Fires are classified into five types: A, B, C, D, and F/K, according to the components involved in combustion. Recognizing fire classes is critical, since each kind demands a unique suppression approach. Proper classification helps decrease risk both life property. The fuel type used determine class, so that appropriate extinguishing agent can be selected. This study takes advantage of recent advances deep learning, employing YOLOv11 variants (YOLO11n, YOLO11s, YOLO11m, YOLO11l, YOLO11x)...

The Internet of Things (IoT) is a widely used technology in automated network systems across the world. impact IoT on different industries has occurred recent years. Many nodes collect, store, and process personal data, which an ideal target for attackers. Several researchers have worked this problem presented many intrusion detection (IDSs). existing system difficulties improving performance identifying subcategories cyberattacks. This paper proposes deep-convolutional-neural-network...

The Lemur Optimizer (LO) is a novel nature-inspired algorithm we propose in this paper. This algorithm’s primary inspirations are based on two pillars of lemur behavior: leap up and dance hub. These principles mathematically modeled the optimization context to handle local search, exploitation, exploration search concepts. LO first benchmarked twenty-three standard functions. Additionally, used solve three real-world problems evaluate its performance effectiveness. In direction, compared six...

Forest fires cause extensive environmental damage, making early detection crucial for protecting both nature and communities. Advanced computer vision techniques can be used to detect smoke fire. However, accurate of fire in forests is challenging due different factors such as shapes, changing light, similarity with other smoke-like elements clouds. This study explores recent YOLO (You Only Look Once) deep-learning object models YOLOv9, YOLOv10, YOLOv11 detecting forest environments. The...

Consumer feedback is highly valuable in business to assess their performance and also beneficial customers as it gives them an idea of what expect from new products. In this research, the aim evaluate different deep learning approaches accurately predict opinion based on mobile phone reviews obtained Amazon.com. The prediction analysing these categorizing positive, negative, or neutral. Different algorithms have been implemented evaluated such simple RNN with its four variants, namely, Long...

In recent years data science has been applied in a variety of real-life applications such as human-computer interaction applications, computer gaming, mobile services, and emotion evaluation. Among the wide range speech recognition (SER) is also an emerging challenging research topic. For SER, studies used handcrafted features that provide best results but failed to accuracy while complex scenarios. Later, deep learning techniques were for SER automatically detect from signals. Deep...

The digital content in smart city, is a critical asset. It can be used for multiple purposes and offers added value layers customization personalization of services. Numerous research issues, related to standardization, annotation, visualization sentiment analysis pose questions. In our approach within greatest context academic we deal with the semi-automatic annotation city contexts, more specifically twitter-content promotion citizen's engagement. We discuss detail feasibility technical...

This paper presents a new contribution of the nonlinear control technique electrical energy in wind system. The sliding mode used to powers DFIG-Generator is connected power grid by two converters (grid side and machine side). proposed model validated using tracking robustness tests with real speed. was developed under Matlab/Simulink, FPGA Loop design DFIG model. By employing co-simulation, purpose test controller for simulated or system its entirety. results obtained cο-simulation show...

Due to the wide availability and usage of connected devices in Internet Things (IoT) networks, number attacks on these networks is continually increasing. A particularly serious dangerous type attack IoT environment botnet attack, where attackers can control systems generate enormous "bot" for generating malicious activities. To detect this several Intrusion Detection Systems (IDSs) have been proposed based machine learning deep methods. As main characteristics include their limited battery...

The purpose of this study is to offer an adaptive hybrid controller for the formation control multiple unmanned aerial vehicles (UAVs) leader‐follower configurations with communication delay. Although numerous studies about exist, very few incorporate delay in their model and are as well. motivation behind article bridge that gap. strategy consists fuzzy logic a Proportional, Integral, Derivative (PID) where fines/tunes PID gains. also integrator raises order system which helps reduce noise...

In this paper, we present a hybrid deep learning model that is based on two-dimensional convolutional neural network (2D-CNN) and bidirectional long short-term memory (Bi-LSTM)to detect non-technical losses (NTLs) in smart meters. NTLs occur due to the fraudulent use of electricity. The global integration meters has proven be beneficial for storage historical electricity consumption (EC) data. proposed methodology learns insights from EC data informs power utilities about presence NTLs....

