【摘要】：地質(zhì)填圖是指對(duì)特定區(qū)域的地質(zhì)年代、地層、構(gòu)造、礦產(chǎn)資源進(jìn)行定義、描述、勘查的過程,其比例尺的選取可以根據(jù)實(shí)際情況選定,最常用的是1:500-1:50000。 傳統(tǒng)地質(zhì)工作對(duì)某些區(qū)域(如我國喜馬拉雅山脈)進(jìn)行實(shí)地勘查時(shí),由于受這些地區(qū)地理環(huán)境險(xiǎn)峻、氣候惡劣等因素影響,會(huì)存在勘查盲區(qū)。傳統(tǒng)地質(zhì)填圖方式中,地質(zhì)工作人員通常選取特殊地質(zhì)點(diǎn)進(jìn)行填圖,如采用1:200000比例尺進(jìn)行填圖時(shí),每兩平方公里的范圍內(nèi)選取一個(gè)地質(zhì)點(diǎn)進(jìn)行觀測(cè),這就存在盲區(qū),無法保證填圖的準(zhǔn)確性。用人工方式填充一幅1:200000地質(zhì)圖時(shí),需一個(gè)地質(zhì)大隊(duì)(200-300人)工作兩年時(shí)間,耗費(fèi)巨大。 數(shù)字地質(zhì)填圖方式,借助遙感數(shù)字圖像對(duì)整個(gè)區(qū)域進(jìn)行全面詳細(xì)、無盲區(qū)地觀察,有利于地質(zhì)工作人員從時(shí)間和空間上對(duì)整個(gè)地質(zhì)填圖區(qū)域有一個(gè)宏觀的認(rèn)識(shí)和把握,使得地質(zhì)填圖成果更加貼近客觀實(shí)際。 近年來,隨著遙感數(shù)字圖像處理技術(shù)在地質(zhì)填圖中不斷發(fā)展進(jìn)步,地質(zhì)工作者越來越意識(shí)到其重要性和不可替代性。目前,遙感數(shù)字圖像經(jīng)過相關(guān)處理方法變換后作為傳統(tǒng)方式地質(zhì)填圖時(shí)的參考。采用遙感數(shù)字圖像處理技術(shù)進(jìn)行地質(zhì)填圖過程中存在許多問題。例如,在進(jìn)行遙感數(shù)字圖像預(yù)處理時(shí),存在像素點(diǎn)亮度值丟失、彩色合成圖像顏色偏藍(lán)、對(duì)比度不適中、不利于目視判讀等問題,導(dǎo)致圖像失真。由于不同衛(wèi)星的不同圖像采集設(shè)備對(duì)圖像中有效像素的影響是一致的。例如,對(duì)不同波段的圖像,如果偏強(qiáng),則所有有效像素全部偏強(qiáng)；如果偏弱,則所有有效像素全部偏弱。商業(yè)處理軟件處理遙感數(shù)字圖像后,新生成的圖像異常亮點(diǎn)過多,對(duì)比度過強(qiáng),這些都影響到遙感數(shù)字圖像原始信息的提取,最終影響到地質(zhì)填圖的實(shí)施和可信度。 目前,所采用的各種遙感數(shù)字圖像對(duì)比度增強(qiáng)和彩色合成增強(qiáng)方法,都存在不能有效解決像素點(diǎn)亮度值經(jīng)過增強(qiáng)后損失的問題。如何將遙感數(shù)字圖像中含有的原始數(shù)據(jù)信息完整無損的提取出來,正是我們要考慮和解決的問題。這直接關(guān)系到主成分分析、小波變換等后續(xù)工作的順利展開,從而為地質(zhì)填圖打下堅(jiān)實(shí)的基礎(chǔ)。 本研究以無損平移雙向?qū)Ρ榷壤焖惴槔碚摶A(chǔ),嘗試提出一種盡量減少像素點(diǎn)亮度值丟失的改進(jìn)算法,為遙感數(shù)字圖像處理后續(xù)工作的展開提供堅(jiān)實(shí)基礎(chǔ)。采用遙感數(shù)字圖像處理技術(shù)直接進(jìn)行地質(zhì)填圖是世界性難題,我們選取浙江省東部沿海區(qū)域中生代酸性侵入巖為研究對(duì)象,首次嘗試?yán)眠b感數(shù)字圖像處理技術(shù)進(jìn)行直接地質(zhì)填圖。
[Abstract]:Geological mapping refers to the process of defining, describing and exploring the geological age, strata, structure and mineral resources of a particular area. The selection of its scale can be selected according to the actual situation, and the most commonly used is 1: 500-1: 50000. During the field exploration of some areas (such as Himalayan Mountains in China) in the traditional geological work, there will be blind exploration areas due to the influence of such factors as the harsh geographical environment and the bad climate in these areas. In the traditional geological mapping method, geological workers usually select special geological sites for mapping. For example, when mapping on a scale of 1: 200000, one geological spot is selected for observation every two square kilometers, and there are blind areas. There is no guarantee of the accuracy of the mapping. Filling a 1: 200000 geological map manually requires two years of work by a geological unit (200-300 people), which is costly. Digital geological mapping, with remote sensing digital images as a means of comprehensive and detailed observation of the whole region without blind area, is conducive to the geological staff to have a macroscopic understanding and grasp of the whole geological mapping area in time and space. The geological mapping results are closer to the objective reality. In recent years, with the development of remote sensing digital image processing technology in geological mapping, geologists are increasingly aware of its importance and irreplaceable. At present, remote sensing digital image is transformed by correlation processing method as the reference of traditional geological mapping. There are many problems in geological mapping using remote sensing digital image processing technology. For example, in the process of remote sensing digital image preprocessing, there are some problems, such as loss of pixel brightness, blue color of color composite image and unsuitable contrast, which is not conducive to visual interpretation, which leads to image distortion. The effects of different image acquisition devices on the effective pixels in the images are consistent. For example, for images of different bands, all effective pixels are strong if they are strong, and all effective pixels are weak if they are weak. After processing remote sensing digital image by commercial processing software, the newly generated image has too many bright spots and strong contrast, which affects the extraction of original information of remote sensing digital image, and finally affects the implementation and credibility of geological mapping. At present, all kinds of remote sensing digital image contrast enhancement and color combination enhancement methods can not effectively solve the loss of pixel brightness value after enhancement. How to extract the original data information from remote sensing digital image is the problem that we should consider and solve. This is directly related to the smooth development of principal component analysis (PCA) and wavelet transform (WT), thus laying a solid foundation for geological mapping. Based on the theory of lossless translation bidirectional contrast stretching algorithm, an improved algorithm is proposed to minimize the loss of pixel brightness value, which provides a solid foundation for the following work of remote sensing digital image processing. It is a worldwide problem to use remote sensing digital image processing technology to directly map geology. We choose Mesozoic acid intrusive rocks in the eastern coastal area of Zhejiang Province as the research object. It is the first time to use remote sensing digital image processing technology for direct geological mapping.
【學(xué)位授予單位】：東北師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：P623.6;TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 施俊法;浙江省諸暨地區(qū)元素地球化學(xué)分布與標(biāo)度律[J];地球科學(xué);2001年02期

