林業巨量資料應用策略之探討

The Application Strategy for Big Data of Forestry

沈玉婷(Yu-ting Shen)

指導教授 ： 邱祈榮

國立臺灣大學/生物資源暨農學院/森林環境暨資源學研究所/碩士(2015年)

https://doi.org/10.6342/NTU.2015.01525

摘要

巨量資料（Big Data）議題本質上並非一個全新的議題，而是基礎過去近十年間資訊技術的發展，在高度數位化的結果長期累積了大量資料，因為資料量的龐大，使得過去的資料處理技術及應用方式受到挑戰，也使得資訊、知識的擷取有了新的思維。
　　本研究首先透過文獻回顧，針對巨量資料定義、技術發展及應用等面向進行闡述：一般對於巨量資料的討論，其特性主要可從資料數量、資料速度及資料多樣化等幾個面向來看，其實務上的定義則依組織而異，因應使用者需求不同而有所差異。在技術面，其包含了從資料的蒐集、整合、操作分析及視覺化的呈現等，同時也涉及到多元資料來源的異質性資訊整合串聯；而對應技術上的議題，又可依資料特性可分為動態及靜態巨量資料兩種類別，希望能分別針對資訊效率及資訊擷取能力做提升。而另外在應用面，目前台灣仍處於發展初期，未來需能面對使用者的需求，以決策發展為主要方向；應用發展過程，需考慮以下幾個面向，包含是：資料取得、技術應用、資料政策與組織改變及能力等。
　　本研究探討國外林業巨量資料應用案例「全球森林觀測網（GFW）專案」的執行成果，該專案結合了分布世界各地有關組織之資料及研究人力，提供了森林覆蓋、森林變遷、火災監測及林業資源管理分析等各種決策資訊，進行全球森林資源狀態監測，推動資源保育；台灣林業雖長期持續進行森林資源調查工作，但在即時性變遷資訊的掌握能力相較明顯不足。本文嘗試提出國內林業巨量資料應用發展策略，並以衛星影像變遷偵測為個案分析，討論如何透過跨年期時序資料，觀察地區的植生特性及變遷情形。最後再依巨量資料分析過程所需面對的各項議題，提出國內後續應用發展的建議。

關鍵字

巨量資料 ； 林業應用 ； 全球森林觀察 ； 森林監測 ； 變遷偵測

並列摘要

“Big Data” issue is essentially not an “all new” issue. It developed by the basis of information technology over the past decade. Because of the high-level digital strategy, large amounts of data has been collected. It makes the past data processing technology and application mode to be challenged, and also makes a new way of thinking of information and knowledge capture.
　Through this study, we review the property of Big Data: its main characteristics are “data volume”, “data velocity” and “data variety” (3V). Actually the definition varies depending on the needs of users. In the technical part, which contains the process of data collection, integration, analysis and visualization of operational presentation, etc. It also related to the heterogeneity of the sources of information integration. Corresponding the issue of technology, the data characteristics can be classified into dynamic and static two categories, and be expected to enhance the efficiency and information are for information capture capabilities. The other side, in application, it is still in the early stages of development of Big Data currently in Taiwan. There are some main development direction to be considered, including that: data acquisition, technical skill, data policy, and also organizational change and capability.
　This study describes Big Data case of forestry applications: “Global Forest Watch (GFW).” The project combines lots of research organizations around the world. It provides information of forest cover, forest change, fires monitoring and decision making tool of forest resource management, in order to monitoring global forest resources and promote resource conservation. In Taiwan, although long-term survey of forest resources, but the ability of get near-real time change information is still insufficient. In this article, using the satellite images try to identify the temporal-changes, discuss how to observe vegetation growth trends from the process of analysis. Finally, try to make suggestions of the future development of Big Data application in Taiwan.

並列關鍵字

Big Data ； Forestry Application ； Global Forest Watch ； Forest Monitoring ； Change Detection

參考文獻

吳岸明、張桂祥（2014）福衛二號影像處理系統研發之回顧與展望。航測及遙測學刊，18(1):1-12。

曾仁鍵 (2004)衛星影像於大肚山地區植被光譜變遷之監測。國立台灣大學森林環境暨資源學研究所碩士論文。

楊純明（2014）鉅量資料及其在農業生產上之應用潛力。作物、環境與生物資訊，11:51-56。

Kumar, Rakesh, Neha Gupta, Shilpi Charu and Kumar Jangi (2014) Architectural Paradigms of Big Data. National Conference on Innovation in Wireless Communication and Networking Technology, 2014.

Manyika, James, Michael Chui, Brad Brown, Jacques Bughin, Richard Dobbs, Charles Roxburgh Angela Hung Byers& McKinsey Global Institute (2011) Big data: The next frontier for innovation, competition, and productivity.