Deep Neural Networks (DNNs) have been successfully used in various fields, nonetheless their lack of clarity poses serious issues in trust, interpretability, and responsibility in critical areas such as healthcare. In this paper, we propose a new hybrid approach of explainability by combining SHapley Additive exPlanations (SHAP) and Local Interpretable Model Agnostic Explanations (LIME) towards improving the interpretability of deep learning models. SHAP offers a global feature importance, which gives a wide-ranging perspective on model behavior while LIME provides local instance-based explanations that clarify individual predictions. Our hybrid approach of SHAP and LIME combines both techniques to improve understanding of model decision making. We evaluated our method with two case studies including Handwritten Digit Recognition (MNIST) and Alzheimer’s disease detection. The experimental results demonstrated that the hybrid approach improved accuracy by $97.92 \%$, increased trust in Al powered decision making, and outperformed other standalone explainability methods in local and global interpretability. The work presented in this document deepens the research on explainable AI (XAI) by proposing a meaningful and straightforward approach that is generally applicable to deep learning systems.

Secure image encryption is necessary to protect our confidential data from malicious access. Chaos-based encryption algorithms have gained significant attention in performing the security of encrypted images because of their complex and random-like behavior. However, the robustness of these encryption schemes against cryptanalysis remains a critical area of research. In this paper, we have designed an innovative deep-learning (DL) method based on the known-plaintext attack (KPA) to focus on the security of chaotic image encryption algorithms. This approach uses a convolutional neural network (CNN) model that has been pre-trained using a set of ciphertext–plaintext pairs to learn chaotic cryptosystem operation mechanisms. Our method is applied to three existing chaotic encryption algorithms. Experimental results show the ability of DL-based models to break different chaotic cryptosystems in contrast to traditional KPA-based cryptanalysis algorithms.

With advances in technology, the issue of object detection and recognition has gained significant recognition in the field of computer vision. There are currently several algorithms that address this growing demand, namely region-based convolutional neural networks (R-CNN) and the You Only Look Once (YOLO) technique. The R-CNN technique encompasses a range of methodologies designed to address object localization and recognition tasks. In addition, the YOLO technique is a distinct set of methodologies that focuses primarily on real-time object recognition and fast performance. The R-CNN and YOLO techniques, in particular, have undergone subsequent improvements, resulting in higher levels of accuracy and performance than their predecessors. The aim of this article is to review these various object detection methods based on CNN.

Abstract Due to its widespread application in the field of public security, anomaly detection in crowd scenes has recently become a hot topic. Some deep learning-based methods led to significant accomplishments in this field. Nevertheless, due to the scarcity of data and the misclassification of queries which most of them suffer to some extent from a sudden and infrequent overfitting. Though, we tried to solve the above problems, understand the long video streams and establish an accurate and reliable security system in order to improve its performance in detecting anomalies. We also referred to the hash technique, which has proven to be the most efficient method used when researching about large-scale image recovery. Thus, this article offers a smart video anomaly detection solution. In this paper, we combine the advantages of both deep hashing and deep auto-encoders to show that tracking changes in deep hash components across time and can be used to detect local anomalies. More precisely, we start with a new technique to minimize the mass of input data and information in order to decrease the time of calculation using a new dynamic frame skipping technique. Then, we propose to measure local anomalies by combining semantic with low-level optical flows to balance the performance and perceptibility. The experimental results illustrate that the proposed methods surpass these baselines for the detection and localization of anomalies.

Human Activity Recognition (HAR) is one of the most important research areas in the fields of health and human-machine interaction. The creation of several artificial intelligence-based models for activity identification has resulted in poor long-term performance in the actual world since these methods are unable to extract spatial and temporal information. Though, Deep learning is starting to replace well-established hand-crafted techniques that rely on expertly built feature extraction and classification techniques in the field of HAR. Nevertheless, it is challenging to acquire an overview of the suitability of these discrete implementations of custom deep architectures for issues ranging from the recognition of manipulative gestures to the identification and segmentation of physical activities. Given these constraints, we develop an innovative Stacked Auto Encoder (SAE) and Extreme Learning Machine (ELM) architecture based on temporal feature to produce a new model for HAR.. This feature is connected to speed of movement and fed to a new neural network architecture as temporal stream. A key step in this approach is the selection of features that characterize the complexity of human actions in time. Results show that the proposed model outperforms many of the existing deep and machine learning techniques.

In today’s era of rapid computer network advancements, an enormous volume of messages is exchanged daily, touching several sectors, including the medical field. While the Internet provides flexible communication capabilities, it also presents many security challenges including data theft, data duplication, data leakage, and copyright protection. One potential solution to address this issue is the application of watermarking. This paper presents a robust watermarking algorithm for medical images that combines the Beta Chaotic Map (BCM), Discrete Wavelet Transform (DWT), and Singular Value Decomposition (SVD). Initially, the watermark undergoes encryption using BCM. Subsequently, the medical image is decomposed into four sub-bands (LL, LH, HL, and HH) through DWT. The low-frequency region LL contains the embedded watermark information, which is obtained by computing the embedding singular value using SVD. In the extraction process, the watermarked medical image is decomposed using DWT, and the watermark is extracted by reversing the SVD. Finally, the watermark is decrypted by applying BCM. To ensure the invisibility of the watermark in the host image and robustness against various attacks (e.g., JPEG compression, JPEG2000 compression, sharpening, noise, and filtering), the algorithm employs metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). Experimental results validate the algorithm’s effectiveness through tests based on Normalized Correlation (NC).

