Informatics Publishing Limited

M. Sivaram ¹, V. Porkodi ¹, Amin Salih Mohammed ¹, V. Manikandan ¹

Affiliations
1 Department of Information Technology, Lebanese French University, IQ

Abstract

Facial Landmark discovery is an imperative issue in numerous PC vision applications about appearances. It is extremely testing as human faces in wild conditions regularly present expansive varieties fit as a fiddle because of various stances, impediments or demeanors. Profound neural systems have been connected to take in the guide from face pictures to confront shapes. To the best of our insight, Recurrent Neural Network (RNN) has not been utilized in this issue yet. In this paper, we propose a technique which uses RNN and Deep Neural Network (DNN) to take in the face shape. To start with, we design a system utilizing Convolutional Neural Network (CNN) to get the underlying Landmark estimation of appearances. At that point, we utilize feed-forward neural systems for neighborhood look where a segment based seeking technique is investigated. By utilizing LSTM- CNN-RNN, the underlying estimation is more dependable which makes the accompanying segment based pursuit doable and exact. Tests demonstrate that the worldwide system utilizing CNN-LSTM-RNN shows signs of improvement results than past systems in the two recordings and single picture. Our technique beats the cutting edge calculations particularly regarding fine estimation of Landmark spots.

Keywords

Facial landmark, Deep Neural Network, Recurrent Neural Network, Convolutional Neural Network.

Full Text

     

V. Manikandan ¹, V. Porkodi ¹, Amin Salih Mohammed ¹, M. Sivaram ¹

Affiliations
1 Department of Information Technology, Lebanese French University, Erbil, IQ

Abstract

Privacy preserving is critical in the field of where data mining is transformed into cooperative task among individuals. In data mining, clustering algorithms are most skilled and frequently used frameworks. In this paper, we propose a privacy-preserving threshold clustering that uses code based technique with threshold estimation for sharing of secret data in privacy-preserving mechanism. The process includes code based methodology which enables the information to be partitioned into numerous shares and handled independently at various servers. The proposed method takes less number of iterations in comparison with existing methods that does not require any trust among the clients or servers. The paper additionally provides experimental results on security and computational efficiency of proposed method.

Keywords

Privacy Preserving, Data Mining, Threshold Cryptography, Fuzzy C-Means Clustering, Vandermonde Matrix, Secure Multiparty Computation.

Full Text

     

V. Manikandan ¹, V. Prabhu ¹

Affiliations
1 Veltech Multitech Dr. Rangarajan Dr. Sakunthala Engineering College, Chennai, IN

Abstract

Image security forms the major area of concern when dealing with copyright protection and Image authentication. The fundamental operation of a digital camera is image capturing and storing. These images stored in the digital camera are subsequently transmitted in various forms over appropriate media. They are always vulnerable to copyright protection and copyright issues. Every picture appearing in newspapers and magazines has been digitally altered to some degree. This brings about the need to secure, copyright, protect and authenticate digital media. Therefore, this paper comprising of cryptography and watermarking will enhance security and authentication of digital images. This paper deals with the incorporation of watermarking and encryption algorithms in Digital cameras for authentication and security of the images taken from the camera. Watermarking provides authentication yet provides access to the data. Encryption provides security to the image and also restricts access to the data. Hence the host image is safeguarded with two layers of protection.

Full Text