Search

This article introduces the basic element to build an augmented reality system allowing to insert a 2D object in a real 3D scene in real time. The first step consists in locating the place where to insert the object using an abstract marker, this marker is a rectangle that surrounds the minimum area of the object's contour detected through its color. This rectangle is a plane surface that provides the position of its four points in the images acquired in real time which allows to have a real time tracking of the detected object. Planar homography describes exactly the relationship between the key points if the scene is flat and only requires four key points to produce an exact solution of the camera position to align a 2D virtual object in a 3D real scene.

This article presents a new image segmentation approach based on the principle of clustering optimized by the meta-heuristic algorithm namely: SCA (Algorithm Sinus Cosine). This algorithm uses a mathematical model based on trigonometric functions to solve optimization problems. Such an approach was developed to solve the drawbacks existing in classic clustering techniques such as the initialization of cluster centers and convergence towards the local optimum. In fact, to obtain an “optimal” cluster center and to improve the image segmentation quality, we propose this technique which begins with the generation of a random population. Then, we determine the number of clusters to exploit. Later, we formulate an objective function to maximize the interclass distance and minimize the intra-class distance. The resolution of this function gives the best overall solution used to update the rest of the population. The performances of the proposed approach are evaluated using a set of reference images and compared to several classic clustering methods, like k-means or fuzzy c-means and other meta-heuristic approaches, such as genetic algorithms and particle swarm optimization. The results obtained from the different methods are analyzed based on the best fitness values, PSNR, RMSE, SC, XB, PC, S, SC, CE and the computation time. The experimental results show that the proposed approach gives satisfactory results compared to the other methods.

Automatic Speech Recognition (ASR) for Amazigh speech, particularly Moroccan Tarifit accented speech, is a less researched area. This paper focuses on the analysis and evaluation of the first ten Amazigh digits in the noisy conditions from an ASR perspective based on Signal to Noise Ratio (SNR). Our testing experiments were performed under two types of noise and repeated with added environmental noise with various SNR ratios for each kind ranging from 5 to 45 dB. Different formalisms are used to develop a speaker independent Amazigh speech recognition, like Hidden Markov Model (HMMs), Gaussian Mixture Models (GMMs). The experimental results under noisy conditions show that degradation of performance was observed for all digits with different degrees and the rates under car noisy environment are decreased less than grinder conditions with the difference of 2.84% and 8.42% at SNR 5 dB and 25 dB, respectively. Also, we observed that the most affected digits are those which contain the "S" alphabet.

In this paper, we present a new technique of tridimensional reconstruction from a sequence of uncalibrated stereo images taken with cameras having varying parameters. At first, our system allows to recover initial coordinates of a set of 3D points. In this context, we have used our method of self-calibration based on the use of unknown 3D scene with its image projections and genetic algorithms to estimate all intrinsic parameters. After that extrinsic parameters are estimated based on classical pose estimation algorithms. Matching points and estimated value of intrinsic and extrinsic parameters are used to estimate initial 3D model that helps us in the initialization step. In order to have a reliable and relevant 3D reconstruction the proposed method is based on good and new exploitation of bundle adjustment (without camera poses initialization) technique based on Levenberg-Marquardt optimization with the aim to estimate our optimal 3D model that has special features compared to the classical case because it masks the pose parameters estimation in the optimization process. Finally, 3D mesh of the 3D scene is constructed with Delaunay algorithm and the 2D image is projected on the 3D model to generate the texture mapping. Experiments is conducted on real data to achieve demonstrate the validity and the performance of the proposed approach in terms of convergence, simplicity, stability and reconstruction quality.

