# 臺灣博碩士論文加值系統

(3.231.230.177) 您好！臺灣時間：2021/07/27 14:11

:::

### 詳目顯示

:

• 被引用:0
• 點閱:210
• 評分:
• 下載:11
• 書目收藏:2
 視覺監視系統之主要目的，在於偵測週遭範圍是否有可疑人物入侵或是環境有無發生異常變化；而移動物體偵測通常是視覺監視系統的第一步驟，後續的目標物追蹤或物件分類等功能，都需依賴移動物體偵測的準確與否；因此，如何從複雜的環境中，更準確且快速地偵測到移動物體為一重要課題。背景相減法為移動物偵測常用的一種方法，其偵測結果的好壞通常取決於系統是否能將環境的變化，即時且適當地更新至背景模型當中；因此，如何在變動的環境下，建立適當的背景模型是本論文探討的主軸。近年來利用統計式方法來建立背景模型的研究越來越受到重視，一般而言，統計式背景模型之建立主要分成兩種方法，一種為有母數法，包括高斯混合模型、空間高斯分佈等；另一種為無母數法，包括核心估測法、K鄰近法等。本論文針對高斯混合模型與核心估測兩種統計式背景模型的建立方式，探討其適用的場合及實際應用上的優缺點，最後以實驗驗證本論文所提出之觀點。
 The main purpose of a visual surveillance system is to detect suspicious objects or erratic changes in the environment. In a visual surveillance system, object tracking and object classification rely on the accuracy of motion detection. Therefore, it is essential to quickly and accurately identify a moving object in an intricate environment. Background subtraction is commonly used in motion detection. The system must update the alteration of the environment into the background model quickly and accurately for best performance. A suitable background model for changing environments was developed in this thesis. In general, there are two methods of constructing a statistical background model. One is the parametric method, which includes the Gaussian mixture model and the spatial distribution of Gaussian. The other is the nonparametric method, which includes the kernel density estimation and the k-nearest neighbors method. In this thesis, several experiments have been conducted to provide a performance comparison between the Gaussian mixture model and kernel density estimation.
 摘要.....................................................................I誌謝...................................................................III目錄...................................................................IV圖目錄...............................................................VI表目錄.............................................................VIII第 1 章 緒論........................................................11.1 前言...............................................................11.2 研究方法與動機...........................................21.3 文獻回顧.......................................................61.4 論文架構.......................................................8第 2 章 高斯混合模型之背景相減法..............102.1 簡介.............................................................102.2 高斯混合模型簡介.....................................122.3 高斯機率密度函數的初始參數估測.........152.4 灰階值的歸類.............................................182.5 背景模型建立與前景偵測.........................212.6 背景模型參數更新.....................................23第 3 章 無母數模型之背景相減法..................303.1 簡介.............................................................303.2 核心密度估測法 ........................................323.3 核心寬度的選擇 ........................................363.4 背景模型建立與前景偵測.........................413.5 錯誤偵測.....................................................433.6 背景更新.....................................................47第 4 章 實驗結果與討論..................................524.1 實驗介紹.....................................................524.2 實驗結果.....................................................534.2.1 GMM法與KDE法之比較.........................534.2.2 實驗場景一：無光源變化之場景..........634.2.3 實驗場景二：有光源變化之場景..........664.2.4 實驗場景三：大雨環境下之道路場景..694.3 實驗討論.....................................................72第 5 章 結論與建議..........................................745.1 結論.............................................................745.2 未來展望與建議.........................................75參考文獻...........................................................76自 述..................................................................82