Outdoor Navigating Scene Labeling Using Importance Factor Based I-RELIEF and Feature Weighted Support Vector Machines
Feature selection plays an important role in terrain classification for outdoor agricultural robot navigation. For terrain classification, the image data usually has a large number of feature dimensions. The better selection of features usually results in higher labeling accuracy. In this work, a novel approach for terrain perception using Importance Factor based I-Relief algorithm and Feature Weighted Support Vector Machines (IFIR-FWSVM) was put forward. Firstly, the weight of each feature for classification was computed by using Importance Factor based I-Relief algorithm (IFIR) and the irrelevant features was eliminated. Then the weighted features were used to compute the kernel functions of SVM and trained the classifier. Finally, the trained SVM was employed to predict the terrain label in the far-field regions. The novelty of this paper was that feature selection was considered for terrain classification and a modified version of I-Relief named IFIR was designed for superpixel-level classification. Experimental results based on DARPA datasets showed that the proposed method IFIR-FWSVM was superior over traditional SVM.
terrain classification robot navigation feature selection machine vision support vector machines (SVM)
Jun Tu Chengliang Liu Mingjun Wang Zhonghua Miao
School of Mechanical Engineering Shanghai Jiaotong University Shanghai, 200240, China College of Mechanical Engineering Ningbo University of Technology Ningbo, 315211, China Mechatronics Engineering and Automation Shanghai University Shanghai, 200072, China
国际会议
哈尔滨
英文
977-981
2012-06-16（万方平台首次上网日期，不代表论文的发表时间）