本文選題：半脆弱水印 + Contourlet變換��； 參考：《測(cè)繪科學(xué)技術(shù)學(xué)報(bào)》2017年03期

【摘要】：針對(duì)遙感影像易被非法篡改的問題,提出了一種用于篡改定位及近似恢復(fù)的遙感影像半脆弱水印方案。首先將二值偽隨機(jī)矩陣作為認(rèn)證水印,通過量化Contourlet變換各方向子帶中絕對(duì)值最大的系數(shù)將其嵌入,然后采用影像4×4分塊后的灰度平均值作為恢復(fù)水印,將其嵌入到最低有效位。水印檢測(cè)不需要原始遙感影像參與,實(shí)現(xiàn)了盲提取。實(shí)驗(yàn)結(jié)果表明,算法對(duì)于噪聲、JPEG壓縮等合理性失真具有一定的魯棒性,而對(duì)于惡意篡改等攻擊具有脆弱性,能夠?qū)崿F(xiàn)篡改區(qū)域的精確定位和近似恢復(fù),使恢復(fù)后的遙感影像滿足基本的視覺要求,可有效保護(hù)遙感影像的安全。
[Abstract]:Aiming at the problem that remote sensing images are easily tampered with illegally, a semi-fragile watermarking scheme for tampering location and approximate restoration is proposed.Firstly, the binary pseudorandom matrix is used as the authentication watermark, which is embedded by quantizing the absolute maximum absolute coefficients in the subband of the Contourlet transform, and then the average gray level of the image is used as the restoration watermark after 4 脳 4 blocks, and the watermark is embedded into the lowest significant bit.The watermark detection does not need the original remote sensing image to participate, realizes the blind extraction.The experimental results show that the algorithm is robust to reasonable distortion such as noise JPEG compression, but vulnerable to malicious tampering attacks, and can achieve accurate location and approximate restoration of tamper regions.The restored remote sensing images can meet the basic visual requirements and can effectively protect the security of remote sensing images.
【作者單位】： 信息工程大學(xué);
【基金】：地理信息工程國(guó)家重點(diǎn)實(shí)驗(yàn)室開放基金項(xiàng)目(SKLGIE2014-M-4-6)
【分類號(hào)】：TP309.7;TP751
，

本文編號(hào)：1767460