edges-master
所属分类：图形图像处理
开发工具：matlab
文件大小：22091KB
下载次数：7
上传日期：2018-03-28 09:38:55
上 传 者：
lohaspig
说明： 边缘检测算子
图像边缘信息主要集中在高频段，通常说图像锐化或检测边缘，实质就是高频滤波。我们知道微分运算是求信号的变化率，具有加强高频分量的作用。在空域运算中来说，对图像的锐化就是计算微分。由于数字图像的离散信号，微分运算就变成计算差分或梯度。图像处理中有多种边缘检测（梯度）算子，常用的包括普通一阶差分，Robert算子（交叉差分），Sobel算子等等，是基于寻找梯度强度。拉普拉斯算子（二阶差分）是基于过零点检测。通过计算梯度，设置阀值，得到边缘图像。
(The image edge information is mainly concentrated in the high frequency section, usually the image sharpening or detecting the edge, the essence is the high frequency filtering. We know that differential operation is the change rate of the signal, and it has the function of strengthening the high frequency component. In the space operation, the sharpening of the image is the calculation of differential. Because of the discrete signal of the digital image, the differential operation becomes the difference or the gradient. There are many edge detection (gradient) operators in image processing, which usually include ordinary first order difference, Robert operator (cross difference), Sobel operator and so on, which are based on finding gradient intensity. The Laplasse operator (two order difference) is based on the zero crossing point detection. By calculating the gradient, the threshold is set to get the edge image.)
文件列表:
Contents.m (2055, 2015-02-16)
boxesData.m (2945, 2015-02-16)
boxesEval.m (5118, 2015-02-16)
edgeBoxes.m (4825, 2015-02-16)
edgeBoxesDemo.m (1598, 2015-02-16)
edgeBoxesSweeps.m (3411, 2015-02-16)
edgesChns.m (1581, 2015-02-16)
edgesDemo.m (1201, 2015-02-16)
edgesDemoRgbd.m (1604, 2015-02-16)
edgesDetect.m (2751, 2015-02-16)
edgesEval.m (3283, 2015-02-16)
edgesEvalDir.m (5852, 2015-02-16)
edgesEvalImg.m (3794, 2015-02-16)
edgesEvalPlot.m (2592, 2015-02-16)
edgesSweeps.m (8831, 2015-02-16)
edgesTrain.m (13669, 2015-02-16)
license.txt (6226, 2015-02-16)
models (0, 2015-02-16)
models\forest (0, 2015-02-16)
models\forest\modelBsds.mat (22640640, 2015-02-16)
private (0, 2015-02-16)
private\correspondPixels.mexa64 (112617, 2015-02-16)
private\correspondPixels.mexmaci64 (113624, 2015-02-16)
private\correspondPixels.mexw64 (32256, 2015-02-16)
private\edgeBoxesMex.cpp (18145, 2015-02-16)
private\edgeBoxesMex.mexa64 (54946, 2015-02-16)
private\edgeBoxesMex.mexmaci64 (50408, 2015-02-16)
private\edgeBoxesMex.mexw64 (43008, 2015-02-16)
private\edgesDetectMex.cpp (8942, 2015-02-16)
private\edgesDetectMex.mexa64 (16614, 2015-02-16)
private\edgesDetectMex.mexmaci64 (17604, 2015-02-16)
private\edgesDetectMex.mexw64 (17408, 2015-02-16)
private\edgesNmsMex.cpp (2309, 2015-02-16)
private\edgesNmsMex.mexa64 (9561, 2015-02-16)
private\edgesNmsMex.mexmaci64 (8744, 2015-02-16)
private\edgesNmsMex.mexw64 (8704, 2015-02-16)
private\spDetectMex.cpp (14761, 2015-02-16)
private\spDetectMex.mexa64 (23125, 2015-02-16)
... ...
###################################################################
# #
# Structured Edge Detection Toolbox V3.0 #
# Piotr Dollar (pdollar-at-gmail.com) #
# #
###################################################################
1. Introduction.
Very fast edge detector (up to 60 fps depending on parameter settings) that achieves excellent accuracy. Can serve as input to any vision algorithm requiring high quality edge maps. Toolbox also includes the Edge Boxes object proposal generation method and fast superpixel code.
