KR960006495A - Mixed Median Filtering Method - Google Patents

Mixed Median Filtering Method Download PDF

Info

Publication number
KR960006495A
KR960006495A KR1019940018935A KR19940018935A KR960006495A KR 960006495 A KR960006495 A KR 960006495A KR 1019940018935 A KR1019940018935 A KR 1019940018935A KR 19940018935 A KR19940018935 A KR 19940018935A KR 960006495 A KR960006495 A KR 960006495A
Authority
KR
South Korea
Prior art keywords
value
median filtering
pixels
variance
pixel
Prior art date
Application number
KR1019940018935A
Other languages
Korean (ko)
Other versions
KR100213019B1 (en
Inventor
이재준
Original Assignee
김광호
삼성전자 주식회사
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 김광호, 삼성전자 주식회사 filed Critical 김광호
Priority to KR1019940018935A priority Critical patent/KR100213019B1/en
Publication of KR960006495A publication Critical patent/KR960006495A/en
Application granted granted Critical
Publication of KR100213019B1 publication Critical patent/KR100213019B1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/14Picture signal circuitry for video frequency region
    • H04N5/21Circuitry for suppressing or minimising disturbance, e.g. moiré or halo
    • H04N5/213Circuitry for suppressing or minimising impulsive noise
    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03HIMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H17/00Networks using digital techniques
    • H03H17/02Frequency selective networks
    • H03H17/0202Two or more dimensional filters; Filters for complex signals
    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03HIMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H17/00Networks using digital techniques
    • H03H17/02Frequency selective networks
    • H03H17/0202Two or more dimensional filters; Filters for complex signals
    • H03H2017/0207Median filters

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Physics (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Image Processing (AREA)
  • Picture Signal Circuits (AREA)

Abstract

본 발명에 의한 혼성 메디안 필터링방법에서는 윈도우내의 부필터링 과정에서 윈도우내의 화소분포에 따라 가중치를 주어 부필터링을 수행한 후 제1 내지 제3단계 메디안 필터링을 수행함으로써 경계 부분의 정보손실을 최소화하면서 임펄스 잡음을 효과적으로 제거할 수 있다.In the mixed median filtering method according to the present invention, impulse is minimized while minimizing information loss at the boundary part by performing sub-filtering after performing sub-filtering by weighting according to the pixel distribution in the window during sub-filtering in the window. Noise can be effectively removed.

Description

혼성 메디안 필터링방법Mixed Median Filtering Method

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음As this is a public information case, the full text was not included.

제1도는 메디안 필터링을 위한 5×5 윈도우를 나타낸 도면이다.1 shows a 5x5 window for median filtering.

제5a도∼5d도는 양방향 메디안 피럴의 윈도우 형태를 나타낸 도면이다.5A to 5D are diagrams showing the window form of the bidirectional median piral.

제7도는 본 발명에 의한 혼성 메디안 필터링방법을 설명하기 위한 도면이다.7 is a diagram for explaining a hybrid median filtering method according to the present invention.

Claims (5)

영상신호에 혼입된 임펄스 잡음을 제거하기 위한 혼성 메디안 필터에 있어서, 메디안 필터링을 의한 N×N 윈도우내 화소의 평균과 분산을 구해 부필터링할 화소에 부여할 가중치를 연산하는 가중치 연산과정; 상기 가중치 연산과정에서 산출된 가중치를 이용하여 부필터링을 수행하는 부필터링과정; 상기 부필터링과정의 출력과 자기 화소를 이용하여 1단계 메디안 필터링을 수행하는 제1메디안 필터링과정; 상기 제1메디안 필터링 과정의 출력과 자기 화소를 이용하여 2단계 메디안 필터링을 수행하는 제2메디안 필터링과정; 및 상기 제2메디안 필터링과정의 출력과 자기 화소를 이용하여 3단계 메디안 필터링을 수행하는 제3메디안 필터링과정을 포함하는 것을 특징으로 하는 혼성 메디안 필터링방법.A hybrid median filter for removing impulse noise mixed in an image signal, comprising: a weighting step of calculating weights to be assigned to pixels to be sub-filtered by obtaining an average and a variance of pixels in an N × N window by median filtering; A subfiltering step of performing subfiltering using the weights calculated in the weighting operation; A first median filtering process for performing a one-step median filtering using the output of the sub-filtering process and the magnetic pixel; A second median filtering process for performing two-step median filtering by using the output of the first median filtering process and the magnetic pixels; And a third median filtering process for performing three-step median filtering using the output of the second median filtering process and the magnetic pixels. 제1항에 있어서, 상기 가중치 연산과정에서는 잡음이 혼입된 영상에서 상기 N×N 윈도우를 이동시키면서 윈도우내의 전 화소들을 평균한 값과 부필터내의 화소들과의 차와 부필터내의 분산치를 비교하여 가중치를 조절하는 것을 특징으로 하는 혼성 메디안 필터링방법.The method of claim 1, wherein in the weighting operation, the average value of all pixels in the window is compared with the difference between the pixels in the subfilter and the variance in the subfilter while moving the N × N window in the noise mixed image. Hybrid median filtering method characterized in that the weight adjustment. 제2항에 있어서, 상기 가중치 연산과정에서는 부필터내의 첫번째 화소값에서 평균치를 뺀 값이 분산치보다 작고, 두번째 화소값에서 평균치를 뺀 값이 분산치보다 큰 경우 상기 부필터를 구성하는 두 화소에 동일한 가중치를 주는 것을 특징으로 하는 혼성 메디안 필터링방법.3. The pixel of claim 2, wherein in the weighting operation, when the value obtained by subtracting the average value from the first pixel value in the subfilter is smaller than the variance value, and the value obtained by subtracting the average value from the second pixel value is larger than the variance value, two pixels constituting the subfilter. Hybrid median filtering method characterized in that to give the same weight to. 제2항에 있어서, 상기 가중치 연산과정에서는 부필터내의 첫번째 화소값에서 평균치를 뺀 값이 분산치보다 크고, 두번째 화소값에서 평균치를 뺀 값이 분산치보다 작은 경우에는 두 화소에 동일하게 0.5의 가중치를 주는 것을 특징으로 하는 혼성 메디안 필터링방법.3. The method of claim 2, wherein in the weighting operation, when the value obtained by subtracting the average value from the first pixel value in the subfilter is greater than the variance value, and the value obtained by subtracting the average value from the second pixel value is smaller than the variance value, the two pixels are equal to 0.5. Hybrid median filtering method characterized in that the weighting. 제2항에 있어서, 상기 가중치 연산과정에서는 부필터내의 첫번째 화소값에서 평균치를 뺀 값이 분산치보다 크고, 두번째 화소값에서 평균치를 뺀 값이 분산치보다 크지만 첫번째 화소값에서 평균치를 뺀 값보다 작은 경우에는 첫번째 화소값에 잡음이 많이 혼입되어 있다고 판단하고 두번째 화소값보다 작은 값의 가중치를 주는 것을 특징으로 하는 혼성 메디안 필터링방법.3. The method of claim 2, wherein in the weighting operation, a value obtained by subtracting the average value from the first pixel value in the subfilter is greater than the variance value, and a value obtained by subtracting the average value from the second pixel value is larger than the variance value but subtracted from the first pixel value If smaller, it is determined that a lot of noise is mixed in the first pixel value, and a weight value of a value smaller than the second pixel value is mixed. ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019940018935A 1994-07-30 1994-07-30 Hybrid median filtering method KR100213019B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019940018935A KR100213019B1 (en) 1994-07-30 1994-07-30 Hybrid median filtering method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019940018935A KR100213019B1 (en) 1994-07-30 1994-07-30 Hybrid median filtering method

