Akram Elkefi

Blind watermarking algorithm based on spiral scanning method and error-correcting codes

This work is a connecting link between the field of digital transmission and (3 Dimension) 3D watermarking. In fact, we propose in this paper a blind and robust watermarking algorithm for 3D multiresolution meshes. This data type, before being watermarked, is divided into GOTs (Group Of Triangles) using a spiral scanning method. At every instant, only one GOT is loaded into memory. It undergoes a wavelet transform. Embedding modifies the wavelet coefficients vector thus generated after being presented in a cylindrical coordinate system. After being watermarked, the current GOT will be released from memory to upload the next GOT. Information is coded using a turbo encoder to generate the codeword to be inserted. Once the entire mesh is scanned, the watermarked mesh is reconstructed. During extraction, the same steps are applied only on the watermarked mesh: our algorithm is then blind. Extracted data are decoded using Error-Correcting Code (turbocode) to correct errors that occurred. The results show that our algorithm preserves mesh quality even with a very large insertion rate while significantly minimizing used memory. Data extraction was done correctly despite the application of various attacks. Our algorithm is robust against most popular attacks such as similarity transformation, noise addition, smoothing, coordinate quantization, simplification and compression.

A Trust Region Optimization Method for Fast 3D Spherical Configuration in Morphing Processes

This paper addressed the problem of Spherical Mesh parameterization. The main contribution of this work was to propose an effective optimization scheme to compute such parameterization, and to have an algorithm exposing a property of global convergence This is the case of trust region spherical parameterization TRSP to minimizing the ratio of inverted triangle, have an efficient spherical parameterization, and to generate bijective and lowly distorted mapping results so the faces have the correct orientation, thus creating a 3d spherical geometry object. Simulation results show that it is possible to achieve a considerable correspondence between the angle and area perspective distortion.

Face Recognition Using HMM-LBP

Despite the existence of many biometric systems such as hand geometry, iris scan, retinal scanning and fingerprints, the face recognition will remain a powerful tool due to many advantages such as his low cost, the absence of physical contact between user and biometric system, and his user acceptance. Thus, a large number of face recognition approaches has been lately done. In this paper, we present a new 2D face recognition approach called HMM-LBP permitting the classification of a 2D face image by using the LBP tool (Local Binary Pattern) for feature extraction. It is composed of four steps. First, we decompose our face image into blocs. Then, we extract image features using LBP. Next, we calculate probabilities. Finally, we select the maximum probability. The obtained results were presented to prove the efficiency and performance of the novel technique.

Multi-layer compression algorithm for 3D deformed mesh based on multi library wavelet neural network architecture

We propose in this paper a 3D mesh compression algorithm for 3D deformation objects to facilitate the transmission of deformed object to another. This algorithm allows eliminating an object in the sequence of deformed objects and reducing the information needed to represent the geometry of a mesh sequence. In our approach, we used Multi Library Wavelet Neural Network architecture (MLWNN) to align features of mesh and minimize distortion with fixed features. The introduced method minimizes the sum of the distances between all the corresponding vertices. It computes deformed ROI (Region Of Interest), updates and optimizes it to align the mesh features. First, our compression was performed using spherical geometrical image obtained by our trust region spherical parameterization. Geometrical images also facilitate compression and level-of-detail control. Second, the spherical wavelet transformation was used to decompose the geometrical image into multi-resolution sub-images characterizing the underlying functions in a local fashion in both spatial and frequency domains. Experimental results show that the progressive compression algorithm yields efficient compression capabilities with minimal set of features used to have good deformation scheme.

A Multiresolution Approach for Blind Watermarking of 3D Meshes Using Spiral Scanning Method

During the last decade, the flow of 3D objects is increasingly used everywhere. This wide range of applications and the necessity to exchange 3D meshes via internet raise major security problems. As a solution, we propose a blind watermarking algorithm for 3D multi-resolution meshes ensuring a good compromise between invisibility, insertion rate and robustness while minimizing the amount of memory used during the execution of our algorithm. To this end, spiral scanning method is applied. It decomposes the mesh into GOTs (a Group Of Triangles). At each time, only one GOT will be loaded into memory to be watermarked. It undergoes a wavelet transform, a modulation then embedding data. Once finished, the memory will be released to upload the next GOT. This process is stopped when the entire mesh is watermarked. Experimental tests showed that the quality of watermarked meshes is kept despite the high insertion rate used and that memory consumption is very reduced (until 24 % of memory reduction). As for the robustness, our algorithm overcomes the most popular attacks in particular compression. A comparison with literature showed that our algorithm gives better results than those recently published.