Unmanned Aerial Vehicles (UAVs) seem to be the most efficient way of achieving intended aerial tasks, according recent improvements. Various researchers from across world have studied a variety UAV formations and path planning methodologies. However, when unexpected obstacles arise during collective flight, might get complicated. The study needs employ hybrid algorithms bio-inspired computations address issues with more stability speed. In this article, two models Ant Colony Optimization...

Electricity theft is considered one of the most significant reasons non technical losses (NTL). It negatively influences utilities in terms power supply quality, grid's safety, and economic loss. Therefore, it necessary to effectively deal with electricity problem. For detecting smart grids (SGs), an efficient state-of-the-art approach designed underlying work based on autoencoder bidirectional gated recurrent unit (AE-BiGRU). The proposed consists six components: (1) data collection, (2)...

Forests are an enduring component of the natural world and perform a vital role in protecting environment. valuable resources to control global warming provide oxygen for survival human life, including wood households. Forest fires have recently emerged as major threat biological processes ecosystem. Unfortunately, almost every year, fire damages millions hectares forest land due late inefficient detection fire. However, it is important identify at initial level before spreads vast areas...

Gender recognition by voice is a vital research subject in speech processing and acoustics, as human voices have many remarkable characteristics. Voice beneficial variety of applications, including mobile health care systems, interactive crime analysis, systems. Several algorithms for been developed, but there still potential development terms the system’s accuracy efficiency. Recent has focused on combining ensemble learning with machine models order to create more accurate classifiers. In...

During the pandemic of coronavirus disease (COVID-19), statistics showed that number affected cases differed from one country to another and also city another. Therefore, in this paper, we provide an enhanced model for predicting COVID-19 samples different regions Saudi Arabia (high-altitude sea-level areas). The is developed using several stages was successfully trained tested two datasets were collected Taif area) Jeddah (sea-level Arabia. Binary particle swarm optimization (BPSO) used...

Extracting the relations between medical concepts is very valuable in domain. Scientists need to extract relevant information and semantic concepts, including protein protein, gene drug drug, disease. These can be extracted from biomedical literature available on various databases. This study examines extraction of that occur diseases drugs. Findings will help specialists make good decisions when administering a medication patient allow them continuously up date their field. The objective...

This paper addresses the path planning and control of multiple colonies/clusters that have unmanned aerial vehicles (UAV) which make a network in hazardous environment. To solve aforementioned issue, we design new novel hybrid algorithm. As seen mission requirement, to combine Maximum-Minimum ant colony optimization (ACO) with Vicsek based multi-agent system (MAS) an Artificially Intelligent (AI) scheme. In order manage different colonies, UAVs form network. The designed method overcomes...

This research offers an improved method for the self-organization of a swarm UAVs based on social learning approach. To start, we use three different colonies and best members i.e., unmanned aerial vehicles (UAVs) randomly placed in colonies. study uses max-min ant colony optimization (MMACO) conjunction with mechanism to plan optimized path individual colony. Hereinafter, multi-agent system (MAS) chooses most optimal UAV as leader each remaining agents, which helps organize positioned into...

A vehicular ad hoc network (VANET) is an emerging technology that improves road safety, traffic efficiency, and passenger comfort. VANETs' applications rely on co-operativeness among vehicles by periodically sharing their context information, such as position speed acceleration, others, at a high rate due to mobility. However, rogue nodes, which exploit the feature share false messages, can disrupt fundamental operations of any potential application cause loss people's lives properties....

Wireless Sensors Networks have been the focus of significant attention from research and development due to their applications collecting data various fields such as smart cities, power grids, transportation systems, medical sectors, military, rural areas. Accurate reliable measurements for insightful analysis decision-making are ultimate goals sensor networks critical domains. However, raw collected by WSNs usually not inaccurate imperfect nature WSNs. Identifying misbehaviours or anomalies...

In this study, a novel control approach for doubly-fed induction generator (DFIG) is developed and applied to improve the system’s dynamic response performance providing high energy quality while avoiding harmonic accumulations. Because of its ease implementation, field-oriented (FOC) frequently used. This has great sensitivity machine’s parametric variations. For reason, adaptive Backstepping (ABC) capable preserving almost all robustness properties. However, analytical formulation problem....