【正文】 編號 ： 本科畢業(yè)設(shè)計(jì)（論文） 題目：（中文） 基于低通濾波的高機(jī)動性視頻目標(biāo)跟蹤 （英文） Enhanced Lowpass Filter Based Vide Predictive Tracking for Target withHigh Mobility 學(xué) 院 專 業(yè) 班 級 學(xué) 號 姓 名 指導(dǎo)教師 職稱 完成日期 寧波大學(xué) 信息 學(xué)院 本科畢業(yè)設(shè)計(jì)（論文） I 誠 信 承 諾 我謹(jǐn)在此承諾：本人所寫的畢業(yè)論文《 基于加強(qiáng)型低通濾波算法的高機(jī)動性視頻目標(biāo)跟蹤 》均系本人獨(dú)立完成，沒有抄襲行為，凡涉及其他作者的觀點(diǎn)和材料，均作了注釋，若有不實(shí)，后果由本人承擔(dān)。 承諾人（簽名）： 年 月 日 基于加強(qiáng)型低通濾波算法的高機(jī)動性視頻目標(biāo)跟蹤 II 摘 要 【 摘要 】 在預(yù)測目標(biāo)移動狀態(tài)過程中，應(yīng)用濾波法是常用的技術(shù)手段。在眾多預(yù)測技術(shù)當(dāng)中，人們常常使用卡爾曼濾波器來跟蹤目標(biāo)在運(yùn)動情況下的軌跡。然而，卡爾曼 濾 波 仍存在一些缺點(diǎn)：用來預(yù)測軌跡時尚缺乏精確度，為了解決這個問題，文中推薦另一種傳統(tǒng)濾波 —— 低通濾波。在文中使用過程中還要考慮目標(biāo)所在的運(yùn)動狀態(tài)?；谶@種情況，則需要在級數(shù)中加入線性項(xiàng)和慣性項(xiàng)算法，這兩種算法分別代 表高機(jī)性和非高機(jī)性兩種狀況。為了證明低通濾波的實(shí)用性，在預(yù)測中融入卡爾曼濾波共同對目標(biāo)檢測跟蹤，實(shí)驗(yàn)表明在預(yù)測軌跡跟蹤質(zhì)量中，所建議的低通濾波對預(yù)測軌跡具有很好的效果比卡爾曼濾波更加有預(yù)測能力，從而證明了它的可行性。 【 關(guān)鍵 詞 】 低通濾波；跟蹤質(zhì)量；卡爾曼濾波；高機(jī)動性 ； 預(yù)測軌跡 寧波大學(xué) 信息 學(xué)院 本科畢業(yè)設(shè)計(jì)（論文） IIIEnhanced Lowpass Filter Based Video Predictive Tracking for Target with High Mobility Abstract 【 ABSTRACT】 In the process of predicting the state of the target mobile , application method of filtering is monly used techniques. Among the many forecasting techniques, people often use the Kalman filter to track the trajectory of the target in the movement. However, Kalman filtering still have some disadvantages: lack of precision for predicting the trajectory when the target is moving, in order to solve this problem, another conventional paper filter is remended lowpass filtering. In the process of using the targets in this paper which should also be considered when the target is in the state of motion. Based on this situation, you need to add linear and inertia algorithms in series, these two algorithms represent a high mobility and nonmobility. To prove the utility of the lowpass filtering, Kalman filter has been added into the join of the forecast for target detection and tracking, trajectory tracking experiments is showing that the quality of the prediction, the proposed lowpass filter to predict the trajectory has a better effect than Kalman filtering predictive ability, thus the proving is feasibility. 【 KEYWORDS】 Lowpass filter。 Tracking quality。 Kalman filter。 High mobility； Predicted trajectory 目 錄 摘 要 .............................................................................................................................................II Enhanced Lowpas s Filter Based Video P redictive Tracking for Target with High Mobility..................................................................................................................................... III Abstract ..................................................................................................................................... III 目 錄 ........................................................................................................................................... III 1 緒論 ........................................................................................................................................ 1 課題的背景 ................................................................................................................ 1 課題研究的意義和目的 .............................................................................................. 1 Matlab 簡介 ............................................................................................................... 2 基于加強(qiáng)型低通濾波算法的高機(jī)動性視頻目標(biāo)跟蹤 IV 論文的研究內(nèi)容 ......................................................................................................... 3 2 高機(jī)動性目標(biāo)識別與跟蹤 ......................................................................................................... 4 低通濾波原理 ............................................................................................................ 4 低通濾波的算法 ......................................................................................................... 4 圖像的處理 ................................................................................................................ 6 圖像分析 ................................................................................................................... 6 低通濾波的預(yù)測跟蹤 .................................................................................................. 7 3 卡爾曼濾波預(yù)測跟蹤 ...............................................................................................................11 卡爾曼濾波的介紹 ....................................................................................................11 卡爾曼濾波預(yù)測原 ......................................................................................