In this paper, a brand-new approach to image encryption is put forth that is based on the new Beta chaotic map, the Beta wavelet, and the Latin square. The proposed strategy is made up of various steps. The Wavelet Beta map is used to produce the random key after generating the Latin square S-box. The encryption stage uses the obtained key. The ciphered images have undergone numerous tests after the encryption procedures, including histogram analysis, information entropy analysis, and differential analysis. The results, which demonstrate that the proposed method has high efficiency and satisfactory security, are promising when compared to earlier systems and demonstrate that it is appropriate for image data transmission.

The results of automatic learning algorithms based on deep neural networks are impressive, and they are extensively used in a variety of fields. However, access to private information, frequently sensitive to confidentiality (financial, medical, etc.), is required in order to use them. This calls for good precision as well as special attention to the privacy and security of the data. In this paper, we propose a novel approach to solve this issue by using Convolutional Neural Network (CNN) model over encrypted data. In order to achieve our contribution, we focus on approximating the often used activation functions that seem to be the key functions in CNN networks which are: ReLu, Sigmoid and Tanh. We start by creating a low-degree polynomial, which is essential for a successful homomorphic encryption (HE). This polynomial which is based on Beta function and its primitive will be used as an activation function. The next step is to build a CNN model using batch normalization to ensure that the data are contained inside a limited interval. Finally, MNIST is used in order to evaluate our methodology and assess the effectiveness of the proposed approach. The experimental results support the efficacy of the proposed approach.

Abstract Affective computing is a key research topic in artificial intelligence which is applied to psychology and machines. It consists of the estimation and measurement of human emotions. A person’s body language is one of the most significant sources of information during job interview, and it reflects a deep psychological state that is often missing from other data sources. In our work, we combine two tasks of pose estimation and emotion classification for emotional body gesture recognition to propose a deep multi-stage architecture that is able to deal with both tasks. Our deep pose decoding method detects and tracks the candidate’s skeleton in a video using a combination of depthwise convolutional network and detection-based method for 2D pose reconstruction. Moreover, we propose a representation technique based on the superposition of skeletons to generate for each video sequence a single image synthesizing the different poses of the subject. We call this image: ‘history pose image’, and it is used as input to the convolutional neural network model based on the Visual Geometry Group architecture. We demonstrate the effectiveness of our method in comparison with other methods in the state of the art on the standard Common Object in Context keypoint dataset and Face and Body gesture video database.

In this article, we investigate the joint task assignment and power control problems for Device-to-Device (D2D) offloading communications with energy harvesting. Exploiting the D2D links for data offloading allows reducing the traffic load of the cellular base stations. The energy consumed by the D2D transmitters for data offloading can be compensated by energy harvesting. The main objective is to maximize the energy efficiency (EE) under energy causality and delay constraints, assuming a harvest–transmit model. Hence, the proposed model results in a nonconvex problem. We first derive an equivalent and more tractable optimization problem by exploiting nonlinear fractional programming, also known as the Dinkelbach method. We propose a layered optimization method by decoupling the EE maximization problem into power allocation and offloading assignment. The first step consists of computing the optimal power values by applying the conjugate gradient method. In the second step, the problem of the D2D pair formation for data offloading amounts to the bipartite graph matching. It can be solved to optimality using the Hungarian algorithm. Extensive simulations were performed on various network scenarios. Numerical results show that the proposed resource allocation scheme achieves remarkable improvements in terms of network EE.

Abstract The emotion recognition field has two major issues. On the one hand, it is difficult to find the same emotion state in different persons since they may express the same emotion state in various ways. On the other hand, it is also hard to seek the difference between expressions of the same person because some emotion states are too subtle to discriminate. The focus of this work is to solve these two problems by proposing a new approach of emotion recognition. This novel approach allows our emotion recognition system to classify 18 emotions (primary emotions and their intensities). First, we proposed textual definitions of the intensity emotions. Then, we created our emotion recognition system, which is composed of three stages: pre-treatment, feature extraction and classification. We used the deep learning for the feature extraction and the fuzzy logic for the classification. The experimental test demonstrates the efficiency of our system for primary emotions and their intensities’ classification compared to other methods.

Abstract Emotion recognition is a key work of research area in brain computer interactions. With the increasing concerns about affective computing, emotion recognition has attracted more and more attention in the past decades. Focusing on geometric positions of key parts of the face and well detecting them is the best way to increase accuracy of emotion recognition systems and reach high classification rates. In this paper, we propose a hybrid system based on wavelet networks using 1D Fast Wavelet Transform. This system combines two approaches: the biometric distances approach where we propose a new technique to locate feature points and the wrinkles approach where we propose a new method to locate the wrinkles regions in the face. The classification rates given by experimental results show the effectiveness of our proposed approach compared to other methods.

Data science becomes creative with generative adversarial networks (GANs) which have had a big success since they were introduced in 2014 by Ian J. Goodfellow and co-authors. In technical term the GANs are based on the unsupervised learning of two artificial neural networks called Generator and Discriminator both trained under the adversarial learning idea. The major goal of GAN is to generate new samples that estimate the potential distribution of real data. Due to its huge success, many modified versions have been proposed in the last two years. We summarize in this review paper GAN’s background, architecture and its application fields. Then, we discuss the different extensions of GAN over the original model and provide a comparative analysis of these techniques.

