「技術短文」衛星影像雲遮蔽區域之移除與填補演算法

A Cloud Removing Algorithm for Multi-temporal Satellite Images

蔡博閎(Po-Hung Tsai) ； 林昭宏(Chao-Hung Lin)

《航測及遙測學刊》 20卷3期 (2016/05) Pp. 217-229

https://doi.org/10.6574/JPRS.2016.20(3).5

摘要

衛星影像在地球表面觀測相關領域上一直是重要的資料之一，近年來也廣泛的應用於自然科學、環境探測、軍事監控、土地利用與規畫等領域上，然而，衛星影像各波段會受到雲層干擾，使得影像無法完整的呈現地表資訊，因此雲遮蔽是目前衛星影像在使用與分析上的一大問題。本文提出一個衛星影像雲層遮蔽區域移除與填補演算法，對於一張有雲遮蔽的衛星影像，藉由前後相鄰時間的影像資訊進行套合與填補，產生一無雲遮蔽的衛星影像。首先使用監督式分類方法尋找雲遮蔽區，接著以本研究所提出的無縫鑲嵌技術，以不同時期衛星影像資訊對遮蔽區域進行資訊填補，在多時期衛星影像選取方面，藉由影像品質評估方法，選取適合的候選影像以進行資料填補。本文提出解帕森方程式(Poisson equation)的自動化無縫遮蔽區域填補演算法，依影像品質評估填補區域的相似度，以提供後續資料使用上的一項參考指標。本研究的實驗影像為Landsat 7 ETM+衛星影像，實驗過程中，對於各種不同類型的地區進行測試，根據實驗結果顯示，本方法可應用在不同特性之區域，並可成功處理大量的雲層遮蔽之衛星影像。

關鍵字

雲遮蔽區域移除 ； 影像無縫鑲嵌 ； 帕森方程式

並列摘要

Satellite images have a widespread use in many fields such as natural science, environment detection, military monitoring, land use and land planning. However, the cloud cover is a problem in the usage of images acquired by optical sensors. The ground information in the regions covered by cloud will miss. Thus, cloud detection and removing is a fundamental research issue with many applications in the field of geodesic and remote sensing. In this study, a cloud removal approach based on information cloning is introduced. The approach removes cloud-contaminated portions of a satellite image, and then reconstructs the information of missing data utilizing temporal correlation of multi-temporal images. The basic idea is to clone information from cloud-free patches to their corresponding cloud-contaminated patches under the assumption that land covers change insignificantly over a short period of time. The patch-based information reconstruction is mathematically formulated as a Poisson equation and solved using a global optimization process. Thus, the proposed approach can potentially yield better results in terms of radiometric accuracy and consistency compared with related approaches. Some experimental analyses on sequences of images acquired by the Landsat-7 Enhanced Thematic Mapper Plus (ETM+) sensor are conducted. The experimental results show that the proposed approach can process large clouds in a heterogeneous landscape, which is difficult for cloud removal approaches. In addition, quantitative and qualitative analyses on simulated data with different cloud contamination conditions are conducted using quality index and visual inspection, respectively, to evaluate the performance of the proposed approach.

並列關鍵字

cloud removing ； seamless replacement ； Poisson equation