필터 지우기
필터 지우기

wavelet transform for feature vector

조회 수: 1 (최근 30일)
Jack Sparrow
Jack Sparrow 2013년 1월 26일
Can someone please explain what the coefficients from Discrete wavelet transform are for and how can they be used to form feature vector for classifier. I have been spending days reading journal articles on it now but still don't understand.

채택된 답변

Wayne King
Wayne King 2013년 1월 26일
Hi Jack The 2D DWT is basically a subband filtering of the image in a combination of highpass and lowpass parts. For example, at level 1, you get 4 images: LL, LH, HL, and HH (L = lowpass, H= highpass)
So LL is lowpass in the x-direction and lowpass in the y-direction. LH is lowpass in the x-direction and highpass in the y-direction.
The wavelet transform then iterates on the LL image to obtain more narrow subbands at successive levels. Another 4 images at each level.
Because features in many real-world signals and images tend to be sparse in these subbands, the wavelet transform coefficients can localize the features of interest and do it more sparsely than other representations. Of course this depends on the type of image (signal).
As one example, images which are piecewise smooth punctuated by abrupt transitions (edges) are sparse in the wavelet domain (represented by relatively few wavelet coefficients) while not so sparsely represented in the Fourier domain (the abrupt transitions require a large number of high frequency terms to approximate).
  댓글 수: 3
Romendra
Romendra 2013년 1월 26일
Jack,
you may want to look into 2 functions appcoef2 and wavedec2. The wavedec2 does a 2d wavelet decomposition and then the appcoef2 can be used to extract the reduced approximate coefficients. If you need the detailed coefficients then u can use detcoef2. Even if you used the wavelet gui you should have been able to get the reduced coefficients Hope this helps.
Thanks
romendra
Jack Sparrow
Jack Sparrow 2013년 1월 27일
thanks. I would do that. sorry for replying late, the time difference is huge

댓글을 달려면 로그인하십시오.

추가 답변 (0개)

카테고리

Help CenterFile Exchange에서 AI for Signals and Images에 대해 자세히 알아보기

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by