Fast mesh compression is becoming a requisite in several applications such as medical imaging and video games. Graphics Processing Units (GPUs) are recently becoming massively parallel devices for Single Instruction, Multiple Data (SIMD) computing, addressing hence greater implementation challenges. Transformation and Quantization (TQ) is considered the second highest workload part of the wavelet-based mesh coding. Therefore, its acceleration will further improve the overall processing speed of the coding. In this paper, an OpenCL (Open Computing Language) acceleration of TQ is proposed. The Butterfly Wavelet Transform (BWT) based on the unlifted scheme is adopted in the transformation method while the embedded deadzone quantization is employed for the wavelet quantization. A chunk rearrangement process is applied for the computation of the neighborhood information needed for the Butterfly subdivision stencils. Accordingly, every chunk proceeds independently the prediction of the wavelet coefficients and their quantization. The key insights behind the proposed TQ method on GPU are a smart memory management and an efficient memory data mapping. Extensive experimental assessments demonstrate the effectiveness of our GPU implementation in terms of memory and runtime costs while preserving the rate distortion performance of the state-ofthe-art Bitplane coder.

Classical signal representation techniques generally use a description of the components on a basis on which the representation of the signal is unique such as wavelets network. Conversely, sparse representations consist in the decomposition of the signal on a dictionary comprising a number of elements much larger than the dimension of the signal. This technique can be widely used for representation, compression, denoising and separation of all types of signals. Consequently, some researches have confirmed that the use of a predefined dictionary is less efficient than a dictionary from training data. So, the idea of this paper is to propose a new technique for the creation of a dictionary using the wavelet decomposition to enhance the sparse representation of images. This technique is based on the combination of sparse coding and the fast wavelet transform algorithms for image representation. Our results obtained using different universal image databases showed greater performances in the representation of images when compared to some methods from the state of the art.

Image classification is an area where deep learning and especially stacked Auto-encoders have really proven their strength. The contributions of this paper lie in the creation of a new classifier to remedy some classification problems. This new method of classification presents a combination of the most used techniques in Deep Learning (DL) and Sparse Coding (SC) in the field of classification. Proposed deep neural networks consist of three stacked Auto-encoders and a Softmax used as an outer layer for classification. The first Auto-encoder is created from a sparse representation of all images of the dataset. The sparse representation of all images represents the decoder part of the first Auto-encoder. Then the transpose of the matrix is applied to get the encoder part. Experiments performed on standard datasets such as ImageNet and the Coil-100 reveal the efficacy of this approach.

This paper introduces a novel approach for human activities recognition (HAR) based on body articulations (joints) that represent the connection between bones in the human body which join the skeletal system such as the knee, shoulder and hand, and which are made to allow different degrees and types of movement. To implement our system, we used PoseNet to extract articulation points, which will be classified employing transfer learning approach to recognize the activity. The created system will be named in the rest of the paper (PTLHAR). The experimental results show that the proposed approach provides a significant improvement over state-of-the-art methods.

One of the major challenges in mobile networks and digital technologies is maintaining the security of real time data. In this regard, the research community developed a lot of works to fulfill this goal by proposing secure image encryption algorithms. However, some of these encryption schemes are not secure enough and lack robustness and security. In this paper, we succeed to reveal the weaknesses of a recently published encryption algorithm that is supposed to be secure and robust. We found that although the proposed network is unable to decrypt the ciphered image, it is able to perform classification on this image. We succeeded to build a deep neural network that can recognize encrypted images with an accuracy of 95.8%. Results demonstrate that our proposed approach is efficient for classifying ciphered images. These results could be valuable for further works into the topic of cryptanalysis using deep learning.

Wireless Sensor Networks (WSNs) are scalable research domain with a multitude of application contexts. Sensor nodes deployment is a decisive step that has a major impact on the performance of the network, since it directly influences the cost, the sensing capability and even the WSNs lifetimes. In this paper, we are interested in the placement problem of sensor nodes for WSNs. First, the issue is formulated as constrained multi-objective optimization problem (MOOP). Then, a novel approach based on Multi-Objective Flower Pollination Algorithm (MOFPA) was proposed. This new method aimed to approximate optimal trade-offs among multiple objective functions, which are enhancing the coverage, reducing the network energy dissipation, maximizing the network lifetime and maintaining the connectivity. Finally, we compared the proposed approach with two popular algorithms, namely, the classic Particle Swarm Optimization (PSO) and Non-dominated Sorting Genetic Algorithm II (NSGA-II). The simulation experiments show that our approach outperforms PSO and NSGA-II.

Device-to-Device communications underlaying cellular networks have great potential to provide a significant improvement in network performance. It allows better resource utilization, increases user throughput and extends the battery life of user equipment compared the conventional cellular network. The fundamental challenge in D2D communications is how to allocate spectral bands and control transmission power optimally, in order to ensure the quality of service (QoS) and increase the energy efficiency. This problem is mixed integer non-linear programming (MINLP) due to the discrete variables. In this paper, we consider the joint power control and channel allocation for underlying D2D communications. We formulated the optimization problem in the form of an iterative combinatorial auction. The proposed algorithm is designed with two rounds. In the first round, each bidder submits its offer for each channel. In the second round, the goal is to improve the results obtained in the first round by considering the interference received from other users. The system power consumption and network lifetime were evaluated through simulations.

Many techniques are used to solve problems of security in the Internet such as cryptography or watermarking. In this context watermarking is a way for protecting copyright and proving authenticity of a digital data. In this paper, a non blind digital watermark scheme is proposed. It is based on Discrete Cosine transformation (DCT), singular Values Decomposition (SVD) and Beta Chaotic Map (BCM). The experimental results show that this scheme is robust against several attacks compared to other algorithms.

