目的提出了一种彩色图像数字水印技术,解决数字水印在大数据量水印、不可见性和鲁棒性之间的关系。方法该算法对彩色QR码水印和彩色载体图像分别进行RGB三通道分解,彩色水印单通图像进行混沌加密;对彩色载体单通道图像进行Contourlet变...目的提出了一种彩色图像数字水印技术,解决数字水印在大数据量水印、不可见性和鲁棒性之间的关系。方法该算法对彩色QR码水印和彩色载体图像分别进行RGB三通道分解,彩色水印单通图像进行混沌加密;对彩色载体单通道图像进行Contourlet变换;按对应通道将水印嵌入到载体图像低频经奇异值分解后的对角矩阵。结果该算法在水印嵌入和提取仿真实验结果为PSNR>50 d B;鲁棒性仿真实验结果表明,该算法在抵抗任意旋转攻击和基本图像处理操作的攻击下,提取出来的彩色QR码颜色未失真,并且均能够成功解码和自动识别。结论该算法具有较强的鲁棒性和较好的不可见性,能够很好地将彩色水印嵌入彩色图像,可应用到实际的数字版权保护中。展开更多
临床医生可通过观察眼底视网膜血管及其分支对人体是否患有疾病进行早期诊断,但由于视网膜中的血管错综复杂,模型在分割时会出现对微细血管分割精确度不足的问题。为此,提出一种结合残差模块Res2-net以及高效通道注意力机制(efficient c...临床医生可通过观察眼底视网膜血管及其分支对人体是否患有疾病进行早期诊断,但由于视网膜中的血管错综复杂,模型在分割时会出现对微细血管分割精确度不足的问题。为此,提出一种结合残差模块Res2-net以及高效通道注意力机制(efficient channel attention,ECA)的D-Linknet模型。首先,利用Res2-net代替基础模型中的残差模块Res-net以提升每个网络层的感受野;其次,在Res2-net中添加一种结合压缩激励(squeeze and excitation,SE)和门通道(gated channel transformation,GCT)的注意力机制模块,改善处于复杂背景下的血管分割效果和效率;在网络的解码层加入ECA确保模型计算的性能,避免因降维导致的精度下降;最后,融合改进的模型输出图与掩膜图细化分割结果。在公开数据集DRIVE、STARE上进行分割实验,模型准确度(accuracy,AC)分别为97.11%、96.32%,灵敏度(sensitivity,SE)为84.55%、83.92%,曲线下方范围的面积(area under curve,AUC)为0.9873和0.9766,分割效果优于其他模型。实验证明了算法的可行性,为后续研究提供科学依据。展开更多
为快速生成准确描述图片内容的语句,提出语义分割和卷积神经网络(convolutional neural network,CNN)相结合的图像描述方法。将图像分类模型和语义分割模型结合为编码器,增强对图像语义信息的利用,采用CNN代替长短时记忆网络(long short...为快速生成准确描述图片内容的语句,提出语义分割和卷积神经网络(convolutional neural network,CNN)相结合的图像描述方法。将图像分类模型和语义分割模型结合为编码器,增强对图像语义信息的利用,采用CNN代替长短时记忆网络(long short term memory,LSTM)作为解码器生成完整描述性语句。通过在MSCOCO数据集上与5种主流算法的对比实验可知,以CNN作为解码器能够大幅提高解码速度,语义信息的增强能够有效提高实验精度,验证了该方法的有效性和可行性。展开更多
A new region of interest, （ROI） coding called partial multiply bitplane alternating shift （PMBAShift） is presented in this paper. In PMBAShift, the partial most significant bitplanes of ROI are shifted up and no o...A new region of interest, （ROI） coding called partial multiply bitplane alternating shift （PMBAShift） is presented in this paper. In PMBAShift, the partial most significant bitplanes of ROI are shifted up and no overlap with any bitplane of background （BG）. The most significant bitplanes of BG and general significant bitplanes of ROI coefficients are shifted up by bitplane-by-bitplane alternating sealing. Finally, the least significant bitplanes of ROI and the general significant bitplanes of BG are obtained in the original position and the least significant bitplanes of BG coefficients are shifted down and no overlap with any bitplane. The new method not only retains advantages of maximum shift （Maxshift） method, but also efficiently compresses multiple ROIs according to different degrees of interest without any shape information.展开更多
Image secret sharing(ISS)is gaining popularity due to the importance of digital images and its wide application to cloud-based distributed storage and multiparty secure computing.Shadow image authentication generally ...Image secret sharing(ISS)is gaining popularity due to the importance of digital images and its wide application to cloud-based distributed storage and multiparty secure computing.Shadow image authentication generally includes shadow image detection and identification,and plays an important role in ISS.However,traditional dealer-participatory methods,which suffer from significant pixel expansion or storing auxiliary information,authenticate the shadow image mainly during the decoding phase,also known as unidirectional authentication.The authentication of the shadow image in the distributing(encoding)phase is also important for the participant.In this study,we introduce a public key based bidirectional shadow image authentication method in ISS without pixel expansion for a(k,n)threshold.When the dealer distributes each shadow image to a corresponding participant,the participant can authenticate the received shadow image with his/her private key.In the decoding phase,the dealer can authenticate each received shadow image with a secret key;in addition,the dealer can losslessly decode the secret image with any k or more shadow images.The proposed method is validated using theoretical analyses,illustrations,and comparisons.展开更多
