Search

Solar panels typically consist of photovoltaic (PV) cells covered by a protective glass coating, which generate electricity when subjected to radiation. However, the capability of electricity gener...

International audience; Illumination, noise, and changes of environments, scales negatively impact on encoding chaotic motions for dynamic texture (DT) representation. This paper proposes a new method to overcome those issues by addressing the following novel concepts. First, different Gaussian-based kernels are taken into account as an effective filtered pre-processing with low computational cost to point out robust and invariant features. Second, a discriminative operator, named Local Rubik-based Pattern (LRP), is introduced to adequately capture both shape and motion cues of DTs by proposing a new concept of complemented components together with an effective encoding method. In addition, it also addresses a novel thresholding to take into account rich spatio-temporal relationships extracted from a new model of neighborhood supporting region. Finally, an efficient framework for DT description is presented by exploiting operator LRP for encoding various instances of Gaussian-based volumes in order to form a robust descriptor against noise, changes of illumination, scale, and environment. Experiments for DT classification on benchmark datasets have authenticated the interest of our proposal.

Background Predictive patient stratification is greatly emerging, because it allows us to prospectively identify which patients will benefit from what interventions before their condition worsens. In the biomedical research, a number of stratification methods have been successfully applied and have assisted treatment process. Because of heterogeneity and complexity of medical data, it is very challenging to integrate them and make use of them in practical clinic. There are two major challenges of data integration. Firstly, since the biomedical data has a high number of dimensions, combining multiple data leads to the hard problem of vast dimensional space handling. The computation is enormously complex and time-consuming. Secondly, the disparity of different data types causes another critical problem in machine learning for biomedical data. It has a great need to develop an efficient machine learning framework to handle the challenges. Methods In this paper, we propose a fast-multiple kernel learning framework, referred to as fMKL-DR, that optimise equations to calculate matrix chain multiplication and reduce dimensions in data space. We applied our framework to two case studies, Alzheimer’s disease (AD) patient stratification and cancer patient stratification. We performed several comparative evaluations on various biomedical datasets. Results In the case study of AD patients, we enhanced significantly the multiple-ROIs approach based on MRI image data. The method could successfully classify not only AD patients and non-AD patients but also different phases of AD patients with AUC close to 1. In the case study of cancer patients, the framework was applied to six types of cancers, i.e., glioblastoma multiforme cancer, ovarian cancer, lung cancer, breast cancer, kidney cancer, and liver cancer. We efficiently integrated gene expression, miRNA expression, and DNA methylation. The results showed that the classification model basing on integrated datasets was much more accurate than classification model basing on the single data type. Conclusions The results demonstrated that the fMKL-DR remarkably improves computational cost and accuracy for both AD patient and cancer patient stratification. We optimised the data integration, dimension reduction, and kernel fusion. Our framework has great potential for mining large-scale cohort data and aiding personalised prevention.

Representation of dynamic textures (DTs), well-known as a sequence of moving textures, is a challenging problem in video analysis due to the disorientation of motion features. Analysing DTs to make them `understandable' plays an important role in different applications of computer vision. In this study, an efficient approach for DT description is proposed by addressing the following novel concepts. First, the beneficial properties of dense trajectories are exploited for the first time to efficiently describe DTs instead of the whole video. Second, two substantial extensions of local vector pattern operator are introduced to form a completed model which is based on complemented components to enhance its performance in encoding directional features of motion points in a trajectory. Finally, the authors present a new framework, called directional dense trajectory patterns, which takes advantage of directional beams of dense trajectories along with spatio-temporal features of their motion points in order to construct dense-trajectory-based descriptors with more robustness. Evaluations of DT recognition on different benchmark datasets (i.e. UCLA, DynTex, and DynTex++) have verified the interest of the authors' proposal.

Efficiently capturing shape and turbulent motions of dynamic textures (DTs) for video description is a challenge in real applications due to the negative influences of the well-known problems: environmental elements, illumination, scale, and noise. In this paper, we propose an efficient and simple framework for DT representation based on the oriented features of high-order Gaussian gradients. Firstly, 2D/3D Gaussian-based filtering kernels in high-order partial derivatives are taken into account the video analysis as a preprocessing step to obtain corresponding gradient-filtered images/volumes. After that, the oriented features, which are robust against the above issues, are extracted by decomposing the Gaussian derivative magnitudes into oriented components. Finally, a shallow local encoding is utilized for structuring spatio-temporal features from these oriented magnitudes. This allows constructing discriminative descriptors with promising performances compared to those based on the non-oriented ones. Experimental results for DT classification task on benchmark datasets have verified the interest of our proposal.

