Order Statistics Filters in Wavelet Domain for Color Image Processing

A new steganographic method for RGB color images using estimation of variance field in the wavelet domain

“…The criteria used to compare different methods were the PSNR (Peak Signal to Noise Ratio), MAE (Mean Absolute Error), Q (Quality index), NCD (Normalized Color Deviation), and HC (Hiding Capacity). The image quality measures (IQM) steganalysis technique was used to measure the robust capabilities of various steganographic algorithms [3,8]. Table 1 presents the performance results of various methods in the 320x320 color images Airplane, House, Lena, Mandrill, Peppers, and Sailboat used as host and secret images.…”

“…The proposed method works in the wavelet domain and provides an analysis of the wavelet coefficients in different decomposition scales to estimate the variance field to distinguish areas where the pixels are considered noisy [3]. The proposed wavelet steganographic method is depicted in Figure 1.…”

“…It has been proven that the stego-images which are achieved by hiding secret data in host images with more variation in their pixel intensities are less detectable by the statistical examinations of steganalyzers increasing the embedding capacity [4]. The redundancy of approaches algorithm uses one level of decomposition with Haar wavelet and it smoothes low frequencies of host image through of a double convolution operation (first of decomposition, and after reconstruction) of the coefficients and samples of host image [3], providing more hiding capacity and detail preservation of proposed method.…”

“…Otherwise, this area is considered free noise and it is not possible to insert data, 2 is a weighting function. This criterion provides good invisibility of hiding data and edges and fine detail preservation of the stego-image [3]. To recover the hidden image, the algorithm of Figure 1 is used again but in the block of input changes to RGB stego-image and in the condition T k   recovers the hidden data.…”