Structure from Motion (SfM) is a 3D reconstruction approach for estimating camera poses and 3D structure from calibrated images. The recovered 3D structure is a sparse 3D point cloud that not permit to well define the shape of the object/scene. We must therefore move to a dense 3D reconstruction that requires a step of the dense matching between pairs of consecutive images, which requires a long calculation time. To reduce computation time, we have proposed an algorithm for the good choice of image pairs that will be used by the Modified Match Propagation (MMP) to improve the sparse 3D reconstruction. These image pairs will be selected on the basis of the result already achieved by SfM. The MMP algorithm will be applied for each image pair to retrieve new matches and their 3D coordinates. The final 3D point cloud is achieved by fusion of results obtained from the image pairs selected. The realistic 3D model is recovered after applying the Poisson surface reconstruction method with texture mapping. The results of the experiments show the speed of the proposed approach without losing quality of 3D reconstructed models.

In this work, we proposed a new approach to encrypt color images using two matrices with size of 16 × 16 whose integer values are between 0 and 255 generated randomly, and the bit-shift operators. These matrices are used to perform the first encryption phase. The first value of the first matrix is calculated from the pixels of each channel (red, green and blue) of the original image; the rest of the values are randomly generated; each value must be unique; the values of the second matrix are unique and generated randomly. The first encryption phase of the original image is done by digraph (two-digit sequence). We take the first digit in the first matrix, the second digit in the second matrix; then, we look in these matrices for the numbers that complete the rectangle. In the second encryption phase, we used a right circular shift of bits; the number of bits to shift is calculated according to a function which considers the values of the two matrices as well as their positions (row and column). Therefore, any change in the two keys (two matrices) will completely change the encrypted image. Our encryption system is resistant against brute force attacks, statistical attacks as well as differential attacks. The results are justified by applying several safety criteria, such as correlation coefficient, entropy and peak signal-to-noise ratio (PSNR). In addition, our method is very sensitive to the change made, either in the original image or in the two keys used for the encryption, which was justified by calculating the number of changing pixel rate (NPCR > 99.69) and the unified averaged changed intensity (UACI > 33.54).

Image mosaicking is a process of assembling multiple images to create an image with a larger field of view. It is used in different studies, but some errors, like ghosting or parallax effects, could occur when the images contain dynamic elements. To avoid the failure of the mosaic and to solve these errors, a new method that searches an optimal seamline for dynamic mosaicking is presented. By finding regions that are similar between the images, and regions that are not alike, the seamline was computed by going through the similar regions and by avoiding the not common regions. To achieve this, a combination of Canny edge detector, and the outliers and inliers from the RANSAC method were used to identify these regions. Then, the regions were incorporated in an intensity difference to create a map that reveals them. Thus, the optimal seamline was computed by going through the similar regions and by avoiding the unalike regions. The experimental results show that the proposed approach is capable of generating robust mosaics against ghosting and parallax effects.

Sparse 3D reconstruction, based on interest points detection and matching, does not allow to obtain a suitable 3D surface reconstruction because of its incapacity to recover a cloud of well distributed 3D points on the surface of objects/scenes. In this work, we present a new approach to retrieve a 3D point cloud that leads to a 3D surface model of quality and in a suitable time. First of all, our method uses the structure from motion approach to retrieve a set of 3D points (which correspond to matched interest points). After that, we proposed an algorithm, based on the match propagation and the use of particle swarm optimization (PSO), which significantly increases the number of matches and to have a regular distribution of these matches. It takes as input the obtained matches, their corresponding 3D points and the camera parameters. Afterwards, at each time, a match of best ZNCC value is selected and a set of these neighboring points is defined. The point corresponding to a neighboring point and its 3D coordinates are recovered by the minimization of a nonlinear cost function by the use of PSO algorithm respecting the constraint of photo-consistency. Experimental results show the feasibility and efficiency of the proposed approach.

