论文部分内容阅读
在巴西的中南部,遥感技术用于矿产勘查由于出露的基岩缺乏而受到限制。该区遭受过很深的风化作用,从而产生了很厚的残积土层,这些土壤上长有自然森林的植物群、开阔的灌木丛、咖啡种植园及供放牧的草。数字图像增强处理已经用于识别从镁铁质和超镁铁质变质火山岩派生的富铁土壤,并将其与混合片麻岩上发育的土壤区别开,确定从风化的硫化物矿体派生的土壤中含异常铁富集的地区。该项研究基于一个15×15km子区,区内有经改造的太古代变质科马提岩/玄武岩中的奥托勒镍硫化物矿床。将与矿床有关的富褐铁矿土壤作为“训练”场,以便对TM图像中的铁氧化物矿物进行波谱增强。已开发了一种图像处理技术,称为“特征指定主组分选择”(FPCS),这种技术有助于识别和选择数据库中的特殊主组分,以增强一定的波谱标记。在试验区,用FPCS综合4组分的彩色显示技术取得了最佳结果,用平衡扩展波段差值及去相关扩展也取得了好结果。精选出的增强技术可用于巴西中南部的一个更大的潜在重要地区,该区与试验区具有相似的地质、气候和土壤条件。
In south-central Brazil, the use of remote sensing for mineral exploration is limited by the lack of exposed bedrock. The area is heavily weathered and produces very thick residual soil layers of flora with natural forests, open shrubs, coffee plantations and grazing grasses. Digital image enhancement processes have been used to identify iron-rich soils derived from mafic and ultramafic metamorphic volcanic rocks and to distinguish them from soils developed on mixed gneiss to identify soils derived from weathered sulfide ore bodies In areas containing abnormal iron enrichment. The study is based on a 15 × 15km subregion within the remodeled Archean metamorphic kamamatite / basalt Ottoal nickel sulfide deposit. The limonite-rich soil associated with the deposit is used as a “training” field for spectral enhancement of iron oxide minerals in TM images. An image processing technique has been developed, called Feature-Specific Principal Component Selection (FPCS), which helps to identify and select specific principal components in a database to enhance certain spectral markers. In the experimental area, the best results were obtained with the FPCS integrated 4-component color display technique, and good results were obtained with balanced extended band differences and decorrelated extensions. The selected enhanced technology is available for a larger, potentially important area in south-central Brazil that has similar geological, climatic and soil conditions to the experimental area.