In the multimedia processing, the extraction and the representation of characteristics are considered as an important step. The extraction of the ideal characteristics having the ability to reflect the intrinsic content of the images as complete as possible is still a difficult problem in computer vision. Little research has focused on this problem. This paper presents a new unsupervised method based on the self-organizing map (SOM) for features extraction. Our method consists of two main steps: Extracting the sub-regions of an image according to their points of interest and using the SOM for the different views of an image such as color, texture and shape. Then, combine them to have finally a "Multi-View" vector characteristic. The proposed method is evaluated on three image classification datasets Cloud, Coil100 and CIFAR-10. The classification accuracies of the proposed method for the three datasets are much higher in comparison with the other methods cited in the literature.

Wireless Mesh Network is a recent network paradigm to provide broadband Internet services for end users. Different from the traditional wireless mobile network, WMN is based on multi-hop structure. The mesh routers (MRs) act as access points for the mesh clients and connect via multi-hops to Internet gateways (IGs) by directing all their Internet traffic to these gateways. We aim to determine the optimum location of gateways, by minimizing the variance of MR-IG hop count and the number of deployed gateways. Guided by a mathematical model, we propose a multi-objective approach based on genetic algorithm. The simulation results highlights the performance of our approach in terms of operating cost and communication delay.

As a powerful technique, sparse coding was adopted by several researchers in different approaches. Particularly in image processing, it has attracted a considerable attention. It can be widely used for representation, compression, denoising and separation of all type of signal. Recent works have confirmed that the use of a predefined dictionary is less efficient than a dictionary from training data. According to this idea, this paper proposes a new technique based on wavelet network to create a dictionary to ameliorate the representation and classification of image using sparse coding technique.

The wireless sensor networks (WSN) consist of a set of sensors that are more and more used in surveillance applications on a large scale in different areas: military, Environment, Health ... etc. Despite the minimization and the reduction of the manufacturing costs of the sensors, they can operate in places difficult to access without the possibility of reloading of battery, they generally have limited resources in terms of power of emission, of processing capacity, data storage and energy. These sensors can be used in a hostile environment, such as, for example, on a field of battle, in the presence of fires, floods, earthquakes. In these environments the sensors can fail, even in a normal operation. It is therefore necessary to develop algorithms tolerant and detection of defects of the nodes for the network of sensor without wires, therefore, the faults of the sensor can reduce the quality of the surveillance if they are not detected. The values that are measured by the sensors are used to estimate the state of the monitored area. We used the Non-linear Auto- Regressive with eXogeneous (NARX), the recursive architecture of the neural network, to predict the state of a node of a sensor from the previous values described by the functions of time series. The experimental results have verified that the prediction of the State is enhanced by our proposed model.

To survive the competition, companies always think about having the best employees. The selection is depended on the answers to the questions of the interviewer and the behavior of the candidate during the interview session. The study of this behavior is always based on a psychological analysis of the movements accompanying the answers and discussions. Few techniques are proposed until today to analyze automatically candidate’s non verbal behavior. This paper is a part of a work psychology recognition system; it concentrates in spontaneous hand gesture which is very significant in interviews according to psychologists. We propose motion history representation of hand based on an hybrid approach that merges optical flow and history motion images. The optical flow technique is used firstly to detect hand motions in each frame of a video sequence. Secondly, we use the history motion images (HMI) to accumulate the output of the optical flow in order to have finally a good representation of the hand‘s local movement in a global temporal template.

The video surveillance is one of the key areas in computer vision researches. The scientific challenge in this field involves the implementation of automatic systems to obtain detailed information about individuals and groups behaviors. In particular, the detection of abnormal movements of groups or individuals requires a fine analysis of frames in the video stream. In this article, we propose a new method to detect anomalies in crowded scenes. We try to categorize the video in a supervised mode accompanied by unsupervised learning using the principle of the autoencoder. In order to construct an informative concept for the recognition of these behaviors, we use a technique of representation based on the superposition of human silhouettes. The evaluation of the UMN dataset demonstrates the effectiveness of the proposed approach.

Deep Convolutional Neural Network (DCNN) can be marked as a powerful tool for object and image classification. However, the training stage of such networks is highly consuming in terms of storage space and time. Also, the optimization is still a challenging subject. In this paper, we propose a fast DCNN based on smart dropout and layer skipping. The proposed approach led to improve the speed of the testing stage as well as image classification accuracy. This was possible thanks to three key advantages: First, the rapid way to compute the features using Fast Beta Wavelet Transform. Second, the proposed intelligent dropout method is based on whether or not a unit is efficiently and not randomly selected. Third, it is possible to classify the image using efficient units of earlier layer(s) and skipping all the subsequent hidden layers directly to the output layer. Our experiments were performed on CIFAR-10 and MNIST datasets and the obtained results are very promising.

Device-to-device (D2D) multicast transmission is an important feature for group-oriented applications. In this paper, we propose a joint channel allocation and power control scheme for the D2D multicast underlay communications. In single-rate multicast, the achieved data rate is determined by the weakest link. Therefore, we formulate this MINLP problem as a maximin optimization problem to guarantee fairness of different multicast groups. Each D2D group can reuse the channel of all cellular users. In order to limit the impact of D2D communications on cellular users' quality of service (QoS), we define an SINR threshold value for the cellular users. The non-linear Perron-Frobenius theory is used to derive a deterministic algorithm for solving the non-convex non-linear fair power control problem. A novel-genetic-algorithm-aided efficient scheme is then proposed to solve the combinatory issue of the joint channel allocation and power control. The proposed scheme is evaluated by using extensive simulations. Numerical results highlight the performance of our approach in terms of sum rate fairness.

