电子测量技术
電子測量技術
전자측량기술
ELECTRONIC MEASUREMENT TECHNOLOGY
2015年
6期
68-72
,共5页
视频降噪%背景差分%Non-local means%时空联合
視頻降譟%揹景差分%Non-local means%時空聯閤
시빈강조%배경차분%Non-local means%시공연합
Video denoising%Background difference%Non-local means%Spatio-temporal combination
针对视频图像的高斯型随机噪声,提出一种背景提取与前景滤波相结合的时空联合视频降噪算法。结合图像膨胀处理和背景差分法将视频图像分为背景和前景部分,前景部分和背景部分分别采用基于Non‐local means filter的时空联合视频降噪算法和时域平均算法进行降噪处理,并将处理之后的前景和背景相加,得到最终的视频图像序列。最后,给出了Non‐local means filter方法和本文降噪方法降噪效果的对比试验。实验结果表明,Non‐local means filter和本文降噪方法降噪后2个测试序列的PSNR分别为33.0043、29.0365和35.8340、31.5261。这说明对于背景固定的监控类视频,该算法在降低算法复杂度、提高实时性的基础上,有效的处理和保留了视频图像的低频信息和高频细节。
針對視頻圖像的高斯型隨機譟聲,提齣一種揹景提取與前景濾波相結閤的時空聯閤視頻降譟算法。結閤圖像膨脹處理和揹景差分法將視頻圖像分為揹景和前景部分,前景部分和揹景部分分彆採用基于Non‐local means filter的時空聯閤視頻降譟算法和時域平均算法進行降譟處理,併將處理之後的前景和揹景相加,得到最終的視頻圖像序列。最後,給齣瞭Non‐local means filter方法和本文降譟方法降譟效果的對比試驗。實驗結果錶明,Non‐local means filter和本文降譟方法降譟後2箇測試序列的PSNR分彆為33.0043、29.0365和35.8340、31.5261。這說明對于揹景固定的鑑控類視頻,該算法在降低算法複雜度、提高實時性的基礎上,有效的處理和保留瞭視頻圖像的低頻信息和高頻細節。
침대시빈도상적고사형수궤조성,제출일충배경제취여전경려파상결합적시공연합시빈강조산법。결합도상팽창처리화배경차분법장시빈도상분위배경화전경부분,전경부분화배경부분분별채용기우Non‐local means filter적시공연합시빈강조산법화시역평균산법진행강조처리,병장처리지후적전경화배경상가,득도최종적시빈도상서렬。최후,급출료Non‐local means filter방법화본문강조방법강조효과적대비시험。실험결과표명,Non‐local means filter화본문강조방법강조후2개측시서렬적PSNR분별위33.0043、29.0365화35.8340、31.5261。저설명대우배경고정적감공류시빈,해산법재강저산법복잡도、제고실시성적기출상,유효적처리화보류료시빈도상적저빈신식화고빈세절。
To solve the random gauss noise in the video ,this paper proposed a spatio‐temporal combination method with the combing of background extraction and foreground filtering .Video images in this paper are divided into two parts :The foreground and the background ,using background difference method combined with image expanding .Then deal with the two parts using spatio‐temporal combination method based on the Non‐local means filter and direct meaning in time domain .And then the output video can be got by adding the foreground and the background which are proposed . At last ,this paper give the comparison of the Non‐local means filter and the method in this paper .Experimental results indicate that the PSNR of the tow test video after denoising reaches 33 .0043、29 .0365 and 35 .8340、31 .5261 . Experiment shows that for the monitoring video with fixed background ,the arithmetic in this paper effectively deal with the low‐frequency information and the high‐frequency details ,on the basis of reducing the complexity and improving the real‐time ability of the method .