Hand gestures have been shown to be an efficient way for human-machine interaction. Existing approaches usually utilize ambient or head/chest-mounted cameras to capture hand images. This paper presents a new way to capture hand gestures using the wrist-worn camera. The wrist-worn device is designed as a watch with an integrated camera that is much easier and comfortable to wear in daily life context. We then collect a dataset of ten hand postures using the designed prototype by ten subjects. In addition, we deploy state-of-the-art lite CNN models (YOLO family, Single Shot Detector-SSD) as posture detectors and classifiers. Experimental results show that with limited camera angles, the postures are highly distinctive and easily discriminated with the highest performance of 98.85% and 97.40% in terms of precision and recall, which motivates a wide range of applications and new research directions for human-machine interaction, wearables, the Internet of Things (IoT) and so on.

Efficiently representing spatio-temporal features of dynamic textures (DTs) in videos has been restricted due to negative impacts of the well-known issues of environmental changes, illumination, and noise. In order to mitigate those, this paper proposes a new approach for an efficient DT representation by addressing the following novel concepts. Firstly, a novel filtering kernel, called Difference of Derivative Gaussians (DoDG), is introduced for the first time based on high-order derivative of a Gaussian kernel. It allows to point out DoDG-based filtered outcomes which are prominently resistant to noise for DT representation compared to exploiting the conventional Difference of Gaussians (DoG). A new framework in low computational complexity is then presented to take DoDG into account video denoising as an effective preprocessing of DT encoding. Finally, a simple variant of Local Binary Patterns (LBPs) is addressed to extract local features from these DoDG-filtered outcomes for constructing discriminative DoDG-based descriptors in small dimension, expected as one of appreciated solutions for mobile applications. Experimental results for DT recognition have verified that our proposal significantly performs well compared to all non-deep-learning methods, while being very close to deep-learning approaches. Also, ours are eminently better than those based on the traditional DoG.

Background: Chronic urticaria is an allergic skin condition that significantly affects the quality of life. Skin prick test determines allergens in order to prevent recurrent urticaria. Materials and methods: 43 chronic urticaria patients visiting the Dermatology Clinic from 09/2017 to 09/2018 were tested for 16 allergens on the skin at the Immunology Department, Hue University of Medicine and Pharmacy. Results: 30 females and 13 males participated in the study (female/male 2.3/1). The average number of disease episodes was 3.1 ± 1.4 times; with the other allergic diseases accompanied by 41.9% of patients; the average severity score was 10.0 ± 2.0 with the serious condition accounting for 60.5%. 86.0% of patients had a positive skin prick test, of which the positive rate for respiratory allergens (mites with the highest rate was 34.9%) was higher than the food allergens (crab 30.2%). There is a correlation between positive skin prick results and atopic allergies. Conclusion: The majority of chronic urticaria patients were severe and had positive skin prick results, in which mites had the highest incidence of allergy. Key words: urticaria, chronic urticaria, skin prick test, allergen, allergic

We present a novel framework for efficient and robust facial feature representation based upon Local Binary Pattern (LBP), called Weighted Statistical Binary Pattern, wherein the descriptors utilize the straight-line topology along with different directions. The input image is initially divided into mean and variance moments. A new variance moment, which contains distinctive facial features, is prepared by extracting root k-th. Then, when Sign and Magnitude components along four different directions using the mean moment are constructed, a weighting approach according to the new variance is applied to each component. Finally, the weighted histograms of Sign and Magnitude components are concatenated to build a novel histogram of Complementary LBP along with different directions. A comprehensive evaluation using six public face datasets suggests that the present framework outperforms the state-of-the-art methods and achieves 98.51% for ORL, 98.72% for YALE, 98.83% for Caltech, 99.52% for AR, 94.78% for FERET, and 99.07% for KDEF in terms of accuracy, respectively. The influence of color spaces and the issue of degraded images are also analyzed with our descriptors. Such a result with theoretical underpinning confirms that our descriptors are robust against noise, illumination variation, diverse facial expressions, and head poses.

Vision based human detection and posture classification are essential components of many computer vision systems in security, advertisement and healthcare. Although human detection has been studied for more than three decades, most of proposed methods focused on standing people. In a daily-life applications, posture of human can vary strongly from standing, sitting, lying to crunching. In addition, human shape also vary according to different viewpoints. This makes more challenges for human detection and posture classification. Due to a smooth transition of postures, it is difficult to determine how many postures should be considered and classified. In this paper, we deploy an unsupervised technique to explore the number of distinctive human postures from a given set of activities. We then resolve the problem of postures classification as a multi-class detection problem using a state of the art convolutional neural network YOLO. The proposed method gives promising results on a dataset of human activities taken from six views. Recall and precision of human posture detection and classification are highly achieved on every viewpoint (99% of recall and precision). The posture classification leads to potential application for fall detection at high frame-rate.