Point matching via corresponding circles (ICC) is a technique for removing outliers (mismatches) from given putative point correspondences in image pairs. It can be used as a basis for a wide range of applications including structure-from-motion, 3D reconstruction, tracking, image retrieval, registration, and object recognition. In this paper, we propose a new method called Fast Identification of point correspondences by Corresponding Circles (FICC) that improves the quality of the rejection mismatches and reduces the cost of computing it. In particular, we propose a new strategy that aims to take better advantage of the corresponding circles and reduces the number of putative points correspondences tested by the corresponding circles in each iteration rather than all set of putative correspondences, as in the original ICC. This reduces the computing time and together with a more efficient tool for rejecting mismatches which leads to significant gains in efficiency. We provide comparative results illustrating the improvements obtained by FICC over ICC, and we compare with many state-of-the-art methods.

In this paper, we present a novel efficient face description method called Electric Virtual Binary Pattern (EVBP). The main idea of EVBP descriptor is to combine Local Binary Pattern (LBP) and our new Model based on the Virtual Electric Field. This model consider the neighborhood of each pixel as a grid of virtual electric charges that are electrostatically balanced. Then, we apply the LBP principle for this neighborhood to generate the new EVBP pixel representation. Based on the four trivial space directions, this representation is computed using the corresponding four electrical interactions. Moreover, the spatially enhanced Local Binary Pattern Histogram (eLBPH) algorithm is employed to extract features. Therefore, the proposed EVBP descriptor led to reduce the features vector size by 93.75%. Consequently, we moved from 255 bin-histograms for LBP to 16 bin-histograms for EVBP descriptor. Extensive experiments were carried on relevant databases have proved the effectiveness of the proposed approach.

This paper introduces a new mesh segmentation approach into semantic parts, most closely resemble those made by humans, which is based on the pixel color of the images used in the 3D reconstruction. This approach allows to segment the mesh into semantic and a much simpler way than most of the mesh segmentation methods that are based on the geometrical characteristics of the mesh. The principle of our method is to establish a link between the color objects of the scene and the mesh while exploiting the link between the interest points of the images brought into play and the vertices of 3D mesh. The results in great part, reflect the efficiency and performance of our method.

In this paper we examined the human voice of 20 adults (20 smokers and 20 non-smokers) to determine the effects of cigarette smoking on formants frequency, pitch, shimmer and jitter based on 3 Amazigh language vowels (A, I, U). The statistical data parameters are collected from male Moroccan speakers aged between 26 and 50 years old. Our results show that, the pitch values of smokers are lower compared to those of non-smokers. Also, smokers’ formants frequency F1 and F2 are close to non-smokers ones for the three considered vowels .Whereas, F3 and F4 are lower in the case of smokers. Shimmer and Jitter analysis showed higher values for these parameters among smoker.

This paper presents a complete pipeline of the reconstruction and the modeling of the unknown complex 3D scenes from a sequence of unconstrained images. The proposed system is based on the formulation of a nonlinear cost function by determining the relationship between 2D points of the images and the cameras parameters; the optimization of this function by a genetic algorithm makes finding the optimal cameras parameters. The determination of these parameters allows thereafter to estimate the 3D points of the observed scene. Then, the mesh of the 3D points is achieved by 3D Crust algorithm and the texture mapping is performed by multiple dependent viewpoints. Extensive experiments on synthetic and real data are performed to validate the proposed approach, and the results indicate that our system is robust and can achieve a very satisfactory reconstruction quality.

In this paper, we present a flexible and fast system for multi-scale objects/scenes 3D reconstruction from uncalibrated images/video taken by a moving camera characterized by variable parameters. The proposed system is based on incremental structure from motion and good exploitation of bundle adjustment. At first, from two selected images, our system allows to recover, in a well-chosen reference, coordinates of a set of 3D points. In this context, we have proposed a new method of self-calibration based on the use of two unknown scene points with their image projections. After that, new images are inserted progressively using 3D information already obtained. Local bundle adjustment is used to adjust the new estimated entities. At some time, we introduce a global bundle adjustment to adjust as best as possible all estimated entities and to have an initial 3D model of quality covering an interesting part of the object/scene. This model will be used as reference for the insertion of the rest of images. The proposed system allows to obtain satisfactory results within a reasonable time.

