本文關(guān)鍵詞： 全景視頻 圖像拼接 異構(gòu)計算 嵌入式GPU ORB　出處：《計算機(jī)研究與發(fā)展》2017年06期 　論文類型：期刊論文

【摘要】：全景視頻是在同一視點拍攝記錄全方位場景的視頻.隨著虛擬現(xiàn)實(VR)技術(shù)和視頻直播技術(shù)的發(fā)展,全景視頻的采集設(shè)備受到廣泛關(guān)注.然而制作全景視頻要求CPU和GPU都具有很強的處理能力,傳統(tǒng)的全景產(chǎn)品往往依賴于龐大的設(shè)備和后期處理,導(dǎo)致高功耗、低穩(wěn)定性、沒有實時性且不利于信息安全.為了解決這些問題,首先提出了L-ORB特征點提取算法,該算法優(yōu)化了分割視頻圖像的特征檢測區(qū)域以及簡化ORB算法對尺度和旋轉(zhuǎn)不變性的支持;然后利用局部敏感Hash(Multi-Probe LSH)算法對特征點進(jìn)行匹配,用改進(jìn)的樣本一致性(progressive sample consensus,PROSAC)算法消除誤匹配,得到幀圖像拼接映射關(guān)系,并采用多頻帶融合算法消除視頻間的接縫.此外,使用整合了ARM A57CPU和Maxwell GPU的Nvidia Jetson TX1異構(gòu)嵌入式系統(tǒng),利用其Teraflops的浮點計算能力和內(nèi)建的視頻采集、存儲、無線傳輸模塊,實現(xiàn)了多攝像頭視頻信息的實時全景拼接系統(tǒng),有效地利用GPU指令的塊、線程、流并行策略對圖像拼接算法進(jìn)行加速.實驗結(jié)果表明,算法在圖像拼接的特征提取、特征匹配等各個階段均有很好的性能提升,其算法速度是傳統(tǒng)ORB算法的11倍、傳統(tǒng)SIFT算法的639倍;系統(tǒng)較傳統(tǒng)的嵌入式系統(tǒng)性能提升了29倍,但其功耗低至10W.
[Abstract]:Panoramic video is a video recording omnidirectional scene from the same viewpoint. With the development of virtual reality (VR) technology and live video technology, However, CPU and GPU are required to have strong processing capability. Traditional panoramic products often rely on large equipment and post-processing, which leads to high power consumption and low stability. In order to solve these problems, L-Orb feature point extraction algorithm is proposed, which optimizes the feature detection region of segmented video image and simplifies the support of ORB algorithm for scale and rotation invariance. Then the feature points are matched by the locally sensitive Hash(Multi-Probe algorithm, and the mismatch is eliminated by using the improved sample consistency progressive sample consensus algorithm, and the frame image splicing mapping relationship is obtained, and the multi-band fusion algorithm is used to eliminate the seam between videos. Using the heterogeneous embedded system of Nvidia Jetson TX1, which integrates ARM A57 CPU and Maxwell GPU, the real-time panoramic mosaic system of multi-camera video information is realized by using the floating-point computing ability of Teraflops and the built-in video capture, storage and wireless transmission module. The parallel strategy of GPU instruction block, thread and stream is used to accelerate the image mosaic algorithm effectively. The experimental results show that the algorithm has a good performance improvement in every stage of image stitching, such as feature extraction and feature matching. The speed of the algorithm is 11 times that of the traditional ORB algorithm and 639 times that of the traditional SIFT algorithm. The performance of the system is 29 times higher than that of the traditional embedded system, but its power consumption is as low as 10W.
【作者單位】： 武漢理工大學(xué)計算機(jī)科學(xué)與技術(shù)學(xué)院;交通物聯(lián)網(wǎng)技術(shù)湖北省重點實驗室(武漢理工大學(xué));佛羅里達(dá)大學(xué)電氣與計算機(jī)工程系;
【基金】：國家自然科學(xué)基金項目(61303029)~~
【分類號】：TP391.41
，

本文編號：1548695