In recent years, bioinformatics has been significantly contributing to patient stratification that is very crucial for early detection of cancer diseases. In particular, stratification or classification of patients is to divide patients into subgroups that will be offered effective treatment regimens. However, current methods have to face two major challenges in analyzing large biomedical datasets when stratifying cancer patients. Firstly, the datasets are very big with a high number of features. Secondly, because the public data is available and heterogeneous, there is a great need of combining multiple data sources, providing more comprehensive and informative datasets. A variety of methods has been proposed to tackle these challenges, but they have often solved one or the other separately. Handling noisy data encountered another difficulty in data integration. In this paper, we have proposed an efficient model, combining of the robust principal component analysis-based dimensionality reduction and feature extraction with classification based on multiple kernel learning. The proposed method resolved the above-mentioned problems in cancer patient stratification. The model obtained high accuracy with 92.92% and significant statistical tests. These results hold great promise, supporting cancer research, diagnosis, and treatment.

In this paper, we propose a technique using the conservation of optical path length and edge ray theorem to design a Fresnel lens, which can achieve high concentration ratio, high uniform irradiance, and small f-number. The structure of the Fresnel lens consists of three parts: Inner part, middle part, and outer part. The design process is carried out by solving the equations of the conservation of optical path length in Matlab™ program. In addition, a wide range of wavelengths is applied to do rays tracing so that the system becomes more suitable in real conditions. In this technique, the Fresnel lens is constructed by many grooves that are built by the ideal Cartesian oval surface. Thus, the optical efficiency of the designed lens is improved. The simulation results by LightTools™ software show that the Fresnel lens has good optical properties such as a high concentration ratio of 900x, f-number = 0.46, high uniform irradiance distribution, and optical efficiency larger than 85%.

In the context of industrial revolution 4.0, grounded robot is frequently utilized in real world. In this paper, an adaptive sliding mode controller for autonomous robot is proposed. This scheme adapts with nonlinear characteristics, variations of load and external disturbance. To enhance the trajectory tracking target, the error system is classified into two sub-system, second-order system related to angular tracking error and third-order system associated with linear one. Firstly, the practical model of grounded robot is introduced to state the technical problem. Later, the dynamic equation of autonomous system is simulated the theoretical characteristics. The stability of the proposed controller is validated by Lyapunov theorem. Finally, the numerical test results are carried out to illustrate the robustness, effectiveness and feasibility of the proposed controller. From these achievements, it can be seen that our approach is proper for various applications such as warehouse management, factory automation, logistics and supply chain.

Nowadays, hand gestures are becoming one of the most natural and intuitive ways of communication between human and computer. To this end, a complex process including hand gesture acquisition, hand detection, gesture representation and recognition must be carried out. This paper presents a method that detects hand and segments hand regions from images captured by a Kinect sensor. As Kinect sensor provides not only RGB images as conventional camera, but also depth and skeleton, in our work, we incorporate multi-modal data from Kinect to deal with hand detection and segmentation. Specifically, we use skeleton to approximately determine hand palm. Then a skin based detector will be applied to discard non-skin pixels from the region of interest. Using depth data helps to limit the human body regions and remove false positive regions from the previous steps. Finally, morphological operations will be applied to fill holes in the hand region. The main advantage of this method is very easy to implement and it performs in real-time on an ordinary computer. We evaluate the proposed method on a dataset of hand gestures captured from different viewpoints. Experiment shows that it provides reasonable accuracy at very high frame rate. It also produces comparable performance in comparison with deep learning based methods.

MicroRNAs (miRNAs) are a class of small non-coding RNAs that are involved in the development of various complex human diseases. A great effort has spent to uncover the relations between miRNAs and diseases for decades. Although most of known miRNA-disease associations are discovered by experimental methods, the experimental methods are in general expensive and time-consuming. Another approach using computational methods to predict potential miRNA-disease associations has been attracted many computer scientists in recent years. However, computational methods suffer from various limitations that affect the prediction accuracy and their applicability. In this paper, we proposed a new computational method that would be able to predict reliable miRNA-disease associations. We integrate different biological data sources such as known miRNA-disease associations, miRNA-miRNA functional similarity, and disease-disease semantic similarity into a miRNA-disease heterogeneous network. The structural characteristics of this network are represented as a feature vector dataset via meta-paths and a binary classification problem is formulated. However, because the number of known miRNA-disease associations is very small, we face with an imbalance data classification problem. To solve this issue, a clustering-based under-sampling algorithm has been proposed. Training classification models using SVMs, we obtained results of 2–5% higher in AUC measures when compared to previous methods. These results implied that our proposed model could be used to discover reliable miRNA-disease associations in the human genome.

