Changes between Version 15 and Version 16 of GSoC/2016/Additional_segmentation_algorithms/mean_shift
- Timestamp:
- Sep 13, 2016, 9:05:35 AM (8 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
GSoC/2016/Additional_segmentation_algorithms/mean_shift
v15 v16 167 167 168 168 [[Image(New_MS_Fig1.jpg)]] 169 169 170 Compare with the original image, there are many small segmentations showing because minsize has been set to 1. 170 171 Then zoom into a smaller region and compare with the original image. 172 171 173 [[Image(New_MS_Fig2.jpg)]] 172 174 173 175 Compare with the original image, most of the features in original image has been segmented out. the noise can be significantly reduced by increasing the threshold of minsize. 176 174 177 [[Image(New_MS_Fig5.jpg)]] 175 178 179 == Segmentation using different range bandwidth and spatial bandwidth == 180 This example uses fixed range bandwidth(hr) value, larger hr value indicates pixel value in a larger range take part in the filtering calculation. 181 [[Image(New_MS_Fig3.jpg)]] 182 183 keeping all else constant, changed the value of spatial bandwidth. A larger spatial bandwidth hardly results in larger segments. 184 185 [[Image(New_MS_Fig4.jpg)]] 186 176 187 == Segmentation using adaptive bandwidth and progressive bandwidth == 177 178 179 188 This example uses -a and -p flag to add adaptive bandwidth and progressive bandwidth respectively 189 190 [[Image(New_MS_Fig6.jpg)]] 191 192 In the zoomed-in image, adaptive bandwidth enhanced the result significantly. Because range bandwidth is the most important parameter for mean shift. 193 194 = REFERENCES = 180 195 [1] http://stackoverflow.com/questions/4831813/image-segmentation-using-mean-shift-explained [[BR]] 181 196 [2] Adams, Rolf, and Leanne Bischof. "Seeded region growing." IEEE Transactions on pattern analysis and machine intelligence 16.6 (1994): 641-647[[BR]]