文件列表:

data (0, 2017-02-07)
data\Image_1.tiff (307776, 2017-02-07)
data\Image_10.tiff (307776, 2017-02-07)
data\Image_2.tiff (307776, 2017-02-07)
data\Image_3.tiff (307776, 2017-02-07)
data\Image_4.tiff (307776, 2017-02-07)
data\Image_5.tiff (307776, 2017-02-07)
data\Image_6.tiff (307776, 2017-02-07)
data\Image_7.tiff (307776, 2017-02-07)
data\Image_8.tiff (307776, 2017-02-07)
data\Image_9.tiff (307776, 2017-02-07)
irisseg_demo.m (5366, 2017-02-07)
irisseg_main.m (4107, 2017-02-07)
modules (0, 2017-02-07)
modules\eyelid (0, 2017-02-07)
modules\eyelid\adjgamma.p (209, 2017-02-07)
modules\eyelid\canny.p (474, 2017-02-07)
modules\eyelid\circle_hough.p (691, 2017-02-07)
modules\eyelid\circle_houghpeaks.p (1254, 2017-02-07)
modules\eyelid\circlepoints.p (235, 2017-02-07)
modules\eyelid\geteyelid.p (2487, 2017-02-07)
modules\eyelid\geteyelid_blocks.p (467, 2017-02-07)
modules\eyelid\hysthresh.p (444, 2017-02-07)
modules\eyelid\lower_eyelid_icb.p (483, 2017-02-07)
modules\eyelid\nonmaxsup.p (562, 2017-02-07)
modules\eyelid\upper_eyelid_icb.p (512, 2017-02-07)
modules\iris (0, 2017-02-07)
modules\iris\getiris_blocks.p (299, 2017-02-07)
modules\iris\iris_boundary.p (1011, 2017-02-07)
modules\iris\iris_boundary_actual.p (2159, 2017-02-07)
modules\iris\iris_boundary_actual_double.p (2662, 2017-02-07)
modules\iris\shiftiris.p (185, 2017-02-07)
modules\normalization (0, 2017-02-07)
modules\normalization\NEWnormaliseiris.p (902, 2017-02-07)
modules\normalization\circlecoords.p (318, 2017-02-07)
modules\normalization\convert_to_polar.p (614, 2017-02-07)
modules\preprocessing (0, 2017-02-07)
modules\preprocessing\RankOrderFilter.p (352, 2017-02-07)
... ...

#### IrisSeg: A Fast and Robust Iris Segmentation Framework for Non-Ideal Iris Images This is the matlab implementation of the paper ***IrisSeg: A Fast and Robust Iris Segmentation Framework for Non-Ideal Iris Images*** This code is provided only for Research Purpose. Algorithms details are given in the paper. #### Prerequisites The code is tested on ***Matlab R2014b*** with *i5* processor and 4GB RAM with Windows OS. #### How to use this code If you want to see the complete cycle of the segmentation process the best way is to use the matlab script **irisseg_demo** after making required changes in it- 1. Set the input folder path (at line 33 in irisseg_demo.m) 2. Set the output folder (at line 34 in irisseg_demo.m) 3. Set the correct image extension (e.g.'tiff', '.JPEG' etc. at line 40 in irisseg_demo.m) Under default settings, this will create a folder named **output** if not already created and inside that four files corresponding to each input file will be created in the following format iris_ irismask_ normalizediris_ normalizedirismask_ The details for the above files are as follows * **iris_filename** This file will be the same size as the input image and shows the segmented iris. Only iris pixels are shown and everything else is white(255,255,255) * **irismask_filename** This is the mask file for the iris segmentation with same size as the input file. All non-iris pixels are black and all the iris pixels are white. * **normalizediris_filename** This is the normalized iris segmented and its width is 512 pixels and the height is *** pixels. The size is irresepective of the input image size. In this file all the iris-pixels are unchanged and the non-iris pixels are white. * **normalizedirismask_filename** This is the mask file for the normalized segmented iris. It is 512 pixels wide and *** pixels height. The size is invariant of the input image and all the iris-pixels are black and the non-iris pixels are white. Most of the files are protected but one can get the idea of high level modules in *irisseg_main.m* **Parameters** Certain settings are cutomizable in *irisseg_demo.m* such as outputpath, inputpath, image extensions, writing of intermediate images etc. Certain parameters of the code are customizable in *irisseg_main.m*, such as- 1. Scale: This version of Code has not been verified to work with other Values for Scale Parameter. Hence We do not recommend Changing value of Scale Parameter. 2. Size of Normalized Iris **Terms and Conditions** This code is provided "as is", without any warranty, and for research/academic purposes only. By downloading the code, you agree with the terms and conditions. **Reference** Please remember to cite following reference [1] if you make use of this code in any publication. [1] Abhishek Gangwar, Akanksha Joshi, Ashutosh Singh, Fernando Alonso-Fernandez and Josef Bigun, IrisSeg: A Fast and Robust Iris Segmentation Framework for Non-Ideal Iris Images, International Conference on Biometrics (ICB), 2016 ### Contact If you have queries/suggestions regarding the code you can contact at * abhishekg [at] cdac [dot] in * akanksha [at] cdac [dot] in To get a copy of GroundTruth data, please contact- Dr. Fernando Alonso-Fernandez (http://islab.hh.se/mediawiki/Fernando_Alonso-Fernandez)

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：