Background: Acne vulgaris is a common disease of teenager which is affected by the knowledge, attitude and practice of acne patients. Aims: To describe knowledge, attitude, practice features about acne vulgaris and its associations among acne patients. Material and Method: From June/2017 to May/2018, 251 acne patients satisfying selective criteria were enrolled in our study and interviewed for all needed information. Likert scale were used to measure patients’ attitude. Results: There were associations between career and knowledge, educational background and knowledge (p

The supervision and feeding of grazing livestock are always difficult missions. Since animals act based on habits, the real-time monitoring data logger has become an indispensable instrument to assist farmers in recognizing the status of livestock. Position-tracked and acoustic monitoring have become commonplace as two of the best methods to characterize feeding performance in ruminants. Previously, the existing methods were limited to desktop computers and lacked a sound-collecting function. These restrictions impacted the late interventions from feeders and required a large-sized data memory. In this work, an open-source framework for a data collector that autonomously captures the health information of farm animals is introduced. In this portable hardware, a Wireless Location Acoustic Sensing System (WiLASS) is integrated to infer the health status through the activities and abnormal phenomena of farming livestock via chew&ndash, bite sound identification. WiLASS involves the open modules of ESP32-WROOM, GPS NEO-6M, ADXL335 accelerometer, GY-MAX4466 amplifier, temperature sensors, and other signal processing circuits. By means of wireless communication, the ESP32-WROOM Thing micro-processor offers high speed transmission, standard protocol, and low power consumption. Data are transferred in a real-time manner from the attached sensing modules to a digital server for further analysis. The module of GPS NEO-6M Thing brings about fast tracking, high precision, and a strong signal, which is suitable for highland applications. Some computations are incorporated into the accelerometer to estimate directional movement and vibration. The GY-MAX4466 Thing plays the role of microphone, which is used to store environmental sound. To ensure the quality of auditory data, they are recorded at a minimum sampling frequency of 10 KHz and at a 12-bit resolution. Moreover, a mobile software in pocket devices is implemented to provide extended mobility and social convenience. Converging with a cloud-based server, the multi-Thing portable platform can provide access to simultaneously supervise. Message Queuing Telemetry Transport (MQTT) protocol with low bandwidth, high reliability, and bi-direction, and which is appropriate for most operating systemsOS, is embedded into the system to prevent data loss. From the experimental results, the feasibility, effectiveness, and correctness of our approach are verified. Under the changes of climate, the proposed framework not only supports the improvement of farming techniques, but also provides a high-quality alternative for poor rural areas because of its low cost and its ability to carry out a proper policy for each species.

In this paper, we propose a new approach based on Gaussian descriptors for action recognition. We first develop a feature representation technique that encodes high-order statistics of local features in two levels, where single Gaussians are used to capture the distributions involved. To deal with the possible loss of information about the distribution of features caused by heterogeneous feature vectors when summarizing them, we use K-means clustering and Sparse Coding to construct some sets of feature vectors over which the summarization is performed. We then present two methods based on depth images and pose data for action recognition. In both methods, the proposed feature representation technique is applied to effectively obtain discriminative action descriptors. Experimental evaluation on the seven benchmark datasets, i.e., MSRAction3D, MSRGesture3D, DHA, SKIG, Florence, UTKinect, and HDM05, shows that our methods achieve very promising results on all the datasets.

Hereditary haemochromatosis (HH) is an autosomal recessive disease, where HFE C282Y homozygosity accounts for 80–85% of clinical cases among the Caucasian population. HH is characterised by the accumulation of iron, which, if untreated, can lead to the development of liver cirrhosis and liver cancer. Since iron overload is preventable and treatable if diagnosed early, high-risk individuals can be identified through effective screening employing artificial intelligence-based approaches. However, such tools expose novel challenges associated with the handling and integration of large heterogeneous datasets. We have developed an efficient computational model to screen individuals for HH using the family study data of the Hemochromatosis and Iron Overload Screening (HEIRS) cohort. This dataset, consisting of 254 cases and 701 controls, contains variables extracted from questionnaires and laboratory blood tests. The final model was trained on an extreme gradient boosting classifier using the most relevant risk factors: HFE C282Y homozygosity, age, mean corpuscular volume, iron level, serum ferritin level, transferrin saturation, and unsaturated iron-binding capacity. Hyperparameter optimisation was carried out with multiple runs, resulting in 0.94 ± 0.02 area under the receiving operating characteristic curve (AUCROC) for tenfold stratified cross-validation, demonstrating its outperformance when compared to the iron overload screening (IRON) tool.

Distributed generators in a microgrid (MG) mostly are renewable energy sources. A system with multiple battery energy storage devices should be used in the MG to improve power supply reliability of these renewable energy sources. This system is defined as an aggregate battery energy storage system (ABESS). To demonstrate the significance of the ABESS in the MG, its operation reliability will be analyzed in this paper. A systematic method is proposed to evaluate reliability performance of the ABESS under different failure operation cases. Besides the used-time-dependent failure rates, voltage-dependent failure rates (VDFR) of critical components in the ABESS such as DC/DC converters, DC/AC inverters, battery modules, and battery charger/controller are also formulated and incorporated in the reliability evaluation. According to differently dynamic operation cases of a microgrid with the ABESS and photovoltaic generation systems, the VDFR of the ABESS will be affected. This paper uses an analytic approach based on Markov models to assess the operation reliability of the whole ABESS. Simulation results are presented and discussed to validate that the operation reliability of the ABESS in the microgrid significantly depends on variously dynamic operating strategies along with the voltage stress.

