Diabetic retinopathy (DR) is a disease that forms as complication of diabetes. It particularly dangerous since it often goes unnoticed and can lead to blindness if not detected early. Despite the clear importance urgency such an illness, there no precise system for early detection DR so far. Fortunately, could be achieved using deep learning including convolutional neural networks (CNNs), which gained momentum in field medical imaging due its capability being effectively integrated into...

Continuous growing interest in IoT applications particularly for a smart city setting has attracted many researchers. E-health networks are the newest area of this research field. On other hand, networking and communications fields witnessing revolution through new concepts Mobile Edge Computing (MEC) characterised by latency sensitivity geographical awareness. Moreover, Software Defined Network (SDN) is an innovative network paradigm that allows programming separation data plane control...

In this paper, we propose a smart system for realtime tracking of airport luggage using mobile applications and smartwatches. We track Kalman-filtered Wi-Fi fingerprints collected by active tags. Information about the flights association with different pieces is inputted pre-flight QR codes. Our uses power management scheme fusing multi-sensor flight data assesses risk battery drain to warn user need recharging. A phone application used arrival allowing passengers rest after long flight....

This paper proposes a smart queue management system for delivering real-time service request updates to clients' smartphones in the form of audio and visual feedback. The proposed aims at reducing dissatisfaction with services medium long waiting times. To this end, allows carriers digital ticket leave areas return time their turn receive service. also improves experience clients choosing stay area by connecting them signal often muted television sets running entertainment programs,...

Mobile government is an innovative research area where efforts are made to advance governmental and public services. As such, in this paper, we propose a crisis management system for real-time emergency notification of users using mobile applications smart watches. We develop intuitive web portal client server architecture agencies easily efficiently notify within the range danger occurrence disaster through SMS or push notifications application watch, if latter available. Moreover, mapping...

Infrared (IR) images are characterized by a lower sensitivity to lighting conditions than the visible spectrum. This opens door relatively untapped research potential of automatic recognition systems that robust shadows and variability in illumination levels or appearance. IR action (AR) is one such application. It remains fairly unexplored domain IR. As such, this paper, we propose use hidden Markov models (HMM) for AR. We also derive mathematical model variational learning Beta-Liouville...

It is not uncommon to lose everyday objects outside your home. Currently, there are very few technological resources help locate lost objects. This paper discusses a multi-platform mobile application that provides solution this issue. Lost and found helps people report through their phones rather than going the procedure of filling up forms. In addition, it has backend server runs an algorithm for object matching using Speeded Up Robust Features (SURF) match images items with items. The...

Dynamic textures (DT) constitute of objects characterized by stationary properties in time such as how leaves move a windy day. Classification DTs has made tremendous impact various domains video synthesis and segmentation. In this paper, we propose the use Fisher kernels with Dirichlet based Beta-Liouville (BL) hidden Markov models (HMM) for DT recognition. Experiments demonstrate promising results on DynTeX Alpha dataset using proposed generative-discriminative approach. To best our...

Mixture models are broadly applied in image processing domains. Related existing challenges include failure to approximate exact data shapes, estimate correct number of components, and ignore irrelevant features. In this study, the authors develop a statistical self‐refinement framework for background subtraction task by using Dirichlet Process‐based asymmetric Gaussian mixture model. The parameters model learned variational inference methods. They also incorporate feature selection...

Abstract Diabetic retinopathy (DR) is a disease that forms as complication of diabetes, It particularly dangerous since it often goes unnoticed and can lead to blindness if not detected early. Despite the clear importance urgency such an illness, there no precise system for early detection DR so far. Fortunately, could be achieved using deep learning including convolutional neural networks (CNNs), which gained momentum in field medical imaging due its capability being effectively integrated...

Recent years have witnessed tremendous advancement in data modelling. In this paper, we focus on Dynamic textures (DT). We propose the use of Fisher kernels with Generalized Dirichlet based hidden Markov models (HMM) for DT recognition. Experiments demonstrate promising results DynTeX Alpha dataset using proposed generative discriminative approach. This is first application HMMs to recognition, best our knowledge.

One of the pillar generative machine learning approaches in time series data study and analysis is hidden Markov model (HMM). Early research focused on speech recognition application with later expansion into numerous fields, including video classification, action recognition, text translation. The recently developed generalized Dirichlet HMMs have proven efficient proportional sequential modeling. As such, we focus investigating a maximum posteriori (MAP) framework for inference its...

In this paper, we propose an automatic platform for mobile process monitoring by utilising wearable computing. The proposed system is implemented and tested the of renewal expired documents, that caters to both clients issuing governmental entities. client-side consists a application synced with smart watch, which serves receive notifications status update on requested document. On other hand, authorities are served using specialised web portals, offer capability manage incoming requests in...

This paper proposes an education robotic platform that aims to improve teaching methods of programming and robotics skills, for both beginner advanced users. We propose innovative consists a versatile set sensors actuators, controlled by utilizing user-friendly visual language through mobile phone interface, or representational state transfer application interface more Suggested form the foundation problem-based learning, emphasizing hands-on experimental assignments activities,...

The visible spectrum is the most widely used modality for video media. Nonetheless, it highly dependent on lighting conditions. Hence, infrared (IR) imaging lower light sensitivity characterisation presents untapped potential robust automatic recognition systems. This applicable to many applications including IR action (AR), which a relatively young field in IR. As such, this study, authors tackle and multimodal AR with proposed utilisation of variational learning Beta‐Liouville (BL) hidden...

In this paper, we propose a framework for automated assessment of participation in classroom or professional meeting discussions using audio analysis. Participation is key to the success businesses and schools; therefore, these establishments aim measure, incentivise, ultimately increase it. Currently, process mostly subjective. The chair teacher makes subjective judgement on level by relying memory recorded notes. unreliability approach creates need objective tools through automation. We...

Challenges in realtime installation of surveillance systems is an active area research, especially with the use adaptable machine learning techniques. In this paper, we propose variational Beta-Liouville (BL) hidden Markov models (HMM) for AR online setup. This proposed incremental framework enables continuous adjustment system better modelling. We evaluate model on visible IOSB dataset to validate framework.

The United Arab Emirates is focusing on cultivating Renewable Energy (RE) to meet its growing power demand. This also brings planning the forefront in regards keen interests renewable constrained economic dispatch. paper takes note of UAE's vision incorporating a better energy mix (RE), nuclear, hybrid system along with existing plants mostly utilizing natural gas; further attention for sound dispatch scenario. describes and delves into usage Genetic Algorithm optimize proposed thermal solar...