Research on Classification of Open-Pit Mineral Exploiting Information Based on OOB RFE Feature Optimization.

Yi Zhou,Shufang Tian,Yao Liu,Chaozhu Li,Jianping Chen

doi:10.3390/s22051948

Abstract

Mineral exploiting information is an important indicator to reflect regional mineral activities. Accurate extraction of this information is essential to mineral management and environmental protection. In recent years, there are an increasingly large number of pieces of research on land surface information classification by conducting multi-source remote sensing data. However, in order to achieve the best classification result, how to select the optimal feature combination is the key issue. This study creatively combines Out of Bag data with Recursive Feature Elimination (OOB RFE) to optimize the feature combination of the mineral exploiting information of non-metallic building materials in Fujian province, China. We acquired and integrated Ziyuan-1-02D (ZY-1-02D) hyperspectral imagery, landsat-8 multispectral imagery, and Sentinel-1 Synthetic Aperture Radar (SAR) imagery to gain spectrum, heat, polarization, and texture features; also, two machine learning methods were adopted to classify the mineral exploiting information in our study area. After assessment and comparison on accuracy, it proves that the classification generated from our new OOB RFE method, which combine with random forest (RF), can achieve the highest overall accuracy 93.64% (with a kappa coefficient of 0.926). Comparing with Recursive Feature Elimination (RFE) alone, OOB REF can precisely filter the feature combination and lead to optimal result. Under the same feature scheme, RF is effective on classifying the mineral exploiting information of the research field. The feature optimization method and optimal feature combination proposed in our study can provide technical support and theoretical reference for extraction and classification of mineral exploiting information applied in other regions.

Highlights

Introduction published maps and institutional affilThe improper exploitation and utilization of mineral resources may mainly be responsible for the deterioration of ecological environments worldwide
Is the principle of Out of Bagdata data (OOB) Recursive Feature Elimination (RFE): Setting RFE as algorithm framework, we firstly extracted several samples through bootstrap re-sampling of random forest to form OOB; decision tree was constructed for each bootstrap sample to constitute a random forest, which was applied in recursion model for feature importance calculation; In the meantime, we introduced cycle iteration of RFE to evaluate the relevance of features and eliminate the features with relative low importance; After that, we conducted random forest repeatedly to calculate the remaining features until one feature left, and could select the optimal feature combination among all features through comparing their determination coefficient and root-mean-square error
Based on a total of 52 features including spectrum, heat, polarization, and texture features, respectively, extracted from ZY-1-02D, Landsat-8, and Sentinel-1, this study creatively proposed an OOB RFE feature combination optimization method and integrates it with

Summary

Introduction

Introduction published maps and institutional affilThe improper exploitation and utilization of mineral resources may mainly be responsible for the deterioration of ecological environments worldwide. Mineral exploiting information, which generally refers to the scope of the mining area and its status (using/discarded/restored), is a basic indicator to monitor mineral activities and evaluate ecological conditions in a regional mining area [1–3]. Collecting mineral exploiting information timely and accurately can provide a solid database for effective mining activity management and ecological restoration [4]. Comparing with traditional field research, remote sensing has several advantages, like high timeliness, extensive coverage, and immediate results; and because of all the above has been applied in mineral exploitation since 1970s [5,6]. Optical image is the most commonly used data source to obtain mineral exploiting information. The China Geological Survey adopted high-resolution remote sensing images to monitor the status of mining areas and land use from 2003 [7]

Methods

Results

Discussion

Conclusion