让你垂涎的计算机视觉 Matlab 源码~ 希望对你有用~

（http://blog.renren.com/share/340039432/17742227002?from=0101010202&ref=timelinefeed&sfet=102&fin=0&fid=26025230354&ff_id=340039432&platform=6&expose_time=1416224932）

（http://blog.renren.com/GetEntry.do?id=498466017&owner=231033016）

不小心浏览到Peter Kovesi的个人主页，打开了这个计算机视觉处理的宝库，在文后有链接，大家可以去下载奥

MATLAB and Octave Functions 
for Computer Vision and Image Processing

Feature Detection via Phase Congruency

 

• Phase congruency is an illumination and contrast invariant measure of feature significance. Unlike gradient based feature detectors, which can only detect step features, phase congruency correctly detects features at all kind of phase angle, and not just step features having a phase angle of 0 or 180 degrees.

• phasecongmono.m This function computes phase congruency via monogenic filters. It has excellent speed and much reduced memory requirements compared to the other phase congruency functions below. However you may prefer the output from phasecong3's oriented filters.
• phasecong3.m This function supersedes phasecong2.m and phasecong.m being faster and requiring less memory.

   

  • Deprecated: phasecong.m Original code for calculating phase congruency in an image. This function also returns a feature type image. Note this function is superseded by phasecong2.m and phasecong3.m and is only here for reference.
  • Deprecated: phasecong2.m Phase congruency code that combines edge and corner detection, and provides better localization. Note this function is superseded by phasecong3.m and phasecongmono.m and is only here for reference.

   

• dispfeat.m This function provides visualisation and statistics of the different feature types found in an image by phasecong. Typically you will find a broad distribution of all feature types between step edges and lines. This function needs edgelink.m (see below).
• odot.m Demonstrates the actions of the 'Odot' and 'Oslash' operators on a 1D signal. These operators allow one to decompose and combine signals in a way that is consistent with the Local Energy model of feature perception.
• spatialgabor.m applies a single oriented Gabor filter to an image.

  phase symmetry image

• phasesym.m Code for calculating phase symmetry. This can be used as a line and blob detector. Phase symmetry is an illumination and contrast invariant measure of symmetry in an image. (A bright circle is not more 'symmetric' than a grey circle as can be the case with some other measures!). (Updated May 2009)
• phasesymmono.m This function computes phase symmetry via monogenic filters. Has excellent speed and much reduced memory requirements compared to phasesym.m However you may prefer the output from phasesym's oriented filters.
• gaborconvolve.m Code for convolving an image with a bank of log-Gabor filters. A pre-processing step for texture analysis, feature detection and classification, etc.
• plotgaborfilters.m A function for plotting log-Gabor filters. This function is useful for seeing what effect the various parameter settings have on the formation of a log-Gabor filter bank used in the functions above.
• monofilt.m An implementation of Felsberg's monogenic filters. This function applies a bank of monogenic filters to an image to obtain the 2D analytic signal over a number of scales. As in gaborconvolve this can be used as a pre-processing step for texture analysis, feature detection and classification, etc.
• An explanation of the implementation of convolution with log-Gabor filters used in the functions above.

References:

• Peter Kovesi, "Symmetry and Asymmetry From Local Phase". AI'97, Tenth Australian Joint Conference on Artificial Intelligence. 2 - 4 December 1997. Proceedings - Poster Papers. pp 185-190.
• Peter Kovesi, "Image Features From Phase Congruency". Videre: A Journal of Computer Vision Research. MIT Press. Volume 1, Number 3, Summer 1999.
• Peter Kovesi, "Edges Are Not Just Steps". Proceedings of ACCV2002 The Fifth Asian Conference on Computer Vision, Melbourne Jan 22-25, 2002. pp 822-827.
• Peter Kovesi, "Phase Congruency Detects Corners and Edges". The Australian Pattern Recognition Society Conference: DICTA 2003. December 2003. Sydney. pp 309-318.

---

Spatial Feature Detection

• canny.m Canny edge detector.
• harris.m Harris corner detector.
• fastradial.m An implementation of Loy and Zelinski's fast radial feature detector.
• gaussfilt.m Wrapper function for convenient Gaussian filtering.
• derivative5.m computes 1st and 2nd derivatives of an image using the 5-tap coefficients given by Farid and Simoncelli. Use this function instead of MATLAB's GRADIENT function for much more accurate results.
• derivative7.m computes derivatives using the 7-tap coefficients given by Farid and Simoncelli.

Reference:

• Scanned images of my photocopy of Harris and Stephens' paper 'A Combined Corner and Edge Detector'.

---

Integral Images

• integralimage.m computes integral image of an image.
• integralfilter.m performs filtering using an integral image.
• intfilttranspose.m transposes an integral image filter specification.
• integaverage.m performs averaging filtering using an integral image. Computation cost is independent of averaging filter size.
• integgaussfilt.m This function approximates Gaussian filtering by repeatedly applying integaverag.m . This allows smoothing at a very low computational cost that is independent of the Gaussian size.
• solveinteg.m This function is used by integgausfilt.m to solve for the multiple averaging filter widths needed to approximate a Gaussian of desired standard deviation.

Reference:

• Tech report describing high speed approximate Gaussian filtering via integral images.