Changes between Version 18 and Version 19 of GSoC/2016/Additional_segmentation_algorithms/mean_shift

Timestamp:

Sep 13, 2016, 6:11:12 PM (8 years ago)

Author:

hao2309

Comment:

--

GSoC/2016/Additional_segmentation_algorithms/mean_shift

@@ -16 +16 @@
 
 '''-a''' [[BR]] Use adaptive bandwidth for mean shift [[BR]]  Range (spectral) bandwidth is adapted for each moving window
-  
+
 '''-p''' [[BR]] Use progressive bandwidth for mean shift [[BR]] Spatial bandwidth is increased, range (spectral) bandwidth is decreased in each iteration
 
@@ -72 +72 @@
 '''--ui'''
     run with the user interface mode
-'''--help'''
-
-'''--verbose'''
-
-'''--quiet'''
+
 
 = DESCRIPTION =
 
-GRASS GIS has the i.segment which provides the possibility to segment an image into objects. This is a basic step in object-based image analysis (OBIA). Currently, the module only provides one segmentation algorithm: region-growing. The code of i.segment was structured in a way that allows addition of other algorithms. It would be more useful and comprehensive to add mean-shift to the i.segment module.
-
 Mean shift segmentation is a local homogenization technique that is very useful for damping shading or tonality differences in localized objects. For the algorithm implementation of this case, basically the algorithm replaces each pixel with the mean of the pixels in a range-r neighborhood and whose value is within a distance d. The Mean Shift usually has 3 important parameters: 1) A distance function for measuring distances between pixels. Usually the Euclidean distance, but any other well-defined distance function could be used. The Manhattan Distance is another useful choice sometimes. 2) A radius (spatial bandwidth). All pixels within this radius (measured according the above distance) will be accounted for the calculation. 3) A value difference (range bandwidth). From all pixels inside radius r, we will take only those whose values are within this difference for calculating the mean [1].
 
 = COMPARISON = 
+== Mean shift ==
+Numerous nonparametric clustering methods can be separated into two parts:hierarchical clustering and density estimation. Hierarchical clustering composes either aggregation or division based on some proximate measure. The concept of the density estimation-based nonparametric clustering method is that the feature space can be considered as the experiential probability density function (p.d.f.) of the represented parameter. The mean shift algorithm can be classified as density estimation. It adequately analyzes feature space to cluster them and can provide reliable solutions for many vision tasks. The basics of mean shift are discussed as below [3].
+
+'''Pros:'''
+* An extremely versatile tool for feature space analysis. Mean-shift algorithm is applicable not only in image segmentation, but also in motion detector, image filtering and etc.
+* Suitable for arbitrary feature spaces.
+'''Cons:'''
+* The range bandwidth (hr) and spatial bandwidth (hs) are the only two factors can control the output, especially the range bandwidth results in a tremendous influence to the result.
+* The computation time is quite long, for the iterations of anisotropic filtering, clustering and merging small objects.
 
 == Seeded Region Growing ==
@@ -95 +98 @@
 * The blocky segment problem could be reduced by splitting in higher level, but the trade off is that the computation time will arise.
 
-== Mean shift ==
-Numerous nonparametric clustering methods can be separated into two parts:hierarchical clustering and density estimation. Hierarchical clustering composes either aggregation or division based on some proximate measure. The concept of the density estimation-based nonparametric clustering method is that the feature space can be considered as the experiential probability density function (p.d.f.) of the represented parameter. The mean shift algorithm can be classified as density estimation. It adequately analyzes feature space to cluster them and can provide reliable solutions for many vision tasks. The basics of mean shift are discussed as below [3].
-
-'''Pros:'''
-* An extremely versatile tool for feature space analysis. Mean-shift algorithm is applicable not only in image segmentation, but also in motion detector, image filtering and etc.
-* Suitable for arbitrary feature spaces.
-'''Cons:'''
-* The range bandwidth (hr) and spatial bandwidth (hs) are the only two factors can control the output, especially the range bandwidth results in a tremendous influence to the result.
-* The computation time is quite long, for the iterations of anisotropic filtering, clustering and merging small objects.
 
 == Other segmentation methods ==