Publications (2)

Publication Number Publication Date
KR960006495A true KR960006495A (en) 1996-02-23
KR100213019B1 KR100213019B1 (en) 1999-08-02

Family

ID=19389570

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019940018935A KR100213019B1 (en) 1994-07-30 1994-07-30 Hybrid median filtering method

Country Status (1)

Country Link
KR (1) KR100213019B1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210070969A (en) * 2009-06-17 2021-06-15 아이디어허브 주식회사 Method for multiple interpolation filters, and apparatus for encoding by using the same
KR20220036686A (en) * 2020-09-16 2022-03-23 이호재 System for plant cultivation

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101561778B1 (en) 2014-05-14 2015-10-19 세메스 주식회사 Method for estimating a background component of captured image and method for detecting a defect of wafer using the same

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210070969A (en) * 2009-06-17 2021-06-15 아이디어허브 주식회사 Method for multiple interpolation filters, and apparatus for encoding by using the same
KR20220036686A (en) * 2020-09-16 2022-03-23 이호재 System for plant cultivation

Also Published As

Publication number Publication date
KR100213019B1 (en) 1999-08-02

Similar Documents

Publication Publication Date Title
EP0797349A3 (en) Signal adaptive postprocessing system for reducing blocking effects and ringing noise
EP0961229A3 (en) Non-linear adaptive image filter forsuppressing blocking artifacts
EP1168824A2 (en) Image processing method for reducing noise and blocking artifact in a digital image
US7551795B2 (en) Method and system for quantization artifact removal using super precision
JPH06121192A (en) Noise removing circuit
EP1093087A3 (en) Removing chroma noise from digital images by using variable shape pixel neighborhood regions
WO1997022948A3 (en) System and method for minimizing blocking artifacts in a filtered image
EP0809209A3 (en) Image enhancing method using lowpass filtering and histogram equalization and a device therefor
EP1174824A3 (en) Noise reduction method utilizing color information, apparatus, and program for digital image processing
EP0767444A3 (en) Antialiasing of silhouette edges
EP1032196A3 (en) Method and apparatus for reducing motion artifacts and noise in video image processing
EP1416464A3 (en) Method of and apparatus for displaying halftone images
CN101472056A (en) Noise reduction device and noise reduction method
US8891898B2 (en) Signal processing device and control program for sharpening images
US20090074318A1 (en) Noise-reduction method and apparatus
JP3171283B2 (en) Noise measurement method and device
CA2090382A1 (en) Moving picture receiving method and apparatus
KR960006495A (en) Mixed Median Filtering Method
EP0884693A3 (en) Interpolation method for binary image
TW200514427A (en) Device and method for estimating the noise of a video signal
WO2002030100A3 (en) Noise reduction in images, by adding a pixel to a region (r), adding surrounding pixel pairs to the region if they fulfil a condition and determine a mean value for the region
JP2003528547A (en) N-dimensional filter and method for N-dimensionally filtering original image pixels
EP1342367B1 (en) Spatial smoothing process and device for dark regions of an image
EP0890921A3 (en) Interpolation method for binary image
WO2002086820A3 (en) Automatic background removal method and system

Legal Events

Date Code Title Description
A201 Request for examination
E701 Decision to grant or registration of patent right
GRNT Written decision to grant
FPAY Annual fee payment

Payment date: 20080429

Year of fee payment: 10

LAPS Lapse due to unpaid annual fee