文摘目的提出了一种彩色图像数字水印技术,解决数字水印在大数据量水印、不可见性和鲁棒性之间的关系。方法该算法对彩色QR码水印和彩色载体图像分别进行RGB三通道分解,彩色水印单通图像进行混沌加密;对彩色载体单通道图像进行Contourlet变换;按对应通道将水印嵌入到载体图像低频经奇异值分解后的对角矩阵。结果该算法在水印嵌入和提取仿真实验结果为PSNR>50 d B;鲁棒性仿真实验结果表明,该算法在抵抗任意旋转攻击和基本图像处理操作的攻击下,提取出来的彩色QR码颜色未失真,并且均能够成功解码和自动识别。结论该算法具有较强的鲁棒性和较好的不可见性,能够很好地将彩色水印嵌入彩色图像,可应用到实际的数字版权保护中。
文摘临床医生可通过观察眼底视网膜血管及其分支对人体是否患有疾病进行早期诊断,但由于视网膜中的血管错综复杂,模型在分割时会出现对微细血管分割精确度不足的问题。为此,提出一种结合残差模块Res2-net以及高效通道注意力机制(efficient channel attention,ECA)的D-Linknet模型。首先,利用Res2-net代替基础模型中的残差模块Res-net以提升每个网络层的感受野;其次,在Res2-net中添加一种结合压缩激励(squeeze and excitation,SE)和门通道(gated channel transformation,GCT)的注意力机制模块,改善处于复杂背景下的血管分割效果和效率;在网络的解码层加入ECA确保模型计算的性能,避免因降维导致的精度下降;最后,融合改进的模型输出图与掩膜图细化分割结果。在公开数据集DRIVE、STARE上进行分割实验,模型准确度(accuracy,AC)分别为97.11%、96.32%,灵敏度(sensitivity,SE)为84.55%、83.92%,曲线下方范围的面积(area under curve,AUC)为0.9873和0.9766,分割效果优于其他模型。实验证明了算法的可行性,为后续研究提供科学依据。
文摘为快速生成准确描述图片内容的语句,提出语义分割和卷积神经网络(convolutional neural network,CNN)相结合的图像描述方法。将图像分类模型和语义分割模型结合为编码器,增强对图像语义信息的利用,采用CNN代替长短时记忆网络(long short term memory,LSTM)作为解码器生成完整描述性语句。通过在MSCOCO数据集上与5种主流算法的对比实验可知,以CNN作为解码器能够大幅提高解码速度,语义信息的增强能够有效提高实验精度,验证了该方法的有效性和可行性。
基金This work was supported by the National Natu-ral Science Foundation of China (No. 40372129 and 59638220)the Youth Teacher Foundation of Bei-jng Normal University.
文摘A new region of interest, （ROI） coding called partial multiply bitplane alternating shift （PMBAShift） is presented in this paper. In PMBAShift, the partial most significant bitplanes of ROI are shifted up and no overlap with any bitplane of background （BG）. The most significant bitplanes of BG and general significant bitplanes of ROI coefficients are shifted up by bitplane-by-bitplane alternating sealing. Finally, the least significant bitplanes of ROI and the general significant bitplanes of BG are obtained in the original position and the least significant bitplanes of BG coefficients are shifted down and no overlap with any bitplane. The new method not only retains advantages of maximum shift （Maxshift） method, but also efficiently compresses multiple ROIs according to different degrees of interest without any shape information.
基金Project supported by the National Natural Science Foundation of China(No.62271496)。
文摘Image secret sharing(ISS)is gaining popularity due to the importance of digital images and its wide application to cloud-based distributed storage and multiparty secure computing.Shadow image authentication generally includes shadow image detection and identification,and plays an important role in ISS.However,traditional dealer-participatory methods,which suffer from significant pixel expansion or storing auxiliary information,authenticate the shadow image mainly during the decoding phase,also known as unidirectional authentication.The authentication of the shadow image in the distributing(encoding)phase is also important for the participant.In this study,we introduce a public key based bidirectional shadow image authentication method in ISS without pixel expansion for a(k,n)threshold.When the dealer distributes each shadow image to a corresponding participant,the participant can authenticate the received shadow image with his/her private key.In the decoding phase,the dealer can authenticate each received shadow image with a secret key;in addition,the dealer can losslessly decode the secret image with any k or more shadow images.The proposed method is validated using theoretical analyses,illustrations,and comparisons.