Predicting stock prices is a challenging task due to the highly stochastic nature of the financial market. Among many proposed quantitative approaches to tackle this problem, machine learning, in recent years, has become one of the most promising methods. However, machine learning is still new to a large part of Vietnamese investors community. This motivated us to take some first steps in using machine learning techniques on Vietnamese stock data, in particular top 20 listed stocks (according to market capitalization) of VN-Index in June 2019. The experimental results suggest that machine learning and hybrid methods give better performances in forecasting stock price fluctuation than ones achieved by traditional methods such as the Autoregressive Integrated Moving-average model. To realize our study, we implement a web-based tool and release its source code.

International audience; An efficient framework for dynamic texture (DT) representation is proposed by exploiting local features based on Local Binary Patterns (LBP) from filtered images. First, Gaussian smoothing filter is used to deal with near uniform regions and noise which are typical restrictions of LBP operator. Second, the receptive field of Difference of Gaussians (DoG), which is exploited in DT description for the first time, allows to make the descriptor more robust against the changes of environment , illumination, and scale which are main challenges in DT representation. Experimental results of DT recognition on different benchmark datasets (i.e., UCLA, DynTex, and DynTex++), which give outstanding performance compared to the state of the art, verify the interest of our proposal.

International audience; An effective model, which jointly captures shape and motion cues, for dynamic texture (DT) description is introduced by taking into account advantages of volumes of blurred-invariant features in three main following stages. First, a 3-dimensional Gaussian kernel is used to form smoothed sequences that allow to deal with well-known limitations of local encoding such as near uniform regions and sensitivity to noise. Second , a receptive volume of the Difference of Gaussians (DoG) is figured out to mitigate the negative impacts of environmental and illumination changes which are major challenges in DT understanding. Finally, a local encoding operator is addressed to construct a discriminative descriptor of enhancing patterns extracted from the filtered volumes. Evaluations on benchmark datasets (i.e., UCLA, DynTex, and DynTex++) for issue of DT classification have positively validated our crucial contributions.

Photonic crystal (PC) is usually realized in materials with high refractive indices contrast to achieve a photonic bandgap (PBG). In this work, we demonstrated an optimization of two-dimensional PCs using a low refractive index polymer material. An original idea of assembly of polymeric multiple rings in a hexagonal configuration allowed us to obtain a circular-like structure with higher symmetry, resulting in a larger PBG at a low refractive index of 1.6. The optical properties of such newly proposed structure are numerically calculated by using finite-difference time-domain (FDTD) method. The proposed structures were realized experimentally by using a direct laser writing technique based on low one-photon absorption method.

Together with the significant leap in technologies, the interference among them make our lives more plentiful. Only a lone major could not last in current era. Instead of that, multidisciplinary approach becomes a novel trend in industrial evolution. In this paper, the collaboration in various fields in one platform is introduced as successful multidisciplinary model. All fields of study profit from this shared hardware, such as cost saving, obsessive partnership, municipal culture or communication. In addition, students are enhanced in their behaviors, for example community awareness, discipline knowledge or entrepreneunal mindset.

Nowadays, mobile robot is very popular and still one of key issues in automation field. The control approach is a core technique which varies from simulation to experiment. Recently, image processing technique gains many significant achievements. In this paper, a method to follow target by using digital camera is presented. Mobile robot become an intelligent system to estimate and track the movement of target. In the test scenario, autonomous robot and human share the workspace. From the results, robot is able to follow worker to complete a mission.

The emerging technologies into a compact unit become one of key trends in engineering design. If the product has more advanced functions, the chance to utilize in a wide range of industry is identified. Only a single field could not succeed in our era. Therefore, in this paper, the integration in many majors into one hardware is presented as successful model. The design of autonomous model involves hardware components and infrastructure. Then, the modeling of autonomous system is simulated on computer to meet the proposed design. The implementation of control strategy into model is to drive autonomous system to complete mission. From these results, it can be seen that the proposed design is feasible, effective and capable in real world.

International audience; An efficient texture modelling framework based on Topological Attribute Patterns (TAP) is presented considering topology related attributes calculated from Local Binary Patterns (LBP). Our main contribution is to introduce new efficient mapping mechanisms that improve some typical mappings for LBP-based operators in texture classification such as rotation invariant patterns (ri), rotation invariant uniform patterns (riu2), and Local Binary Count (LBC). Like them, the proposed approach allows contrast and rotation invariant image description using more compact descriptors by projecting binary patterns to a reduced feature space. However, its expressiveness, and then its discrimination capability, is higher, since it includes additional information, related to the connected components of the binary patterns. The proposed mapping, evaluated and compared with different popular mappings, validates the interest of our approach. We then develop Complemented Patterns of Topological Attributes (CTAP) that generalise TAP model and exploit complemented information to further enhance its discrimination capability, and evaluate it on different texture datasets.

