とする請求項8記載の画像処理装置。
[Claims]
1. A feature extraction step for extracting feature points of an image, and a feature extraction step.
A matching process that uses the feature points of the above image for matching,
An image processing method comprising a compression step of compressing the image based on the result of the matching step.
2. A feature extraction means for extracting feature points of an image,
Matching means for matching using the feature points of the above image and
An image processing apparatus including a compression means for compressing the 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 image processing apparatus according to claim 2, wherein the matching means detects the symmetry of a set indicating a feature point cloud consisting of all the feature points of the image.
6. The matching means calculates the centroid of the set, which is a two-dimensional set of points, obtains the distances between all the feature point groups constituting the set and the centroid, and the radius of the set. The function is calculated, it is determined whether or not the density of the feature point group located on an arbitrary radius defined by the calculated radius function is 1 or more, and the density of the feature point group located on the radius is determined. When is determined to be 1 or more, an arbitrary feature point located on the radius is selected, it is determined whether or not the axis at the selected feature point is symmetric, and the axis is symmetric. If it is determined that the axis is not symmetric, the transformation that makes the set symmetric is reported, and if it is determined that the axis is not symmetric, the set is an equal mapping. The image processing apparatus according to claim 5, wherein the symmetry of the set is detected by reporting.
7. The matching means calculates the convex closure of the feature points of the set when it is determined that the concentration of the feature point group located on the radius is not 1 or more, and uses the convex closure. The image processing apparatus according to claim 6, wherein all the transformations are listed, the logical product of all the listed transformation sets is calculated, and the calculated logical product is reported.
8. The matching means calculates a centroid of the set which is a point set of an arbitrary dimension having an order d of 3 or more, and determines the distance between all the feature point groups constituting the set and the centroid. The radius function of the set is calculated, and it is determined whether or not the density of the feature point group located on an arbitrary radius defined by the calculated radius function is d-1 or more, and the radius is on the radius. When it is determined that the concentration of the feature point group located in is d-1 or more, the above set is selected by selecting the minimum value of the radius having at least d-1 points and at most 2d-3 points. The image processing apparatus according to claim 5, wherein the symmetry of the image processing apparatus is detected.
9. The matching means calculates the convex closure of the feature points of the set when it is determined that the concentration of the feature point group located on the radius is not d-1 or more, and performs the convex closure. The image processing apparatus according to claim 8, wherein all the transformations are enumerated by using the method, and the logical product of all the enumerated transformation sets is calculated.

JP28495699A 1999-10-05 1999-10-05 Method and device for image processing Abandoned JP2001109886A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP28495699A JP2001109886A (en) 1999-10-05 1999-10-05 Method and device for image processing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP28495699A JP2001109886A (en) 1999-10-05 1999-10-05 Method and device for image processing

Publications (2)

Publication Number Publication Date
JP2001109886A JP2001109886A (en) 2001-04-20
JP2001109886A5 true JP2001109886A5 (en) 2006-05-18

Family

ID=17685266

Family Applications (1)

Application Number Title Priority Date Filing Date
JP28495699A Abandoned JP2001109886A (en) 1999-10-05 1999-10-05 Method and device for image processing

Country Status (1)

Country Link
JP (1) JP2001109886A (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6248647B2 (en) * 2014-01-22 2017-12-20 富士通株式会社 Image collation method, image processing system, and program

Similar Documents

Publication Publication Date Title
Zeng et al. Image retrieval using spatiograms of colors quantized by gaussian mixture models
Tang et al. Histogram of oriented normal vectors for object recognition with a depth sensor
Chen et al. 3D free-form object recognition in range images using local surface patches
TWI281126B (en) Image detection method based on region
Herdiyeni et al. Combination of morphological, local binary pattern variance and color moments features for indonesian medicinal plants identification
CN105335722A (en) Detection system and detection method based on depth image information
CN107423768A (en) The image Hash sequence generating method combined based on SURF and PCA
CN108596197A (en) A kind of seal matching process and device
CN107392142B (en) Method and device for identifying true and false face
EP1267275A4 (en) Image retrieval/distribution system and image retrieval/distribution method
KR20080088778A (en) System and method for image identifier based on singular value decomposition and feature point
CN112801846B (en) Watermark embedding and extracting method and device, computer equipment and storage medium
CN104966081B (en) Spine image-recognizing method
CN104462382A (en) Trademark image inquiry method
CN104680158A (en) Face recognition method based on multi-scale block partial multi-valued mode
Trung et al. Blind inpainting forgery detection
CN109118449B (en) Current carpet type identification system
CN110599478A (en) Image area copying and pasting tampering detection method
CN110633749A (en) Three-dimensional point cloud identification method based on improved viewpoint feature histogram
JP2001109886A5 (en)
CN104766323B (en) A kind of Point matching method of remote sensing images
CN103310134A (en) Vector data watermark anti-counterfeiting method based on geographical semantics support
CN103679170B (en) Method for detecting salient regions based on local features
WO2015168362A1 (en) Method and apparatus for processing block to be processed of urine sediment image
Patil et al. Image hashing by LoG-QCSLBP