基于機器視覺的目標檢測與跟蹤研究

以下是資料介紹,如需要完整的請充值下載. 本資料已審核過,確保內容和網頁里介紹一致.  
無需注冊登錄,支付后按照提示操作即可獲取該資料.
資料介紹:

基于機器視覺的目標檢測與跟蹤研究(任務書,開題報告,外文翻譯,論文15000字)
摘要
隨著計算機技術的飛速發展,目標檢測與跟蹤技術有了很大的提高,它已經成為智能監控、智能交通等領域的重要構成部分。該技術對于行車安全和行人安全有著非常重要的意義,特別是行人的檢測與跟蹤。在我們生活的城市里,交通環境是比較復雜的。
本文首先研究了運動目標檢測,當背景是靜態時,其方法主要是背景減除法、三幀差分法、光流法,然后分析了各自的特點與不足。當背景是動態時,該部分是本文的核心部分之一,提出了改進的灰度投影算法,能夠估計運動背景的平移,縮放等參數,并且準確地在動態背景下對運動目標檢測其次是人形目標識別,在完成運動目標檢測的基礎上,通過寬高比,周長,矩形度等圖形特征識別運動目標中的人形目標。最后是運動目標跟蹤,提出了基于多特征與Kalman濾波融合的Meanshift跟蹤算法,能夠對多個目標進行快速有效地跟蹤,同樣是本文核心部分。通過實驗可以證明本文所提出的算法無論在靜態還是動態的背景下,都能完成運動目標準確實時地檢測跟蹤。
關鍵詞:檢測,跟蹤,動態背景,行人監控
Abstract
With the fast advance of computer technology, target detection and tracking technology has been greatly improved. It has become an important part of intelligent monitoring, intelligent transportation and other fields. This technology is very important for traffic safety and pedestrian safety, especially for pedestrian detection and tracking. In the cities where we live, the traffic environment is relatively complex.
Firstly, this paper studies the detection of moving objects. When the background is static, the main methods are background subtraction, three frame difference and optical flow. Then, the characteristics and shortcomings of each method are analyzed. At that time, this part is one of the core parts of this paper. An improved gray projection algorithm is proposed, which can estimate the parameters of moving background such as translation and scaling, and accurately detect moving target in dynamic background, followed by human object recognition. When it is completing moving target detection, the width-height ratio, circumference, rectangularity and other graphical features are adopted. Feature recognition of human-shaped objects in moving targets. Finally, we propose algorithms to track shifting based on motion tracking target for several properties and fusion to wait. The Kalman, which can track multiple targets quickly and efficiently. It is also the core part of this paper. Experiments show that the proposed algorithm can detect and track moving objects accurately and real-time in both static and dynamic environments.
Key words: detection, tracking, dynamic background, pedestrian monitoring
 

基于機器視覺的目標檢測與跟蹤研究
基于機器視覺的目標檢測與跟蹤研究


目錄
第1章 緒論    1
1.1 課題的研究背景和意義    1
1.2關鍵技術發展現狀    1
1.2.1目標檢測技術的發展現狀    1
1.2.2目標跟蹤技術的發展現狀    2
1.2.3行人檢測與跟蹤技術的發展現狀    3
1.3論文的章節安排    4
1.4本章小結    4
第2章 目標檢測研究    6
2.1基本檢測方法    6
2.1.1背景減除法    6
2.1.2幀差法    6
2.1.3光流法    6
2.2動態背景下運動目標檢測    7
2.2.1背景運動補償    7
2.2.2改進的三幀差分法    8
2.3實驗結果分析    9
2.4本章小結    10
第3章人形目標識別    11
3.1人形目標特征選擇    11
3.2人形目標特征提取    12
3.2.1確定目標邊界    12
3.2.2提取目標的周長    12
3.2.3計算目標的長寬比    13
3.2.4計算目標的矩形度    13
3.3運動行人識別算法    13
3.4本章小結    14
第4章目標跟蹤研究    15
4.1基于多特征與Kalman濾波結合的Meanshift算法    15
4.1.1Meanshift算法的基本原理    15
4.1.2多特征提取    15
4.1.3Meanshift跟蹤算法    17
4.1.4 Kalman濾波    18
4.1.5多特征與Kalman濾波結合的Meanshift算法    19
4.2實驗結果分析    21
4.3本章小結    22
第5章總結與展望    23
5.1總結    23
5.2展望    23
參考文獻    24
致謝    25

302宿舍白露全文阅读