In previous works, the perceived safety and comfort are currently not the principal objectives of all industries, especially robotics system. It might lead not to take psychological safety into consideration of adjusting robot behavior, hence, the human-robot interaction lacks of ease and naturalness. In this paper, a novel framework of human’s zones to ensure safety for social interactions in human-machine system is proposed. In the context of service robot in hospital, machine should not produce any actions that may induce worry, surprise or bother. To maintain the comfortable interaction, an algorithm to update human’s state into personal space is developed. Then, a motion model of robot is demonstrated with assumption of the reference path under segmentation. Dynamic Window Approach is employed for motion planning while Optimize Cost function searches the shortest path in a graph. To validate our approach, three test cases (without human-aware framework, with basic model of human’s zone and with extended personal space) are carried out in the same context. Moreover, three interactive indicators, for instance collision index (CI), interaction index (CII) and relative velocity of robot (Vr), are analyzed in different situations. Lack of human-aware framework, robot might break all thresholds and meet the potential collisions. While robot with basic model of human’s zone in its perception maintains the physically safe thresholds but not socially, it respects whole criterions in both physical constraints and social relations. As a result, our findings are useful for robot’s navigation in presence of human while the socially comfortable interaction is guaranteed.

Currently, integrated trends play a key role in every aspect of automation applications. In particular, if the mechanization of agriculture becomes a competitive factor among farmers or nations, then the multi-functional transportation of agricultural products is inevitable in global trade. In sustainable transportation, the challenge of overcoming stable control in harsh environments, such as through imprecise parameters or varying loads, should be addressed. In this paper, a novel controller for a nonholonomic mechanical system able to adapt to uncertainties is proposed. Based on the multi-functional autonomous carrier (MAC), the system configuration of the kinematic and dynamic model is launched in order to identify the unstable problems that arise when tracking the trajectory. To solve these troubles, the decoupled formation of a MAC system has been investigated by considering two second-order components, namely a linear speed-based sub-system and angular speed-based sub-system. To stabilize the whole system using the Lyapunov theory, the advanced control techniques are studied. To validate the proposed approach, a series of test scenarios have been carried out. From the superior performance of numerous trials, it is clear that our approach is effective, feasible, and reasonable for the advanced control of agricultural applications.

Understanding the chaotic motions of dynamic textures (DTs) is a challenging problem of video representation for different tasks in computer vision. This paper presents a new approach for an efficient DT representation by addressing the following novel concepts. First, a model of moment volumes is introduced as an effective pre-processing technique for enriching the robust and discriminative information of dynamic voxels with low computational cost. Second, two important extensions of Local Derivative Pattern operator are proposed to improve its performance in capturing directional features. Third, we present a new framework, called Momental Directional Patterns, taking into account the advantages of filtering and local-feature-based approaches to form effective DT descriptors. Furthermore, motivated by convolutional neural networks, the proposed framework is boosted by utilizing more global features extracted from max-pooling videos to improve the discrimination power of the descriptors. Our proposal is verified on benchmark datasets, i.e., UCLA, DynTex, and DynTex++, for DT classification issue. The experimental results substantiate the interest of our method.

The increase of obesity, its related diseases and the high incidence of metabolic diseases as a whole, constitute a major public health problem on a global scale. New strategies that allow for the discovery of novel metabolic disease-related genes are necessary to develop new treatments. In this paper, we proposed an efficient method to predict metabolic disease genes, solving the problem of imbalanced data. The method combined protein-protein interactions and miRNA-target interactions to construct integrated networks, whose topological properties can be used as features to train machine learning classifiers. We applied different strategies to optimize imbalanced class. The best model of gradient boosting achieved a significant F1-score of 0.82. When testing the model with non-disease genes, we predicted 549 candidates, out of which 123 were validated indirectly from literature to be related to metabolic diseases. The remaining genes’ functions were investigated by gene enrichment analysis, revealing their association with diseases known to co-occur with metabolic diseases, such as cancer and cardiovascular conditions. These results indicated that this method contributed to the identification of novel metabolic disease-related genes.

Recent years, farming issue is one of the important problems and need to be considered. To handle this issue, it is so essential that we need figure out solutions to assure the growth of crops effectively. In this paper, we propose the new solution to plant automatically and monitor remotely. Firstly, the model of plantation is described in depth. Then, CPU plays a central role to handle the irrigation and aftercare the whole system. It is implemented several sensors and actuators to manipulate the environmental farm. The system parameters are updated in cloud frequently. Furthermore, all of plantations is supervised by digital cameras. The host PC will analyze the image to predict the process of development in agriculture. Therefore, via Internet of Things (IoT) and image processing technology, farmers are able to estimate the best harvesting time and assure the mass production of food.