Implementation of skin color selection prior to Gabor filter and neural network to reduce execution time of face detection

This paper proposes a face detection system based on the skin color, the Gabor filter and the neural network. The use of Gabor filters and neural networks for face recognition is not new. However, the principal focus of the proposed paper is the implementation of skin color selection prior to Gabor filters and neural networks on order to reduce computation time. First, we analyze the skin color to extract skin areas which have an important probability to be faces. This technique robust to the lighting variation allows extracting, from an image, skin areas. We utilize this method to avoid wrong detection and to help the system detect the face in the right areas and minimize the research time. Second, to extract features, we propose a technique using the Gabor filter applied on the localized skin space. Finally, the vectors of the face features obtained by the Gabor filter are used as the input of a neural network classifier which classifies an input image pixel as a face or nonface pixel. Some results are shown to approve our approach efficiency.

Crypto-watermarking system for safe transmission of 3D multiresolution meshes

The idea of digitizing documents to be archived or shared has given rise to a variety of new data types such as 3D meshes. The sharing of this data type between remote users, using high-speed computer networks and remote multimedia databases, poses great security problems. As a solution, we propose, in this paper, a new crypto watermarking algorithm. The originality of this work lies in joining cryptography with digital watermarking to secure 3D multiresolution meshes. To this end, three steps should be executed. The first is the watermark preparation. It consists, firstly, in applying the Secure Hash Algorithm 1 algorithm to generate an electronic signature of the mesh source. Secondly, the logo undergoes an encryption using Advanced Encryption Standard algorithm. To end this step, the signature and the encrypted logo pass through a convolutional encoder to obtain a codeword. As for the second step, it is called mesh preparation and it consists in applying a spiral scanning method to the mesh to split it into Groups Of Triangles. For each Group Of Triangles, a wavelet transform is applied to generate the corresponding Wavelet coefficients vector. Finally, embedding data occurs using the cylindrical coordinate system, a modulation and the least Significant Bit method. The experiment of our algorithm proves that it allows a very high insertion rate without influencing the mesh quality. Our algorithm also minimizes the amount of memory used. Moreover, it is robust against the most popular attacks. Our results show that our algorithm presents an improvement in comparison with recently published results.

Join Cryptography and Digital Watermarking for 3D Multiresolution Meshes Security

The main objective of this paper is to combine cryptography with digital watermarking to secure 3D multiresolution meshes. The result is a new crypto-watermarking algorithm which contains 3 parts: the first part is called watermark preparation and it aims to encrypt the logo using AES (Advanced Encryption Standard) algorithm, to combine the encrypted logo with a binary sequence obtained by transforming the description of the host mesh using ASCII (American Standard Code for Information Interchange) code and to encode the hole watermark using a convolutional encoder. As for the second part, it is said mesh preparation and it consists in applying a wavelet transform to the mesh in order to generate a wavelet coefficient vector. All these coefficients will be presented in the spherical coordinate system. Finally, the third part of our algorithm intervenes to insert the data in the mesh using the LSB (Least Signifiant Bit) method. Found results prove that we are able to insert a high amount of data without influencing the mesh quality. The application of the most popular attacks does not prevent a correct extraction of data already inserted. Our algorithm is, then, robust against these attacks.

Optimization techniques of static 3D triangular mesh compression: A survey

In the last decade, the extensive evolution of 3D graphic applications has induced developpers to enhance the 3D mesh compression techniques. Therefore, 3D models are compressed, transmitted and rendered more and more in real-time and with high quality. Moreover, many out-of-core algorithms are proposed to process complex objects. In this paper, we review the major existing technologies for the single-rate and progressive 3D mesh compression. Then, representative out-of-core approaches are surveyed in details. Finally, we represent some parallel schemes exploiting the evolution of many-core GPU architecture.

Sphere-Tree Semi-regular Remesher

Surface meshes have become widely used since they are frequently adopted in many computer graphic applications. These meshes are often generated by isosurface representations or scanning devices. Unfortunately, these meshes are often dense and full of redundant vertices and irregular sampling. These defects make meshes not capable to support multiple applications; such as display, compression and transmission. To solve these problems and reduce the complexity, the mesh quality connectivity regularity must be ameliorated. Its improvement is called re-meshing. This paper presents a novel re-meshing approach based on Sphere-Tree construction. First, we approximate the original object with a dictionary of multi-dimensional geometric shapes spheres called Sphere-Tree which is, then, re-meshed. Finally, we use a refinement step to avoid artifacts and produce a new semi-regular mesh.