If you use the Structured Edge Detection Toolbox, we appreciate it if you cite an appropriate subset of the following papers:
@inproceedings{DollarICCV13edges,
author = {Piotr Doll\'ar and C. Lawrence Zitnick},
title = {Structured Forests for Fast Edge Detection},
booktitle = {ICCV},
year = {2013},
}
@article{DollarARXIV14edges,
author = {Piotr Doll\'ar and C. Lawrence Zitnick},
title = {Fast Edge Detection Using Structured Forests},
journal = {ArXiv},
year = {2014},
}
@inproceedings{ZitnickECCV14edgeBoxes,
author = {C. Lawrence Zitnick and Piotr Doll\'ar},
title = {Edge Boxes: Locating Object Proposals from Edges},
booktitle = {ECCV},
year = {2014},
}
###################################################################
2. License.
This code is published under the MSR-LA Full Rights License.
Please read license.txt for more info.
###################################################################
3. Installation.
a) This code is written for the Matlab interpreter (tested with versions R2013a-2013b) and requires the Matlab Image Processing Toolbox.
b) Additionally, Piotr's Matlab Toolbox (version 3.26 or later) is also required. It can be downloaded at:
http://vision.ucsd.edu/~pdollar/toolbox/doc/index.html.
c) Next, please compile mex code from within Matlab (note: win***/linux*** binaries included):
mex private/edgesDetectMex.cpp -outdir private [OMPPARAMS]
mex private/edgesNmsMex.cpp -outdir private [OMPPARAMS]
mex private/spDetectMex.cpp -outdir private [OMPPARAMS]
mex private/edgeBoxesMex.cpp -outdir private
Here [OMPPARAMS] are parameters for OpenMP and are OS and compiler dependent.
Windows: [OMPPARAMS] = '-DUSEOMP' 'OPTIMFLAGS="$OPTIMFLAGS' '/openmp"'
Linux V1: [OMPPARAMS] = '-DUSEOMP' CFLAGS="\$CFLAGS -fopenmp" LDFLAGS="\$LDFLAGS -fopenmp"
Linux V2: [OMPPARAMS] = '-DUSEOMP' CXXFLAGS="\$CXXFLAGS -fopenmp" LDFLAGS="\$LDFLAGS -fopenmp"
To compile without OpenMP simply omit [OMPPARAMS]; note that code will be single threaded in this case.
d) Add edge detection code to Matlab path (change to current directory first):
>> addpath(pwd); savepath;
e) Finally, optionally download the BSDS500 dataset (necessary for training/evaluation):
http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/
After downloading BSR/ should contain BSDS500, bench, and documentation.
f) A fully trained edge model for RGB images is available as part of this release. Additional models are available online, including RGBD/D/RGB models trained on the NYU depth dataset and a larger more accurate BSDS model.
###################################################################
4. Getting Started.
- Make sure to carefully follow the installation instructions above.
- Please see "edgesDemo.m", "edgeBoxesDemo" and "spDemo.m" to run demos and get basic usage information.
- For a detailed list of functionality see "Contents.m".
###################################################################
5. History.
Version NEW
- now hosting on github (https://github.com/pdollar/edges)
- suppress Mac warnings, added Mac binaries
- edgeBoxes: added adaptive nms variant described in arXiv15 paper
Version 3.01 (09/08/2014)
- spAffinities: minor fix (memory initialization)
- edgesDetect: minor fix (multiscale / multiple output case)
Version 3.0 (07/23/2014)
- added Edge Boxes code corresponding to ECCV paper
- added Sticky Superpixels code
- edge detection code unchanged
Version 2.0 (06/20/2014)
- second version corresponding to arXiv paper
- added sharpening option
- added evaluation and visualization code
- added NYUD demo and sweep support
- various tweaks/improvements/optimizations
Version 1.0 (11/12/2013)
- initial version corresponding to ICCV paper
###################################################################
近期下载者：
相关文件：
收藏者：