Automatic identification of television programs in the TV stream is an important task for operating archives and represent a principal source of multimedia information.. The goal of the proposed approach is to enable a better exploitation of this source of video by multimedia services (i.e., TV-On-Demand, catch-up TV), social community, and video-sharing pla forms (Vimeo, Youtube, Facebook...) This paper presents a new spatio-temporal approach to identify the programs in TV stream using deep learning in two main steps. A database for video of visual jingles is constructed for training. In the test we use same jingles program type in order to identify the various program types in the TV stream. The main idea of identification process consists in using the principal of auto-encoder. After presenting the proposed approach, the paper overviews the encouraging experimental results on several streams extracted from different channels and composed of several programs. Comparison experiments to similar works have been carried out on the TRECVID 2017 database. We show significant improvements to TV programs identification exceed 95 %.

In the medical field, research studies have shown that Augmented Reality (AR) based surgical training has a good potential in making the learning process more active. However, lack of displaying the correct occlusion between the real surgical instrument and virtual organ greatly limits the trainee surgeons understanding and reduces the overall system usability. In this paper, we propose a novel mutual occlusion handling method based on surgical instrument tracking and 3D positioning approach in AR environment. Therefore, we introduce a monocular image processing based paradigm that aims at (1) tracking the instrument using both background subtraction and Hough transform method (2) calculating the 3D position of the instrument using the geometry of perspective projection (3) comparing the 3D coordinates of the real instrument with the virtual organ to achieve a realistic AR rendering system. The experimental results show that our approach is highly accurate and can handle the mutual occlusion automatically in real time.

Copy-move image forgery is the act of cloning a restricted region in the image and pasting it once or multiple times within that same image. This procedure intends to cover a certain feature, probably a person or an object, in the processed image or emphasize it through duplication. Consequences of this malicious operation can be unexpectedly harmful. Hence, the present paper proposes a new approach that automatically detects Copy-move Forgery (CMF). In particular, this work broaches a widely common open issue in CMF research literature that is detecting CMF within smooth areas. Indeed, the proposed approach represents the image blocks as a sparse linear combination of pre-learned bases (a mixture of texture and color-wise small patches) which allows a robust description of smooth patches. The reported experimental results demonstrate the effectiveness of the proposed approach in identifying the forged regions in CM attacks.

This paper presents a new technique for detection of dental caries that is a bacterial disease that destroys the tooth structure. In our approach, we have achieved a new segmentation method that combines the advantages of fuzzy C mean algorithm and level set method. The results obtained by the FCM algorithm will be used by Level sets algorithm to reduce the influence of the noise effect on the working of each of these algorithms, to facilitate level sets manipulation and to lead to more robust segmentation. The sensitivity and specificity confirm the effectiveness of proposed method for caries detection.

Currently, there are several fall detection systems based on video analysis. However, these systems have not yet reached the desired level of appropriateness and robustness. To reduce the risk of falling in insecure environments, a new method is developed in this paper to detect and predict human fall detection. We adopt, in this approach, a Block Matching motion estimation algorithm based on acceleration and changes of the human body silhouette area, which are obtained from a single surveillance camera. It presents an algorithm to accelerate the fall detection system on based on a local adjustment of the velocity field.

Automatic identification of television programs in the TV stream is an important task for operating archives. This article proposes a new spatio-temporal approach to identify the programs in TV stream into two main steps: First, a reference catalogue for video features visual jingles built. We operate the features that characterize the instances of the same program type to identify the different types of programs in the flow of television. The role of video features is to represent the visual invariants for each visual jingle using appropriate automatic descriptors for each television program. On the other hand, programs in television streams are identified by examining the similarity of the video signal for visual grammars in the catalogue. The main idea of the identification process is to compare the visual similarity of the video signal features in the flow of television to the catalogue. After presenting the proposed approach, the paper overviews encouraging experimental results on several streams extracted from different channels and compounds of several programs.

In this paper, we proposed a deep self-organizing map model (Deep-SOMs) for automated features extracting and learning from big data streaming which we benefit from the framework Spark for real time streams and highly parallel data processing. The SOMs deep architecture is based on the notion of abstraction (patterns automatically extract from the raw data, from the less to more abstract). The proposed model consists of three hidden self-organizing layers, an input and an output layer. Each layer is made up of a multitude of SOMs, each map only focusing at local headmistress sub-region from the input image. Then, each layer trains the local information to generate more overall information in the higher layer. The proposed Deep-SOMs model is unique in terms of the layers architecture, the SOMs sampling method and learning. During the learning stage we use a set of unsupervised SOMs for feature extraction. We validate the effectiveness of our approach on large data sets such as Leukemia dataset and SRBCT. Results of comparison have shown that the Deep-SOMs model performs better than many existing algorithms for images classification.

Extreme learning Machine is a well known learning algorithm in the field of machine learning. It's about a feed forward neural network with a single-hidden layer. It is an extremely fast learning algorithm with good generalization performance. In this paper, we aim to compare the Extreme learning Machine with wavelet neural networks, which is a very used algorithm. We have used six benchmark data sets to evaluate each technique. These datasets Including Wisconsin Breast Cancer, Glass Identification, Ionosphere, Pima Indians Diabetes, Wine Recognition and Iris Plant. Experimental results have shown that both extreme learning machine and wavelet neural networks have reached good results.

