Software Listing of Author : "Su Dongcai"

Implementation of the level set method proposed in[1](ACWE) while add the regularity term[2] to avoid re-initialization. The core function to implement ACWE is 'acwe.m',while 'demo_acwe.m' is for demonstration purpose. Acknowledgement: Li chunming for his clearly coded distribution: http://www.mathworks.com/matlabcentral/fil...evel-set-for-im Reference: [1]. T. Chan and L. Vese, 'Active contours without edges,' IEEE Trans. Image Process., vol. 10, no. 2, pp. 266 277, Feb. 2001 [2]. Minimizaion of region-scalable fitting energy for image segmentation. IEEE Trans. Image Processing, vol. 17 (10), pp.1940-1949, 2008.by Li chunming etc.

  • Platform: Matlab, Scripts

This file implement the modified hausdorf distance[2], and apply it to solve a simple application--tile images classification with shape variance. See the 'description.pdf' for more details. [1] D. P. Huttenlocher, G. A. Klanderman, and W. J. Rucklidge, 'Comparing images using the Hausdorff distance', IEEE Trans. PAMI, vol. 15, pp. 850- 863, 1993. [2] Marie-Pierre Dubisson and A.K.Jain. A modified hausdorff distance for object matching

  • Platform: Matlab, Scripts

This file adapt a state of art structure 'disjoint set' to label objects in an binary 2D image. Which have similar effect as the matlab function 'bwlabel' in case of 8-connected objects. %please run the following commands to duplicate the screen shot: mex vcBwlable.cpp %compile the cpp file im = imread('coins.png'); L = djs_bwlable(im2bw(im)); %reference syllabus about Disjoint set: http://www.google.com.hk/url?sa=t&rct=j&q=disjoint+set&source=web&cd=4&ved=0CDwQFjAD&url=http%3A%2F%2Fwww.cs.iupui.edu%2F~xkzou%2Fteaching%2FCS580%2FDisjointSets.ppt&ei=vlEaT_iSKuWTiQfym6T3Cw&usg=AFQjCNG4v--8rDjH2j1f620owySUkZcRyA

  • Platform: Matlab, Scripts

Example: Counting the number of nuts as shown in the photo: count 3 by 3, reminder: 2 count 5 by 5, reminder: 3 count 7 by 7, reminder: 2 then the requested total number of nuts X can be calculated as follow command: x = sunzTheorem([2, 3, 2], [3, 5, 7]); The motivation and key ideas of the program is described in "program guild.pdf", Note: just throw out the designing ideas in a huristic way not for vigorous proof purpose.

  • Platform: Matlab, Scripts

An implementation of [1], it works in follows: 1. Compute the Jmap. 2. Compute the valleys of Jmap. 3. Valleys growing as the oversegmentation. 4. Region merging. Example usage: %labels = colorImgSeg('imagessflowg.jpg', 12, 5, 10,0.35, 0.02, 0.88); To run the code, you have to firstly generate the necessary dlls in "local_min", "Jmap", "seededRegionGrowing" and "Region_merging" in each folder, run the corresponding compling m-file to generate dlls, which are: local_min_complie, Jmap_complie, seededRegionGrowing_complie, and RegionMerging_complie respectively.And then copy the corresponding dlls to the parent directory reference: [1]Yining Deng, B.S. Manjunath, Hyundoo Shin, "Color Image Segmentation," cvpr, vol. 2, pp.2446, 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99)- Volume 2, 1999...

  • Platform: Matlab, Scripts

This is a popular problem in MSRA interview. interviewer: How do you know about the operation system? interviewee: Very!...... interviewer: Well, can you make the cpu time curve act as a straight line? can you make it act as a sin curve? interviewee: ..... Example: %complie: mex cConductCpuTime.cpp % conductCpuTime(200, 300, 5); Then press Ctrl+Alt+Delete to check the cpu time Curve

  • Platform: Matlab, Scripts

This is an implementation of [1]. The result is slightly different from website[1], because we apply adjacent neighborhood rather than K nearest neighborhood. The following is an example usage: %complie:(only once) GraphSeg_complie %Read an image img = imread('imagesrice.jpg'); %segmentation [L, contour] = graph_segment(img, 1, 3, 100); %display result subplot(2, 1, 1), imshow(img), title('original image'); subplot(2, 1, 2), imshow(label2rgb(L)),title('segmented result'); [1]Efficient Graph-Based Image Segmentation http://people.cs.uchicago.edu/~pff/segment/

  • Platform: Matlab, Scripts

This file is a fast computation of Jmap, which introduce in [1]. Jmap is an edge indicator which insensitive to noise and texture. sample usage: 1.compile: mex scale_sum.cpp 2. read an image: img = imread('tiger.jpg'); 3.Jmap computation: Jmap = cmp_Jmap(img, 3, 7); 4. display image: subplot(1, 2, 1), imshow(img, []), title("original image"); subplot(1, 2, 2), imshow(Jmap, []), title("Jmap"); reference: [1]Yining Deng, B.S. Manjunath, Hyundoo Shin, "Color Image Segmentation," cvpr, vol. 2, pp.2446, 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99)- Volume 2, 1999

