Hamdy M. Kelash

Faces and Facial Features Detection in Color Images

Human face detection is concerned with finding the location and size of every human face in a given image. Face detection plays a very important role in human computer interaction field. It represents the first step in a fully automatic face recognition, facial features detection, and expression recognition. There are many techniques used in face detection, each one has its advantages and disadvantages. The face detection system presented in this paper is a hybrid of known algorithms. First stage of the proposed method is applying skin detection algorithm to specify all skin locations in the image. Second, extract face features like eyes, mouth and nose. At the last, a verification step is applied to ensure that the extracted features are facial features. In experiments on images having upright frontal faces with any background our system has achieved high detection rates and low false positives

Hiding data in video sequences using steganography algorithms

This paper focuses on the utilization of digital video/images as cover to hide data. The proposed steganography algorithm based on color histograms for data embedding into Video clips directly, where each pixel in each video frame is divided in two parts, the number of bits which will be embedded in the right part are counted in the left part of the pixel. This algorithm is characterized by the ability of hiding larger size of data and the ability of extracting the written text without errors, besides it gives a high level of authentication to guarantee integrity of the video/ images before being extracted. Furthermore, the data were embedded inside the video/ images randomly which gave the video/ images a higher security and resistance against extraction by attackers.

Utilization of Steganographic Techniques in Video Sequences

The fast growth in the demand and consumption of the digital multimedia content in the previous period has led to some effective concerns over issues such as content security, digital rights management, and authenticity. Hiding Data is an important way of recognizing copyright protection for digital multimedia. Digital video/ images became very common choice for hiding data ( hold large amounts of data) and the selection of suitable pixels in the video frames, which are used to store the secret data is very important for an effective and successful embedding process . If pixels are not selected sensibly, undesired spatial and temporal perception problems occur in the stego-video. This paper will concentrate on the use of digital video/images as cover to hide data. The proposed steganography algorithm based on color histograms for data embedding into Video Sequences directly, where each pixel in each video frame is divided in two parts, the number of bits which will be embedded in the right part are counted in the left part of the pixel. This algorithm is characterized by the ability of hiding bigger size of data and the ability of extracting the written text without errors, besides it gives a high level of authentication to guarantee integrity of the video/ images before being extracted. Additionally, the data were embedded inside the video/ images randomly which gave the video/ images a difficult security and resistance against extraction by intruders.

Secure Wireless Image Communication Using LSB Steganography and Chaotic Baker Ciphering

The popularity of wireless multimedia applications in real life and the open nature of wireless media for access make the security of confidential multimedia over wireless channels a major task. This paper proposes a two multi-layer security algorithm using steganography with encryption that strengthens the image security through its transmission over a wireless channel. The steganography conceals the data existence within a cover medium, while the encryption hides the data meaning by using specific algorithms. The suitability of image as a cover medium is due to abundance of redundant bits, and bulk data capacity. The chaos based encryption algorithms are adopted in this paper. In this work a confidential image is embedded into a cover image using least significant bit steganography, and then the output stego image is encrypted using a 2-D Logistic map or 2-D chaotic Baker map. The encrypted stego image is then transmitted over a wireless AWGN channel and the reverse processes are implemented to extract the received secret image. Several experiments have been carried out to examine the suitability of the proposed algorithms for securing high sensitivity images through transmission over wireless channels at different noise levels.

Content Verification of Encrypted Images Transmitted Over Wireless AWGN Channels

In this paper, a content-based image verification scheme is presented. This scheme is suitable for verifying the integrity of images transmitted over insecure networks. The Discrete Cosine Transform is used to embed a block-based mark for each block in another block according to a specific algorithm. To achieve higher protection, the marked image is then encrypted using 2-D chaotic Baker map before being transmitted over the communication channel. The encrypted marked image is transmitted over a wireless Additive White Gaussian Noise (AWGN) channel without error correction codes. At the receiver side, a decryption process is performed, and the marks embedded in each block are extracted to detect suspicious forgery activities. Simulation results show the suitability of the proposed scheme for applications with sensitive data types such as military and nuclear applications.