The matching points extracted from images play a very important role in many applications and particularly in computer vision. The use of point sets as being characteristics that describe the entire images brought into play, it greatly contributes to the reduction of the execution time, unlike the use of all the information contained in these images. The major problem of the matching process is the possibility to generate a large number of false correspondences, or outliers, in addition to a limited number of true correspondences (inliers). The objective of this paper is to propose a robust algorithm to eliminate or reduce the false correspondences, or outliers, among the putative set extracted from stereoscopic images. The principle of our method is based on the notion of belonging to the corresponding circles and the concept of similarity of stereoscopic images. The results largely reflect the efficiency and performance of our algorithm in comparison to the other used methods in this framework like RANSAC algorithm.

Automatic speech recognition is a technology that allows a computer to transcribe in real time spoken words into readable text. In this work an HMM automatic speech recognition system was created to detect smoker speaker. This research project is carried out using Amazigh language for comparison of the voice of normal persons to smokers one. To achieve this goal, two experiments were performed, the first one to test the performance of the system for non-smokers for different parameters. The second experiment concern smokers speakers. The corpus used in this system is collected from two groups of speaker, non-smokers and smokers native Morocan tarifit speakers aged between 25 and 55 years. Our experimental results show that we can use our system to make diagnostic for smoking people and confirm that a speaker is smoker when the observed recognition rate is below 50%.

This paper proposes a new appearance model for human tracking based on Mean Shift framework. The proposed method uses a novel target representation by using joint Color-Texture features and Foreground-Weighted Histogram (CTFWH) for a more distinctive and effective target representation. Our contribution is threefold: firstly, to exploit the texture information of the target, we have used joint color-texture histogram to represent the target. Local Binary Pattern (LBP) technique is employed to identify texture features in the target region. Secondly, we have proposed a representation model of the foreground region named Foreground-Weighted Histogram (FWH), in order to exploit the significant features of the foreground region and to use it for selecting only the salient parts from the target model. Thirdly, we propose a simple method to update the foreground model due to the important foreground changes over the tracking process. Hence, by combining these concepts we generate new features for target representation and human tracking. The proposed method is designed for human tracking in complex scenarios and tested for comparative results with existing state-of-the-art algorithms. Experimental results on numerous challenging video sequences verify the significance of the proposed approach in terms of robustness and performance to complex background, illumination and appearance changes, similar target and background appearance, presence of distractors, target and camera motion, occlusions and large background variation.

In this approach, we will process a self-calibration problem of camera characterized by varying parameters. Our approach is based on estimating of homography of the plane at infinity and depths of interest points. This estimation is made from resolution of nonlinear equation system that is formulated from the projection of some points of the scene in the planes of different images. The relationships established between the homography of the plane at infinity and matches, between images, and those established between points of the 3D scene and their projections, in image planes, allow formulating the second nonlinear equations. This system is minimized by using the Levenberg-Marquardt algorithm to estimate the intrinsic parameters of used camera. This approach has several strong points: i) The use of any cameras (having varying parameters), ii) The use of any scenes (3D) and iii) the use of a minimum number of images (two images only). Experiments and simulations show the performance of this approach in terms of stability, accuracy and convergence.

Interactions between trophic and halieutic populations, have often been ignored in most mathematical models that are interested in prey–predator dynamics. However, the presence of vegetation in hydraulic environments, is important, and it has a direct impact on the life cycle of many fishes. In fact, aquatic populations are essential for providing oxygen, food and also shelter for grazers. For this, we suggest a mathematical and optimization approach, in an attempt to discuss the possibility of finding effective harvesting control strategies that aim to optimize the fishing efforts without affecting the trophic-halieutic populations and compromising with the interests of fishermen. The model is in the form of three ordinary differential equations, and which describes dynamics of multi-species of a fishery that includes grazer–predator fishes and aquatic plants, during fishing periods. We study the stability of the proposed differential system, and we suggest harvesting optimal control approaches for the environmental sustainability and bioeconomic cases after the introduction of two control functions in the model and which represent the efforts of fishing in grazer and predator populations respectively. The two optimal controls are characterized in the first case, based on Pontryagin’s maximum principle, and we seek their analytical formulations in the second case, when they are singular functions. Hence, these two optimal control approaches lead to two two-point boundary value problems which are resolve numerically based on the forward-backward sweep method with the incorporation of iterative Runge-Kutta fourth order progressive-regressive schemes.

