tvkpz / MCITF

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

MCITF

source code for our paper Saliency detection for strip steel surface defects using multiple constraints and improved texture features by Guorong Song, Kechen Song and Yunhui Yan.

The code is run on Matlab 2018a on Windows 10/64 bit

Lower version of Matlab may cause several errors

Usage:

  1. put the test images into file '\Image'
  2. run 'demo_MCITF.m'
  3. the output saliency maps are stored in the file '\OUT'

Note:

  1. The file 'Results' contains the experimental results of all the methods presented in our paper.
  2. The file 'SD-saliency-900' contains the source images of strip steel and the corresponding the pixel-wise binary maps (ground truth).

About


Languages

Language:MATLAB 46.7%Language:HTML 36.7%Language:C++ 12.8%Language:C 3.7%Language:Makefile 0.1%Language:CSS 0.1%