多重水印快速加密技术在图像深度传感器中的应用

Application of Multi Watermark Fast Encryption Technology in Image Depth Transduce

针对票据中的数字、签名等关键内容容易被篡改的问题,研究了金融电子票据中高效率的快速多重数字水印加密方法。首先,利用卷积神经网络构建图像深度传感器来识别票据中的关键信息区域,以减少水印加密的运算数据量,提高金融票据自动处理效率。针对传统的网络结构易导致过拟合的问题,提出了利用票据图及其差分特征,构建适合CNN网络的多通道图像输入特征,能充分挖掘图像内在联系;进一步改进了传统的CNN网络结构,把所有卷积层的输出连接为一层,构成包含各层信息的融合特征,输入网络的全连接层进行分类识别。实验结果表明,改进后的CNN识别算法,相较传动CNN、DNN等算法,其性能均有明显提升,能够更加高效的进行多个关键区域的内容识别,从而高效的进行多重数字水印的加密,提高金融票据处理的安全性和运算效率。

We study the efficient watermark encryption methods for electronic bills,in order to better protect the key areas in the content.First,the convolutional neural network(CNN)is used to identify key areas in electronic bills,the processed data amount is small.To solve the overfitting problemof convolutional CNN,we first propose the use of image data and their differential features to construct multi-channel image input features suitable for CNN networks,and fully exploit the intrinsic characteristics of the electronic bills.Then,the traditional CNN network structure is improved and the outputs of all convolutional layers are connected as One layer constitutes a fused feature that contains information at each layer,and is input into the network’s fully connected layer for classification and identification.Experiments indicate that the improved CNN recognition algorithm has significantly improved recognition rate compared with CNN and DNN algorithms and leads to more efficient encryption on the key areas.