摘要:移动机器人的研究已经成为当前机器人研究领域的一个热点,在移动机器人研究的相关技术中,正确感知当前环境是顺利完成后续任务的一个重要前提。因此对地表图像进行分类具有非常重要的研究意义。61151
LBP(Local Binary Pattern, 局部二值模式)是一种用来描述图像的局部纹理特征的算子,显然它的作用是进行特征提取,而且提取的特征是图像的纹理特征,而且是局部纹理特征,所以可以用于地表分类算法。LBP算子具有旋转不变性和灰度不变性两个特点,从而在分类过程中减少了光照等外界环境的影响,提高了分类的正确率。
本文将LBP算子用于地表分类。实验表明,基于LBP算子的地表分类是可行的,能够在提高分类器效率并提高识别率。
毕业论文关键词:纹理特征提取 局部二值模式 地表分类
Title: Terrain classification based on Local Binary Pattern
Abstract:Mobile robot has been a hot spot in robotics.In the related technologies of mobile robot,perception of surroundings is the important premise for following tasks.Therefore,the study on terrain classification is great-significant.
LBP(Local Binary Pattern)is a Operator used to describe the local image texture features,Obviously, it is the role for feature extraction, and the extracted feature is the texture features of the image, and the local texture features; classification algorithms can be used for the surface.LBP operator has gray-scale invariance and rotation invariance, resulting in the classification process to reduce the influence of light and other external environment, to improve the classification accuracy.
This article will LBP operator for surface classification. Tests show that the proposed classification of surface LBP operator is viable, it is possible to improve the efficiency of the classifier to improve the recognition rate.
Keywords:Texture feature extraction ,local binary patterns ,Terrain classification
1. 引言(或绪论) 1
1. 1 课题背景及意义 1
1. 3 地表特征识别 2
1. 4 本文的主要工作与结构安排 3
2. LBP特征简介 4
2. 1 纹理概述 4
2. 2 基本的LBP算子 4
2. 3 LBP算子的发展与演化 5
2.4 本章小结 9
3. 地表分类方法 10
3. 1 RGB颜色空间 10
3. 3 纹理特征 11