In the developing countries, aging of construction sites, such as tunnel or bridge, becomes one of dangerous problems in society. Though this is a critical topic for researchers and inventors, very few studies address the problem and meet conditions, for example low-cost, feasible, effective and not require expert to operate. In response, this paper introduces a new technique to navigate the local coordinate of an unmanned aerial vehicles (UAVs) which is strained by the light tether from a known address on the ground. By constructing and analyzing the mechanical structure, main frame of system is able to work in the outdoor environment. Later, there are several sensor levels that feed-back information to CPU. Hence, mainboard robustly adjusts the speed of motor depend on working condition. Especially, the overall system is done in practical experiments to verify the proposed design.

In the field of logistics, the storehouse management plays an important role. It is difficult to handle a large warehouse only with human. Therefore, the implementation of path tracking AGV robot is investigated as an automated solution. From the requirements of warehouse service, the hardware structure of robot is demonstrated in this work. Then, the software design is built to operate in real-time. Besides, overall system is scheduled to realize in various cases. The monitoring program in host PC will track the actual position of AGV along the trajectory. From the experimental design of AGV robot, the feasibility and capability of design and control approach which is proposed in this paper are proved.

In the era of the fourth industrial revolution, instead of human, machine or robot plays an important role in working place. Especially, it is difficult to manage a large warehouse only with human. In this paper, the open platform of automated guided vehicle is introduced to meet the demands of logistics field. The design of hardware is ensured that vehicle is able to lift up and down goods without problems. There are three levels of software design such as firmware, middleware and monitoring software. In lowest level, firmware drives servo motors directly, receives signals from sensors and carried out the control command. The mission of middleware is to protect communication lacking traffic jam. In host PC, monitoring software displays internal parameters of system, provide executed status or caution. In theoretical analysis, the controller is integrated in microprocessor to track the pre-defined trajectory as well as guarantee the smooth motion. From the experimental results, the design and control method that proved in this work are feasible and capable.

During the last decade, sparse representations have been successfully applied to design high-performing classification algorithms such as the classical sparse representation based classification (SRC) algorithm. More recently, collaborative representation based classification (CRC) has emerged as a very powerful approach, especially for face recognition. CRC takes advantage of SRC through the notion of collaborative representation, relying on the observation that the collaborative property is more crucial for classification than the l 1-norm sparsity constraint on coding coefficients used in SRC. This paper follows the same general philosophy of CRC and its main novelty is the application of a virtual collaborative projection (VCP) routine designed to train images of every class against the other classes to improve fidelity before the projection of the query image. We combine this routine with a method of local feature extraction based on high-order statistical moments to further improve the representation. We demonstrate using extensive experiments of face recognition and classification that our approach performs very competitively with respect to state-of-the-art classification methods. For instance, using the AR face dataset, our method reaches 100% of accuracy for dimensionality 300.

Emerging evidence indicates that migrants from Plasmodium falciparum endemic regions are at risk of delayed presentation of P. falciparum malaria. We report three cases of P. falciparum malaria occurring years after arrival in Europe. All patients were originally from Sub-Saharan Africa. Two subjects had controlled human immunodeficiency virus infection and one was a pregnant woman. We performed a literature review of all published cases of delayed presentation of P. falciparum in migrants and identified 32 additional cases. All cases but one originate from sub-Saharan Africa. There was a median time of 36 months between the last visit to a malaria-endemic country and clinical malaria (range: 3 months to 10 years). Pregnancy was the most frequently reported risk factor (11/35 or 31.4%). Parasitemia was ≤ 0.1% in 38% of cases (11/29 reported), and no death was reported. The underlying possible mechanisms for this delayed presentation in migrants from an endemic area probably include the persistence of submicroscopic parasitemia combined with decaying P. falciparum-specific immunity. Suspicion of P. falciparum delayed malaria should remain high in migrants, mainly from sub-Saharan Africa, even without a recent travel history, especially in those presenting risk factors for impaired parasite clearance or distinct immune responses such as pregnancy and HIV infection. In these patients, new prevention and screening strategies should be studied and blood safety policies adapted.

In this paper, we propose a new approach based on distribution descriptors for action recognition in depth videos. Our local features are computed from binary patterns which incorporate the shape and motion cues for effective action recognition. Given pixel-level features, our approach estimates video local statistics in a hierarchical manner, where the distribution of pixel-level features and that of frame-level descriptors are modeled using single Gaussians. In this way, our approach constructs video descriptors directly from low-level features without resorting to codebook learning required by Bag-of-features (BoF) based approaches. In order to capture the spatial geometry and temporal order of a video, we use a spatio-temporal pyramid representation for each video. Our approach is validated on six benchmark datasets, i.e. MSRAction3D, MSRGesture3D, DHA, SKIG, UTD-MHAD and CAD-120. The experimental results show that our approach gives good performance on all the datasets. In particular, it achieves state-of-the-art accuracies on DHA, SKIG and UTD-MHAD datasets.