Mosaic reconstruction is a stitching process of multiple images, of a particular scene, in a single frame that provides a larger amount of information compared to the separate images. Nowadays, image mosaic is a key tool that has invaded different fields and disciplines such as photography, virtual environment, medicine, etc. In this work, we propose a new pre-processing approach of multi-scale images we have named MSIP (Multi-Scale Image Pre-processing), invariant to scale changes and based on the distance between the matched points detected by SIFT. Its main purpose is to correct the scale difference between images to reduce outliers and alignment errors. The experimentation and statistical analysis, on a real database, show the robustness of our approach by improving the quality of mosaic results.

In this article, we propose a new method of image stitching that computes, in a robust manner, the transformation model applied to creating a panorama that is close to reality. The random selection of matching points used in existing methods, using Random Sample Consensus (RANSAC) or the threshold of the execution process (iteration number) cannot generally provide sufficient precision. Our approach, in this regard, comes to solve this problem. The calculation of the transformation model is based on the VORONOI diagram that divides images into regions to be used in the matching instead of control points. In this case, the transformation estimation will be based on the regions seeds that provide the best correlation score. Among the advantages of our method is solving problems related to outliers that can, in existing methods, affect the reliability of the mosaic. The results obtained are satisfactory in terms of stability, quality, execution time and reduction of the computational complexity.

The purpose of the work presented in this paper is to describe new method of 3D reconstruction from one or more uncalibrated images. This method is based on two important concepts: geometric constraints and genetic algorithms (GAs). At first, we are going to discuss the combination between bundle adjustment and GAs that we have proposed in order to improve 3D reconstruction efficiency and success. We used GAs in order to improve fitness quality of initial values that are used in the optimization problem. It will increase surely convergence rate. Extracted geometric constraints are used first to obtain an estimated value of focal length that helps us in the initialization step. Matching homologous points and constraints is used to estimate the 3D model. In fact, our new method gives us a lot of advantages: reducing the estimated parameter number in optimization step, decreasing used image number, winning time and stabilizing good quality of 3D results. At the end, without any prior information about our 3D scene, we obtain an accurate calibration of the cameras, and a realistic 3D model that strictly respects the geometric constraints defined before in an easy way. Various data and examples will be used to highlight the efficiency and competitiveness of our present approach.

This work proposes a method of camera self-calibration having varying intrinsic parameters from a sequence of images of an unknown 3D object. The projection of two points of the 3D scene in the image planes is used with fundamental matrices to determine the projection matrices. The present approach is based on the formulation of a nonlinear cost function from the determination of a relationship between two points of the scene and their projections in the image planes. The resolution of this function enables us to estimate the intrinsic parameters of different cameras. The strong point of the present approach is clearly seen in the minimization of the three constraints of a self-calibration system (a pair of images, 3D scene, any camera): The use of a single pair of images provides fewer equations, which minimizes the execution time of the program, the use of a 3D scene reduces the planarity constraints, and the use of any camera eliminates the constraints of cameras having constant parameters. The experiment results on synthetic and real data are presented to demonstrate the performance of the present approach in terms of accuracy, simplicity, stability, and convergence.

The Mean Shift procedure is a popular object tracking algorithm since it is fast, easy to implement and performs well in a range of conditions. However, classic Mean Shift tracking algorithm fixes the size and orientation of the tracking window, which limits the performance when the target's orientation and scale change. In this paper, we present a new human tracking algorithm based on Mean Shift technique in order to estimate the position, scale and orientation changes of the target. This work combines moment features of the weight image with background information to design a robust tracking algorithm entitled Scale and Orientation-based Background Weighted Histogram (SOBWH). The experimental results show that the proposed approach SOBWH presents a good compromise between tracking precision and calculation time, also they validate its robustness, especially to large background variation, scale and orientation changes and similar background scenes.

