中文摘要：

通过设定图像预测错误门限,并结合支持向量机（Support Vector Machines,SVM）对图像数据进行分类,提出一种二维（2-D）马尔科夫链模型的信息隐藏检测系统。在嵌入率为0.1bpp时,分别应用扩频（spreadspectrum,SS）和量化索引调制（Quantization Index Modulation,QIM）进行实验,系统实现数字水印的正确检测率超过90%;而应用LSB方法,在嵌入率为0.01bpp-0.3bpp条件下,系统实现的数字水印正确检测率在50%~90%以上。

英文摘要：

An information hiding detection method is proposed combining 2-D Markov chain of thresholded prediction-error image and Support Vector Machines as classifier.As embedding data rate being 0.1 bpp（bits per pixel）,experimental investigation utilizing spread spectrum（SS） and a generic QIM method data hiding method respectively,correction detection rates are all above 90%.For generic LSB method,the method achieves a detection rate from 50% to 90% above with 0.01bpp-0.3bpp various embedding data rates.