本文選題：視頻穩(wěn)像　切入點(diǎn)：SURF算法　出處：《電視技術(shù)》2017年06期 　論文類型：期刊論文

【摘要】：針對(duì)老電影的幀間抖動(dòng)現(xiàn)象,研究一種基于SURF的全局抖動(dòng)修復(fù)方法。該方法首先依據(jù)老電影抖動(dòng)的特點(diǎn)建立對(duì)應(yīng)的數(shù)學(xué)模型,然后運(yùn)用加速健壯特征(SURF)算法進(jìn)行特征點(diǎn)匹配,并針對(duì)攝像機(jī)是否存在主觀運(yùn)動(dòng)提出相應(yīng)的修復(fù)方案,同時(shí)運(yùn)用運(yùn)動(dòng)閾值策略去除誤匹配,最后計(jì)算位移均值作為全局運(yùn)動(dòng)參數(shù),通過反向平移運(yùn)動(dòng)參數(shù)完成抖動(dòng)修復(fù)。在VS2013中使用Open CV實(shí)現(xiàn)了相關(guān)算法,并詳細(xì)介紹了系統(tǒng)的功能和框架。實(shí)驗(yàn)給出的電影去抖動(dòng)效果證明了該方法的有效性。
[Abstract]:Aiming at the jitter between frames of the old movies, a global dithering repair method based on SURF is studied. Firstly, the corresponding mathematical model is established according to the characteristics of the jitter of the old movies, and then the feature points are matched by the accelerated robust feature surf algorithm. At the same time, the motion threshold strategy is used to remove the mismatch, and the displacement mean value is calculated as the global motion parameter. The relative algorithm is implemented by using Open CV in VS2013, and the function and framework of the system are introduced in detail. The experimental results show the effectiveness of the method.
【作者單位】： 上海大學(xué)上海電影學(xué)院;泰山學(xué)院文學(xué)與傳媒學(xué)院;
【基金】：國家自然科學(xué)基金項(xiàng)目(61402278) 上海市科委科技攻關(guān)項(xiàng)目(14511108400) 教育部人文社會(huì)科學(xué)項(xiàng)目(16YJC760056) 上海大學(xué)電影學(xué)高峰學(xué)科項(xiàng)目(2016GF001) 山東省社會(huì)科學(xué)規(guī)劃研究項(xiàng)目(15DWYJ06)
【分類號(hào)】：J931;TP391.41
，

本文編號(hào)：1641403