Aiming at rareness of navigation solutions for vehicle platform in unapplicable-GPS or low quality of GPS environment, this paper investigated an approach of vision-based path planning and navigation by a stereo camera. Firstly, camera is calibrated to get its parameters. In this stage, host PC must deterime locations of all objects in working area. When start point and target point is given out, host based on A∗ algorithm generates trajectory for mobile platform to prevent obstacles. Later, vehicle is driven to track the path planning. Additionally, host is able to update status and re-construct trajectory to avoid dynamic obstacles. The low-cost hardware architecture of mobile platform is built-in to verify the design. From the experiments, it can be seen clearly that vehicle runs on the optimized trajectory and navigated successfully.

Exploring and integrating data from heterogeneous sources have attracted much interest in recent years. However, one of the greatest challenges is that a lot of data are highly dimensional and diverse. In order to effectively combine multiple data sources, it is essential to reduce the number of dimensions and boost computational performance. This could be accomplished by combining multiple kernel learning with dimensionality reduction. In this paper, we propose an improved multiple kernel learning framework, referred to as fMKL-DR, that optimize equations to calculate matrix chain multiplication. To reach this conclusion, we performed several comparative evaluations on various biomedical data sets. The results demonstrate that, compared to previous work, the fMKL-DR remarkably improves computational cost. Therefore, the proposed framework is beneficial to the manipulation and integration of huge and complex datasets.

An effective framework for dynamic texture recognition is introduced by exploiting local features and chaotic motions along beams of dense trajectories in which their motion points are encoded by using a new operator, named \(\mathrm {LVP}_{full}\text {-TOP}\), based on local vector patterns (LVP) in full-direction on three orthogonal planes. Furthermore, we also exploit motion information from dense trajectories to boost the discriminative power of the proposed descriptor. Experiments on various benchmarks validate the interest of our approach.

In the cancer research, a number of stratification methods have been successfully applied and have assisted in the treatment process. Currently, various data types related to the cancer patients have been measured and collected. This fact leads to a great need of data integration for obtaining more comprehensive cancer study. Most of the previous work is based on a single data type and employed a tailor-made method for a specific data type. In this paper, we have proposed an efficient approach, using multiple kernel learning methods, to better stratify cancer patients. We integrated the three most related-cancer data, including gene expression, DNA methylation, and miRNA expression. The model has combined multiple kernel learning methods and dimensionality reduction. The achieved results demonstrated that our integrative model was more accurate than the ones based on single data type. Our work holds a great promise to contribute to theoretical cancer research and effectively support the prevention and prognosis.

This paper concerntrates on the analysis, design and control methodology in the machining servo system. The overall architecture of motion system is investigated to realize the drawbacks of traditional scheme. Then, this research proposes the suitable design of controller due to rapid calculation, high precision and friendly graphical user interface. The diagram of Fuzzy self-tuning PID and feedforward is constructed to drive the servo machine. The hardware platform of controller has been accomplised such as schematic, PCB artworks and soldering. Later, the middleware and firmware have been closed to promote the execution between controller and computer. The intelligent algorithm is embedded in board level by TMS320C6727 to overcome the problems, for example the existing nonlinear properties, external disturbances or servo lag phenomena. Based on the theoretical design, several simulation results are provided to verify the contributions. From the experimental performance, it can be seen clearly that this controller smoothly manipulate the servo motor, ensure real-time performance and high tracking position.

Most of quadcopter operate in the outdoor environment where is often complex and unpredicted. In order to control the quadcopter in unknown outdoor environment, it should be designed an excellent filter to estimate complete state vector which illustrates the movement of rigid body. In this paper, two filters such as Complementary and Kalman are investigated to compare the performance. Quadcopter collect only measurements from a low-cost inertial measurement unit, IMU-MPU6050. The raw data is put into filter to estimate the state vector in system. In the simulations, these filters are studied to implement in the model of quadrotor. Later, the hardware platform of quadcopter is built to denote experimental results in order to validate the effective design of Complementary filter and Kalman filter in quadcopter.

International audience; This paper proposes a new local descriptor for action recognition in depth images using second-order directional Local Derivative Patterns (LDPs). LDP relies on local deriva- tive direction variations to capture local patterns contained in an image region. Our proposed local descriptor combines different directional LDPs computed from three depth maps obtained by representing depth sequences in three orthogonal views and is able to jointly encode the shape and motion cues. Moreover, we suggest the use of Sparse Coding-based Fisher Vector (SCFVC) for encoding local descriptors into a global representation of depth sequences. SCFVC has been proven effective for object recognition but has not gained much attention for action recognition. We perform action recognition using Extreme Learn- ing Machine (ELM). Experimental results on three public benchmark datasets show the effectiveness of the proposed approach.