| 张永彬.露天灰岩矿纹理特征分析及面向对象的分类[J].地质与勘探,2018,54(2):348-357 |
| 露天灰岩矿纹理特征分析及面向对象的分类 |
| Texture analysis and object-oriented classification of open limestone mines |
| 投稿时间：2017-07-06 修订日期：2017-12-02 |
| DOI：10.12134/j.dzykt.2018.02.013 |
| 中文关键词: 矿产资源开发 遥感 面向对象 决策树 纹理分析 露天灰岩矿 |
| 英文关键词: mine exploitation,remote sensing,object-oriented,decision-making tree,texture analysis,open limestone mines |
| 基金项目:河北省教育厅项目（编号：ZD20131040和YQ2014016）资助 |
|
| 摘要点击次数: 1943 |
| 全文下载次数: 895 |
| 中文摘要: |
| 矿产资源过度开发造成了严重的环境污染，可持续发展面临严峻挑战，矿山治理迫在眉睫。遥感信息提取技术能够便捷地获取矿山开发信息，为矿山治理提供依据。然而，目前信息提取的精度和自动化程度仍有提升空间，并且就矿产类型而言，关于石灰岩矿开采信息提取的研究甚少。针对以上现状，文章以唐山市的一处灰岩矿山为例，以Landsat-8影像作为基础数据，通过计算不同地物的纹理信息，包括地物均值、方差、信息熵、偏离度和数据范围，构造了能够突出露天灰岩矿开采区域的纹理指数模型(texture variance index,简称TVI)。结合已有的光谱和空间特征，设定准确的阈值并建立决策树，以面向对象方法为依托，进行露天灰岩矿山开采边界的信息提取。结果表明，将纹理方差指数作为灰岩矿山开发信息提取的规则之一，总体精度、用户精度和Kappa系数与原有决策树相比皆有所提高。以文中灰岩矿开采范围提取结果为基础，根据高分辨率影像的纹理特征，进一步实现了新矿区与待复垦旧矿区的分类，总体精度约为0.916。 |
| 英文摘要: |
| Excessive exploitation of mineral resources has caused serious environmental pollution. The sustainable development of economy is facing severe challenge. To protect the seedlings is a matter of great urgency and mine management is at all imminent. It is convenient using remote sensing information extraction technique to obtain information of mine development, which can provide basis for mine management. However, there is still room for improvement in terms of extraction accuracy and automaticity. On the part of mineral types, research on the limestone exploitation information extraction remains very limited. In this paper, a limestone mining area in Tangshan is taken as an example and Landsat-8 imagery is used as basic data. By calculating the texture information of different land features, including average grads, variance, entropy, deviation and data range, the texture variance index (TVI) model which can highlight the open limestone mine is constructed. Combined with existing spectral and spatial characteristics, this paper sets up accurate threshold and constructs a decision tree to extract mining boundary information. The results show that the overall accuracy, user precision, and Kappa coefficient are all improved compared to the previous decision tree when using texture variance index as a rule of limestone mining information extraction. Based on the extraction results of limestone mining boundaries and the texture characteristic of high-resolution images, this paper further?makes a classification of new mining and old mining areas, of which the overall accuracy is about 0.916. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
