baggiogoal369的专栏

OTSU最大类间方差法是由日本学者大津展之于1979年提出的，是一种自适应的阈值确定的方法,又叫大津法，简称OTSU。它是按图像的灰度特性,将图像分成背景和目标两部分。背景和目标之间的类间方差越大，说明构成图像的两部分的差别越大,当部分目标错分为背景或部分背景错分为目标都会导致两部分差别变小。因此,使类间方差最大的分割意味着错分概率最小。

图像处理测试 opencv

sivp-0.5.2.zipSIVP - Scilab Image and Video Processing toolbox About SIVP ------------ SIVP stands for Scilab Image and Video Processing toolbox. SIVP intends to do image and video processing tasks. SIVP is meant to be a useful, efficient, and free image and video processing toolbox for Scilab. SIVP homepage: http://sivp.sourceforge.net/ SIVP used some source code of SIP (http://siptoolbox.sourceforge.net/), thanks to Ricardo Fabbri and other SIP developers. Dependencies ------------ Scilab >= 5.2 (http://www.scilab.org/) OpenCV >= 1.0.0 (http://sourceforge.net/projects/opencvlibrary/) (if you want video support, OpenCV should be compiled with ffmpeg) TODO ------------ More image processing functions Bugs ------------ Bugs can be reported at http://sourceforge.net/tracker/?func=add&group_id=134898&atid=731349