2 于慶文,李超嶺,張克信,楊東來,朱云海,葛夢(mèng)春;數(shù)字地質(zhì)填圖研究現(xiàn)狀與發(fā)展趨勢(shì)[J];地球科學(xué);2003年04期

3 周國華,吳小勇,周建華;浙北地區(qū)土壤元素有效量及其影響因素研究[J];第四紀(jì)研究;2005年03期

4 薛臘梅;趙希濤;張耀玲;高萬里;;遙感技術(shù)在東昆侖新生代地質(zhì)填圖中的應(yīng)用[J];地質(zhì)力學(xué)學(xué)報(bào);2010年01期

5 鄒長毅;劉應(yīng)漢;邵躍;李應(yīng)桂;;銅巖山地區(qū)原生地球化學(xué)特征及找礦評(píng)價(jià)[J];地質(zhì)找礦論叢;2006年02期

6 李春海;吳昌志;郭坤一;包超民;邢光福;;浙江五畝山地區(qū)鎢礦容礦構(gòu)造及找礦方向[J];地質(zhì)找礦論叢;2008年01期

7 武文波;康停軍;姚靜;;基于IHS變換和主成分變換的遙感影像融合[J];遼寧工程技術(shù)大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年01期

8 周軍,高鵬,田勤虎,劉磊,李得成;新疆巴里坤ETM數(shù)據(jù)遙感地質(zhì)填圖的探索[J];國土資源遙感;2005年03期

9 王欽軍;藺啟忠;黎明曉;魏永明;王黎明;;一種突出構(gòu)造信息的多光譜遙感變換方法[J];光譜學(xué)與光譜分析;2009年07期

10 韓春成;林和平;周濤;趙哲;;遙感數(shù)字圖像增強(qiáng)方法研究[J];長春師范學(xué)院學(xué)報(bào);2012年12期

，

本文編號(hào)：2419809