Enhancement of IR Images using Homomorphic Filtering in Fast Discrete Curvelet Transform (FDCT)

This paper presents an efficient model for enhancement of Infrared images. The proposed method is employing homomorphic filtering in the Fast Discrete Curvelet Transform (FDCT). It based on mixing the advantages of FDCT for representing curves and clarifying features on it with Homomorphic filtering which is an efficient method for enhancing the appearance of IR images. Also, we investigate the influence of FDCT coefficients on the Infrared images. The proposed method gives good results with respect to objective and subjective measurements when using only the first and last FDCT coefficients. General Terms Infrared images Enhancement.

Applying decentralized approach based on mobile agent for improving e-transaction system

In this paper, we propose a solution which can not only satisfy security demands like confidentiality and integrity for data carried during e-transaction, but also solve the bottleneck problem caused by the trusted third party (TTP) due to compromise or unavailability of this party. Applying decentralized approach based on mobile agent (MA) technologies improves e-transaction system by reducing the traffic around administrator management station and alleviates the processing load created by management protocols. Also it improved the performance of the dangerous or real time systems such as a nuclear power plant cooling system and an automated flight control system. However, despite of many advantages of mobile agent, its systems still face a lot of security threats, which restrain the usage of this mechanism. To solve this dilemma we incorporate (RC5) algorithm into our system because of its exceptional simplicity which makes it easy to implement and analyze. Indeed, like the RSA algorithm, the encryption steps of (RC5) can be written on the back of an envelope. (RC5) algorithm is fast, requires low memory and easy to implement. This provides an opportunity for great flexibility in both the performance characteristics and the level of security.

Building a secure e-transaction system based on mobile agent

In this paper, we propose a solution for building a secure e-transaction system based on a combination of theorems like proxy, cross validation and blind signature schemes with mobile agent technology. We make the identity of mobile agent unknown for all the service hosts participating in the e-transaction to ensure the equity of service provision. The proposed solution is designed with the aim of providing: (1) the parity or equity of service provision to all agent hosts. (2) chasing and punish of malicious mobile agents in case of attacking a service host as (3) allows agent host to authenticate or validate the product or the service which he is about to receive is certainly the one he request or pay for, (4) a solution for the problem of conspired truncation attack which the mobile agent faces during its dynamic migration between service hosts. The security properties of the proposed solution are analysed and compared with the most related work.

Comparative Study on Various Random Walk Techniques for Left Ventricle Cavity Segmentation

Objective of image segmentation is to group regions with coherent characteristics. This paper presents a comparative study on Random walk techniques for Left ventricle segmentation. In this paper, LV cavity segmentation is demonstrated for each technique using multi-slice MR cardiac images. The quality of segmentation process is measured by comparing the resulted images of different Random walk techniques using DICE coefficient, PSNR and Hausdorff distance. The comparison includes measuring the execution time of each technique. General Terms Image segmentation, Random walk techniques, LV cavity segmentation

An Enhanced Hierarchical Key Management Scheme for MANETs

Security is one of the major issues in Mobile Ad-hoc Networks (MANETs). Their natural characteristics make them vulnerable to numerous severe attacks. In this paper, we present an enhanced hierarchical key management scheme for secure group communications in MANETs. The proposed approach primarily aims at improving security and complexity, especially complexity in the level of ordinary members for joining or leaving processes to achieve the most possible dynamic characteristic of MANETs without neglecting network security. The proposal scheme is designed with the potential of developing an efficient model interested in keying applied on hierarchical structure to increase the security and make no need to apply rekeying of all group members in different sub-levels as made by hierarchical key management scheme (HKMS). Also, it reduces complexity of processing load, memory usage and improves resources. In the proposed model, each communication between two nodes have a unique key to make it more secure with encrypting messages by different keys for more than one time and support node flexibility. Experiments are carried out on the proposed scheme to show the effect of complexity on each node in different grades and results are compared with traditional HKMS.