Color is a significant visual characteristic for both human vision and computer processing. Global color descriptors characterize an image by its color distribution or histogram, and discard information about object location as well as content of different colors. In this paper, we proposed a local color descriptor based on indexed matrix wavelet analysis and fuzzy decision support system (FDSS), which we called “Fuzzy Indexed Color (FIC)”. First, the indexed map of each image is analysed using Fast wavelet transform to capture the most relevant color feature content for each color channel R, G and B. Then, a FDSS is proposed for image matching in order to get more flexibility and reliability in making decision. The proposed FIC was evaluated using Google color, ebay data, Soccer and Flower datasets; and the results are very promising.

Developing an automatic arabic sign language recognition system is of great importance, it can be used as a communication means between hearing-impaired and other people.

The success of the deep learning and specifically learning layer by layer led to many impressive results in several contexts that include neural network. This gave us the idea to apply this principle of learning on wavelet network because it is an active research topic at the moment. This paper present our approach for image classification by the combination of two techniques of learning: the wavelet network and the deep learning. We try to classify images in a supervised way following by an unsupervised learning using the principle of autoencoder. Experiments on two databases COIL-100 and MNIST show that our approach gives good results for the two classifiers that we used.

Recent research on image encryption algorithms has been increasingly based on chaotic systems, but the drawbacks of small key space and weak security in chaotic cryptosystems are obvious. In this paper, new chaotic maps based on beta function were created. The generation of different pseudo random sequences was carried out to shuffle the position of the image pixels and to confuse the relationship between the encrypted image and the original image, thereby significantly increasing the resistance to attacks. The proposed system has the advantage of high security analysis such as key space, statistical and sensitivity analysis.

Face detection has been one of the most studied topics in the computer vision literature due to its relevant role in applications such as video surveillance, human computer interface and face image database management. Here, we will present a face detection approach which contains two steps. The first step is training phase based on Adaboost algorithm. The second step is the detection phase. The proposed approach presents an enhancement of Viola and Jones’ algorithm by replacing Haar descriptors with Beta wavelet. The obtained results have proved an excellent performance of detection not only when a face is in front of the camera but also when it is oriented towards the right or the left. Moreover, thanks to the start period needed for the detection, our approach can be applied during a real time experience.

This paper presents a novel hand posture recognizer based on separator wavelet networks (SWNs). Aiming at creating a robust and rapid hand posture recognizer, we have contributed by proposing a new training algorithm for the wavelet network classifier based on fast wavelet transform (FWN). So, the contribution resides in reducing the number of WNs modeling training data. To make that, inspiring from the adaboost feature selection method, we thought to create SWNs (n-1 WNs for n classes) instead of modeling each training sample by its wavelet network (WN). By proposing the new training algorithm, the recognition phase will be positively influenced. It will be more rapid thanks to the reduction of the number of comparisons between test images WNs and training WNs. Comparisons with other works, employing universal hand posture datasets are presented and discussed. Obtained results have shown that the new hand posture recognizer is comparable to previously established ones.

Human gait is an attractive modality for recognizing people at a distance. Gait recognition systems aims to identify people by studying their manner of walking. In this paper, we contribute by the creation of a new approach for gait recognition based on fast wavelet network (FWN) classifier. To guaranty the effectiveness of our gait recognizer, we have employed both static and dynamic gait characteristics. So, to extract the static features (dimension of the body part), model based method was employed. Thus, for the dynamic features (silhouette appearance and motion), model free method was used. The combination of these two methods aims at strengthens the WN classification results. Experimental results employing universal datasets show that our new gait recognizer performs better than already established ones.

This work fits into the context of the interpretation of automatic gestures based on computer vision. The aim of our work is to transform a conventional screen in a surface that allows the user to use his hands as pointing devices. These can be summarized in three main steps. Hand detection in a video, monitoring detected hands and conversion paths made by the hands to computer commands. To realize this application, it is necessary to detect the hand to follow. A classification phase is essential, at the control part. For this reason, we resorted to the use of a neuro-fuzzy classifier for classification and a pattern matching method for detection.

In last years, the emergence of 3D shape in face recognition is due to its robustness to pose and illumination changes. These attractive benefits are not all the challenges to achieve satisfactory recognition rate. Other challenges such as facial expressions and computing time of matching algorithms remain to be explored. In this context, we propose our 3D face recognition approach using 3D wavelet networks. Our approach contains two stages: learning stage and recognition stage. For the training we propose a novel algorithm based on 3D fast wavelet transform. From 3D coordinates of the face (x,y,z), we proceed to voxelization to get a 3D volume which will be decomposed by 3D fast wavelet transform and modeled after that with a wavelet network, then their associated weights are considered as vector features to represent each training face . For the recognition stage, an unknown identity face is projected on all the training WN to obtain a new vector features after every projection. A similarity score is computed between the old and the obtained vector features. To show the efficiency of our approach, experimental results were performed on all the FRGC v.2 benchmark.

Driving security is an important task for human society. The major challenge in the field of accident avoidance systems is the driver vigilance monitoring. The lack of vigilance can be noticed by various ways, such as, fatigue, drowsiness and distraction. Hence, the need of a reliable driver’s vigilance decrease detection system which can alert drivers before a mishap happens. In this paper, we present a novel approach for vigilance estimation based on multilevel system by combining head movement analysis and eyes blinking. We have used Viola and Jones algorithm to analyse head movement and a classification system using wavelet networks for eyelid closure measuring. The contribution of our application is classifiying the vigilance state at multi level. This is different from the binary-class (awakening or hypovigilant state) existing in most popular systems.