In this paper, we address a problem of reconstruction of three-dimensional scenes from images taken by cameras, with varying parameters, from different views. This method is based on the projection of 3D points in the image planes. The relationships between the matches and the camera parameters are used to formulate a nonlinear equation system. This system is transformed into an objective function, which is minimized by a genetic algorithm to estimate the intrinsic and extrinsic camera parameters. Finally, the coordinates of 3D points of the scene are obtained by solving a linear equation system. The experiments on synthetic and real data show the quality of this work and the good obtained results.

In this article we present a practical self-calibration method of camera having variable intrinsic parameters. This method is an extension of a method which we have already published. The main idea of our method addressed in this paper is the use of interest points automatically detected in two images of an unknown 3D scene to self-calibrate the camera. These interest points are the projections of the vertices of unknown 3D parallelograms (each triplet of interest points is the projections of three vertices of a unique 3D parallelogram). The projection of the vertices of parallelograms in the two image planes of the camera allows us to obtain two equations based on the intrinsic camera parameters. Seeing that our camera can change its intrinsic parameters from a view to another so the number of unknowns is ten hence the need to use at least five parallelograms in the scene to obtain other equations in function of intrinsic camera parameters. The paper tests the performance of the proposed approach on the both synthetic and real data.

In this paper, we are interested in the problem of the mosaic of images from a method of self-calibration of CCD cameras with varying intrinsic parameters. We use a new method of self-calibration to estimate the intrinsic parameters of the cameras. This method is based on the use of a 3D scene containing an unknown equilateral triangle. The plane of the considered triangle permits us to simplify the equations of self-calibration and to estimate the camera intrinsic parameters. The proposed approach is tested on real images of the same point in space, taken by different orientations of the camera. We show the performance and robustness of the intrinsic parameters estimated by this method on the image mosaic problem.

In this paper, we present a new method for multi-view 3D reconstruction based on the use of a binocular stereo vision system constituted of two unattached cameras to initialize the reconstruction process. Afterwards , the second camera of stereo vision system (characterized by varying parameters) moves to capture more images at different times which are used to obtain an almost complete 3D reconstruction. The first two projection matrices are estimated by using a 3D pattern with known properties. After that, 3D scene points are recovered by triangulation of the matched interest points between these two images. The proposed approach is incremental. At each insertion of a new image, the camera projection matrix is estimated using the 3D information already calculated and new 3D points are recovered by triangulation from the result of the matching of interest points between the inserted image and the previous image. For the refinement of the new projection matrix and the new 3D points, a local bundle adjustment is performed. At first, all projection matrices are estimated, the matches between consecutive images are detected and Euclidean sparse 3D reconstruction is obtained. So, to increase the number of matches and have a more dense reconstruction, the Match propagation algorithm, more suitable for interesting movement of the camera, was applied on the pairs of consecutive images. The experimental results show the power and robustness of the proposed approach.

In this paper we present a technique of camera self-calibration using the improved genetic algorithm based on the Kruppa equations which will be developed in the case where the cameras are characterized by the varying intrinsic parameters. A real coding of genetic algorithm is used to solve this problem. The solutions of the intrinsic parameters of various cameras are encoded in a vector of real values. New genetic operators are used to obtain the solutions of the next generation. An optimal estimate of the intrinsic cameras parameters is obtained by minimizing the cost function by using a modified genetic algorithm. Compared with traditional optimization methods, the camera self-calibration by this approach can avoid being trapped in a local minimum, and converges quickly toward the optimal solution without initial estimate of the cameras parameters. Our study is performed on synthetic and real data to demonstrate the validity and performance of the presented approach. The results show that the proposed technique is both accurate and robust.