کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
1180139 1491522 2016 12 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Neighborhood mutual information and its application on hyperspectral band selection for classification
موضوعات مرتبط
مهندسی و علوم پایه شیمی شیمی آنالیزی یا شیمی تجزیه
پیش نمایش صفحه اول مقاله
Neighborhood mutual information and its application on hyperspectral band selection for classification
چکیده انگلیسی


• Combine information entropy with neighborhood rough set and propose a new measure.
• A band selection method based on neighborhood mutual information was proposed.
• The algorithm was applied to three (soybean, maize and rice) hyperspectral datasets.
• Proposed algorithm was compared to neighborhood dependency measure based algorithm, GA and UVE algorithm.
• Two classification models (Extreme Learning Machine and Random Forests) were built.

Band selection is considered to be an important processing step in handling hyperspectral data. In this work, we combined Shannon's information entropy with neighborhood rough set and proposed a new measure, called neighborhood mutual information. With the proposed measure which can evaluate the significance of bands for classification, a forward greedy search algorithm for band selection was constructed. To assess the effectiveness of the proposed band selection technique, two classification models (Extreme Learning Machine and Random Forests) were built. The proposed algorithm was compared to neighborhood dependency measure based algorithm, genetic algorithm and uninformative variable elimination algorithm on three (soybean, maize and rice) hyperspectral datasets between 400 nm and 1000 nm wavelengths. Experimental results show that the proposed method can effectively select key bands and obtain satisfactory classification accuracy.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Chemometrics and Intelligent Laboratory Systems - Volume 157, 15 October 2016, Pages 140–151
نویسندگان
, , , , , ,