Image classification is an important task within the field of computer vision. In this paper we propose a new wavelet network classifier (WNC) based on the cascaded architecture. This classifier is characterized by its new learning approach and its novel architecture which brings a novel robust test way. So, our contributions in this paper reside in two major points. The first one is the proposition of a new training algorithm which overcomes lacuna detected in the latest version of WN learning approach. Hence, our new approach creates separator WNs discriminating classes (n -- 1 WNs to classify n classes) instead of creating a WN for each training image. This contribution makes very rapid the classification process by reducing the number of comparisons between test images WNs and training WNs. The second contribution is the proposition of a novel architecture which brings a new test approach radically different to those employed in ancient WN versions. By the new architecture which is based on the cascade notion, we aim at reducing the number of kernels employed in the approximation of test images. Experiments, using well known benchmarks, show that our new classifier is very robust and rapid compared to already existing ones.

The problem addressed in this paper is the classification and recognition of Arabic handwritten characters. As a solution, we present a Neural Network (NN) architecture based on Fast Wavelet Transform (FWT) and Adaboost algorithm. FWT is used to extract character's features, based on Multi-Resolution Analysis (MRA) at different levels of abstraction. These features are used to calculate inputs of hidden layer. After this first step, the features are filtered, using Adaboost algorithm, to select the best corresponding ones to each shape of input characters. The reported results are tested on Arabic handwritten characters dataset with 6000 characters. The classification rate for the different groups of characters are 93.92%. Additionally, the speed of the classification algorithm is tested and reported.

This paper gives a review of the deep learning history and proposes a new approach to supervised image classification by the combination of two techniques of learning: the wavelet network and the deep learning. This new approach consists of performing the classification of one class versus all the other classes of the dataset by the reconstruction of a convolutional deep neural wavelet network. This network is obtained using a series of stacked auto-encoders and a linear classifier. The experimental test of our approach performed on "COIL-100" dataset demonstrates that our model is remarkably efficient for image classification compared to a known classifier.

Image data is of immense practical importance in medical informatics. In teeth-related radiograph research, the information of teeth shape is the most critical factor for achieving highly automated diagnosis. Automated image segmentation, which aims at automated extraction of region boundary features, plays a fundamental role in understanding image content for searching and mining in medical image. Therefore, accurate segmentation is an essential but difficult task due to low contrast between regions of interest and uneven exposure of the dental X-ray image. To address this problem, several segmentation approaches have been proposed in the literature, with many of them providing rather promising results. In this paper, we will look at a model by Chan-Vese that detects objects not defined by gradient. We will then implement this algorithm on the GPU and see what kind of speedup we can get compared to serial CPU implementations. Finally we will quantity our results as well as make a qualitative evaluation of the method with respect to how it performs for segmenting medical images.

Speech recognition is a specialized pattern recognition task with several applications such as vocal command system, dictating machines, and understanding systems. In recent years, research on pattern recognition has increased by developing various methods and algorithms for different applications. In this paper, we proposed a novel training algorithm based on the fast Beta wavelet transform for speech recognition. This approach has many advantages compared to other algorithms. The majority of the old approaches need to inverse matrix, which can be computationally intensive. However, the new algorithm is computed by the iterative application of fast wavelet transform to compute connection weights. To highlight our approach, we compared its experimental results to those of the old ones.

In this paper we invoke a new approach for the multi-objective routing problems in Wireless Sensor Networks (WSNs). Our approach improves more than one single Quality of Services (QoS) exigency such as energy consumption and delay. However, the classical routing protocols in conventional network optimize a single objective or QoS parameters. The proposed approach adapted a multi-objective evolutionary algorithm (MOEA), specifically, the improved Strength Pareto Evolutionary Algorithm (SPEA2) in order to improve the QoS in WSNs. Our simulation results show that the SPEA2 algorithms are efficient in solving routing problems and are capable of finding the Pareto optimal Set. Additionally, we demonstrate that this approach provides better trade-off solutions in comparison to the classical routing protocols.

In this paper, a new approach to supervised image classification is suggested. It's conducted by the combination of two techniques of learning: the wavelet network and the deep learning. This new approach consists of performing the classification of one class versus all the other classes of the dataset by the reconstruction of a convolutional deep neural wavelet network. This network is obtained using a series of stacked auto-encoders and a linear classifier. Finally, a local contrast normalization and an intelligent pooling are applied to our network. The experimental test of our approach performed on Arabic Printed Text Image (APTI) dataset demonstrates that our model is remarkably efficient for image classification compared to a known classifier.

This paper presents an emotion recognition system based on Beta wavelet network using the Fast Wavelet Transform in order to improve the performance of this network. The proposed system can be summarized in two main steps: training stage and classification stage. Comparing with many algorithms which suffer from the low classification rates and the long executing time the rates given by our experimental results show the effectiveness of the FWT.

Face emotion recognition is one of the most important and rapidly advanced active research areas of computer science. A new method for facial expression recognition based on wavelet network classifier is proposed in this paper. It allows us the detection of six basic emotions other than the neutral one: (Joy, surprise, sadness, anger, fear and disgust) The process is composed of three principle steps: face detection, features extraction and classification. The effectiveness of our proposed algorithm is experimentally demonstrated by using well-known test database: the extended cohen-kanade database.

MapReduce has become a dominant parallel computing paradigm for storing and processing massive data due to its excellent scalability, reliability, and elasticity. In this paper, we present a new architecture of Distributed Beta Wavelet Networks {DBWN} for large image classification in MapReduce model. First to prove the performance of wavelet networks, a parallelized learning algorithm based on the Beta Wavelet Transform is proposed. Then the proposed structure of the {DBWN} is itemized. However the new algorithm is realized in MapReduce model. Comparisons with Fast Beta Wavelet Network {FBWN} are presented and discussed. Results of comparison have shown that the {DBWN} model performs better than {FBWN} model in classification rate and in the context of training run time.

