No abstract

This paper derives a child and adulthood classification technique by integrating the statistical and structural approaches. The structural approaches are derived on a 3 x 3 window based on Local binary pattern (LBP) approach. The proposed approach divides the LBP in to two structural patterns. The present paper derives two distinct patterns called Left Diagonal (LD) and Right Diagonal (RD) LBP's. The given image is converted into binary by comparing the average value of the 3 x 3 neighborhood with its neighbors. Then LD-LBP and RD-LBP codes are evaluated. The range of these code values will be 0 to 2 3 -1, since three pixels form the above proposed patterns. Based on LD and RD-LBP the present paper derived left and right diagonal-GLCM (LRD-GLCM) and features are evaluated.For efficient age classification chi-square distance method is used. To overcome the data dependency problem, the proposed method is implemented on three different facial namely FG-NET, Google and scanned images. The experimental results indicate a significant age classification rate over the existing methods.

To extract local features efficiently Jhanwar et al. proposed Motif Co-occurrence Matrix (MCM) [23] in the literature. The Motifs or Peano Scan Motifs (PSM) is derived only on a 2*2 grid. The PSM are derived by fixing the initial position and this has resulted only six PSM's on the 2*2 grid. This paper extended this approach by deriving Motifs on a 3*3 neighborhood. This paper divided the 3*3 neighborhood into cross and diagonal neighborhoods of 2*2 pixels. And on this cross and diagonal neighborhood complete Motifs are derived. The complete Motifs are different from initial Motifs, where the initial PSM positions are not fixed. This complete Motifs results 24 different Motifs on a 2*2 gird. This paper derived cross diagonal complete Motifs matrix (CD-CMM) that has relative frequencies of cross and diagonal complete Motifs. The GLCM features are derived on cross diagonal complete Motifs texture matrix for efficient face recognition. The proposed CD-CMM is evaluated face recognition rate on four popular face recognition databases and the face recognition rate is compared with other popular local feature based methods. The experimental results indicate the efficacy of the proposed method over the other existing methods.

In this present work a wavelet based watermarking technique is discussed. The proposed method transforms the image into wavelet coefficients by using DWT. A Simplified Significant Wavelet Tree (SSWT) is formed with wavelet coefficients (other than at lowest level) at higher level subband descending towards lower levels. The proposed scheme quantizes the SSWT coefficients to embed a bit of watermark into the frequency part of the image. In the watermark extraction the wavelet trees are formed on the received image to retrieve the watermark bits. The proposed scheme uses adaptive casting energy at different levels and hence achieves high robustness. Various attacks were performed to test the performance and the proposed method has shown high robustness against these attacks.

No abstract