A projection based method for scratch and wire removal from digital images

Assignee: Sony Corp

•

Filing: 1996-10-17

•

Granted: —

•

Expires: 2016-10-17

📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract The present invention is directed to a method of scratch removal from digitized images wherein a "scratched" portion of a digitized image which requires repair or noise removal, is identified (Step 2) and a binary mask is generated (Step 3) which distinguishes the defined "scratched" portion from the other portions of the digitized image. A repair window (Step 4) is defined which delineates the portion of the digitized image desired to be repaired of scratches or other noise. A sample window having the same dimensional attributes as the repair window, and preferably free of any scratch or other noise, is then defined (Step 5). The sample window is preferably selected so as to be as close in resemblance to the repair window as possible. Image data of the repair window and the sample window are converted (Step 6, 7) into the frequency domain and then processed (Step 8) to yield appropriate pixel data for transfer into the repair window so as to remove the scratch noise in such a way that alignment of edges, as well as textures and other details are maintained in the resulting digitized image without degrading existing non-scratched image areas.

Inventors

👤 Anil c/o Sony Corp. Hirani
👤 Takashi c/o Sony Corp. Totsuka

Bibliographic Details

Application Number —

Jurisdiction EP

Data Source google_patents

Last Synchronized 2026-11-09 11:51

CPC Classifications

G; G06; G06T; G06T5/00; G06T5/77; G06T5/10; G06V; G06V10/00; G06V10/20; G06V10/30; G06T2207/00; G06T2207/20; G06T2207/20048; G06T2207/20056

Claims Analysis

Total Claims: 2

Claim 1

Independent Claim 225 words

A method of removing scratch noise from a digitized image comprising the steps of: inputting (20) digitized image data; displaying said digitised image data on a display device (10); identifying a scratch noise area in said digitized image data displayed on said display device (10); generating (35) binary mask data which distinguishes pixels within the area of said identified scratch noise from pixels within the remainder of said displayed image; storing said binary mask data in a storage memory (25); defining a repair window area on said displayed image; said repair window area contains identified scratch noise areas; storing in (25) memory data representing values of pixels within said repair window area; defining a sample window area within said displayed image; said sample window is chosen so as to resemble the features and values of said repair window area; calculating (35) the FFT of said repair window data; calculating the FFT of said sample window data; generating new image data in accordance with said repair window FFT data, said sample window FFT data; calculating the IFFT of said new image data; conforming the values of said new image data to predefined limits; generating new repair window data in accordance with said new image data, unscratched repair window data and said binary mask data; and replacing said repair window data with said new repair window data.

Claim 2

Independent Claim 203 words

An apparatus for removal of scratch noise from a digitized image comprising: an input device (20) for receiving digitized image data; display device (10) for displaying digitized image data; scratch identifying means for identifying scratch noise within a digitized image displayed on said display device; processor (35) for generating binary mask data which distinguishes identified scratch noise within a digitized image from other areas of said digitized image; means for identifying repair window and sample window areas; first storage memory (25) for storing data representing said identified repair window area; second storage memory for storing data representing said identified sample window area; third storage memory for storing said binary mask data, and said processor (35) further comprises means for calculating fast fourier transform data for said repair window data and said sample window data, means for generating new image data in accordance with the fast fourier transform data of said sample and repair windows, means for calculating inverse fast fourier transform data for said new image data; means for conforming values of said new image data to predefined limits, and means for generating new repair window data in accordance with said new image data, unscratched repair window data and said binary mask data.

References Cited (Backward Citations)

No citation records indexed.

USPTO Annuity & Maintenance Windows

Maintenance schedule is calculated for granted US utility patents.

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag

Select Category

Notes (Optional)