Road safety is an important challenge for human society. Fatigue at the wheel is a serious problem causing thousands of road accidents each year. The major issue in the field of driving security systems is driver vigilance control. In this paper, we present a new method for driver's vigilance level measurement using a multi parameter system based on head movement estimation, eyes blinking analysis and face emotion recognition based on wavelet networks classification system.

This paper focuses on the issue of emotion recognition. It describes an emotion recognition system based on facial expression which contains four steps: detection of face's elements, localization of feature points, their tracking during a movie and facial expression classification. The first step is realized by the famous Viola and Jones algorithm. To localize feature points we have developed an automatic and easy method. To track them we used the optical flow. Finally the classification step is based on wavelet network using Fast Wavelet Transform FWT. The experimental results demonstrated the efficiency of our system.

This work is in the field of human-computer communication, namely in the field of gestural communication. The objective was to develop a system for gesture recognition. This system will be used to control a computer without a keyboard. The idea consists in using a visual panel printed on an ordinary paper to communicate with a computer.

This paper aims at developing a novel approach for speech recognition based on wavelet network learnt by fast wavelet transform (FWN) including a fuzzy decision support system (FDSS). Our contributions reside in, first, proposing a novel learning algorithm for speech recognition based on the fast wavelet transform (FWT) which has many advantages compared to other algorithms and in which major problems of the previous works to compute connection weights were solved. They were determined by a direct solution which requires computing matrix inversion, which may be intensive. However, the new algorithm was realized by the iterative application of FWT to compute connection weights. Second, proposing a new classification way for this speech recognition system. It operated a human reasoning mode employing a FDSS to compute similarity degrees between test and training signals. Extensive empirical experiments were conducted to compare the proposed approach with other approaches. Obtained results show that the new speech recognition system has a better performance than previously established ones.

This paper presents a new cascaded hybrid Wavelet Network Classifier (CHWNC) designed for hand gesture recognition in real time applications. This paper contains two key contributions. The first is the amelioration of our previous works in the classification domain employing wavelet networks (WN). Precisely, by ameliorating the training way of the latest wavelet network classifier (WNC) version by representing each training class by one WN instead of creating a WN for each training image. This contribution makes very rapid the test phase by reducing the number of comparisons between test images WNs and training WNs. The second contribution is the proposition of a new wavelet network architecture including the cascade notion which decomposes the WN on a set of stages. The new architecture has as aim not only to make recognitions robust and rapid but also to reject as fast as possible gestures which must not be considered by the system (spontaneous gestures). Experiments, based on a well known hand posture dataset, show that our method is very robust and rapid compared to already existing ones.

This paper presents a virtual object control method of augmented reality scene. We have based on control approach on speech recognition. The idea came from human-machine interaction. The speech recognition system is based on wavelet network. In this paper, we have briefly described the used toolkit to do with the augmented reality. Then, we present the speech recognition approach the training and recognition approach. Finally, we present the results.

Elastic Bunch Graph Matching EBGM is a face recognition algorithm that is distributed with CSU's Evaluation of Face Recognition Algorithms System. The algorithm recognizes novel faces by first localizing a set of landmark features and then measuring similarity between these features. Both localization and comparison uses Gabor jets extracted at landmark positions. In order to improve the performance of the face recognition system[7][8], we have associated Beta filters to the EBGM technique. This choice of Beta filters is advanced by the performance of these functions in many applications of classification and pattern recognition.

In this paper, a novel learning algorithm of wavelet networks based on the Fast Wavelet Transform (FWT) is proposed. It has many advantages compared to other algorithms, in which we solve the problem in previous works, when the weights of the hidden layer to the output layer are determined by applying the back propagation algorithm or by direct solution which requires to compute the matrix inversion, this may cause intensive computation when the learning data is too large. However, the new algorithm is realized by iterative application of FWT to compute the connection weights. Furthermore, we have extended the novel learning algorithm by using Levenberg–Marquardt method to optimize the learning functions. The experimental results have demonstrated that our model is remarkably more refreshing than some of the previously established models in terms of both speed and efficiency.

Image classification is an important task in computer vision. In this paper, we propose a supervised method for image classification based on a fast beta wavelet networks (FBWN) model. First, the structure of the wavelet network is detailed. Then, to enhance the performance of wavelet networks, a novel learning algorithm based on the Fast Wavelet Transform (FWTLA) is proposed. It has many advantages compared to other algorithms, in which we solve the problem of the previous works, when the weights of the hidden layer to the output layer are determinate by applying the back propagation algorithm or by direct solution which requires to compute matrix inversion, this may be intensive computation when the learning data is too large. However, the new algorithm is realized by the iterative application of FWT to compute connection weights. In the simulation part, the proposed method is employed to classify images. Comparisons with classical wavelet network classifier are presented and discussed. Results of comparison have shown that the FBWN model performs better than the previously established model in the context of training run time and classification rate.

3D shape of face has recently emerged as a major research in face biometrics. However, while it is reputed to be relatively invariant to lighting conditions and pose, one still needs to cope with facial expression variations for a reliable face recognition solution and running time of the matching algorithms for fast identification software. We present in this paper our solutions to overcome these limitations. We propose a new method of 3D facial recognition based on wavelet networks. Firstly, depth image is preprocessed in order to crop the useful area of the face image. Secondly, a compact and representative biometric signature is produced by means of wavelet networks. Finally, the matching of two faces is made by computing Euclidean distance between their two corresponding signatures. To show the efficiency and accuracy of our approach, a subset taken from FRGC v2 dataset is used to made evaluations.