  • Platform: Matlab, Scripts

Let r_scale be the neighborhood radius scale, a brute search method would require N*r_scale to detect the valleys with neighborhood size of r_scale in 1d signal or N*r_scale*r_scale complexity in 2d image. Which is too slow in large data size. In this file, the complexity is reduced in N*log(r_scale) in both 1d and 2d data. Notes: To run the code, firstly you have to generate the dll using: local_min_complie;

  • Platform: Matlab, Scripts

%the function work in 4 steps: %1. compute the inverse eucilidean distance trasform ("inv_distance") of "bw" %2. compute the ridges(watersheds) of "inv_distance" by watershed algorithm %3. compute the gradient of ridges "rigid_grad" %4. compute the final skeleton by Thresholding, dilation and thining "rigid_grad". %The meaning of "gradient of ridges" is defined due to %Nicholas R. Howe and Alex Telea: "A point on the skeleton sits at the center of a circle that touches the edge of the figure at multiple points. %The intensity in the gray-level skeleton image ("rigides_grad" in this file) is based on the shortest distance you would have to travel around the perimeter of %the figure to connect the most distant two points." %Acknowledgement: %[1] This work was inspired by Nicholas R. Howe. Please go to his website...

  • Platform: Matlab, Scripts

To duplicate results of the screenshot, please run: test_WHMM This script is an implementation of Reference [1], including 2 parts: 1. 1D signal denoising (page 9~11) 2. 1D random process(RP) classification (page 12) Reference: [1] Wavelet-Based Statistical Signal Processing Using Hidden Markov Models: MS Crouse, RD Nowak, RG Baraniuk - IEEE transactions on signal processing, 1998 - dsp.rice.edu. Available at: http://scholarship.rice.edu/bitstream/hand....PDF?sequence=1 Acknowledgment: The author wish to thank Prof. Justin Romberg for his 'hmt1d' toolbox and his kindly help of how to use it.

  • Platform: Matlab, Scripts
New Reviews

Print Inspector

Whether you are using a printer connected to your local desktop computer or a print server sometimes you need to see whats going on in your printer by having a close look at the print jobs sent by the users. Print Inspector is such type software ...


Disk Genius

Disk Genius was formerly known as Partition Guru which is an all-in-one solution for disk partition management, data recovery and disk repairing. It can recover lost files from partition, backup and restore partition table, clone partition or clone ...


Safe In Cloud

It is quite difficult to find any computer users nowadays who do not need to use the web, for any purpose. Whether it is for education or entertainment, the web is your best resource. You also need to use various online services for banking, ...


PictBear

Nowadays, a majority of PC users resort to digital image editing once in a while. It may be required for retouching image captured in smart phones or for creating an album in Facebook. For some people, tinkering with digital images is a pastime too. ...


Panda Cloud Cleaner

PC security has become a tricky and complicated subject both for home users and network supervisors, with complexity and risk factors multiplying over the years. Using a single antivirus app may not be enough anymore. That explains why many users ...


SnapCrab

Nearly every PC users need to take screenshots from time to time, whether it is for personal or professional needs. While using the basic Windows screenshot capture method is available, it is not adequate for everyone. When you want to capture ...


BitKiller

Dealing with various types of data efficiently holds the key to success in any computing task. Apart from generating new data and modifying existing files, it becomes necessary to delete existing files at times. While most users resort to using ...


New Downloads

@SwIt Printfil

Print from DOS, Unix, Linux to
Windows printers: USB, GDI,
PDF, e-mail, network, IP, fax.
Capture LPT1:-LPT9: & ...

MindView

MindView is a professional
mind mapping software that
allows you to visually
brainstorm, organize and
present ideas. With ...

Project Timer

Project Timer, a project timer
for the Windows taskbar
notification area. Project
Timer is a simple but powerful
windows ...

SoftPerfect WiFi Guard

A specialised network scanner
that helps to protect and keep
your WiFi network secure. It
scans your wireless network at
...

Express Burn Plus for Mac

Express Burn Plus for Mac is
an ultra fast burner to help
save you time. Burn audio,
video, or files to CD, DVD.
Express ...

Devart ODBC Driver for
PostgreSQL

Devart ODBC Driver for
PostgreSQL is a
high-performance connectivity
tool with enterprise-level
features for accessing ...

PhotoPad Pro Edition for
Mac

PhotoPad Pro Edition for Mac
Editing Software allows you to
easily edit digital photos and
other pictures. PhotoPad Pro
...

Devart ODBC Driver for
MongoDB

Devart ODBC Driver for MongoDB
is a high-performance
connectivity tool with
enterprise-level features for
accessing ...

BestCrypt Container
Encryption

To protect your privacy,
comply with regulations and
prevent data breaches,
BestCrypt delivers
\'on-the-fly\' data ...

Aml Maple

Aml Maple has a new kind of
view on keyboard layout
indicators: simple, modern,
flexible. In addition, you can
change the ...