The threat to people's lives and property posed by fires has become increasingly serious. To address the problem of a high false alarm rate in traditional fire detection, an innovative detection method based on multifeature fusion flame is proposed. First, we combined motion color as preprocessing stage. This saves lot computation time screening candidate pixels. Second, although irregular, it certain similarity sequence image. According this feature, novel algorithm centroid stabilization...

Offshore drilling platforms (ODPs) are critical infrastructure for exploring and developing marine oil gas resources. As these platforms' capabilities expand, deploying intelligent surveillance services to ensure safe production has become increasingly important. However, the unique geographical locations harsh environmental conditions of ODPs pose significant challenges processing large volumes video data, complicating implementation efficient systems. This study proposes a Cloud-Edge...

Abstract Neural-like computing models are versatile mechanisms in the field of artificial intelligence. Spiking neural P systems (SN for short) one recently developed spiking network inspired by way neurons communicate. The communications among essentially achieved spikes, i. e. short electrical pulses. In terms motivation, SN fall into third generation models. this study, a novel variant systems, namely with self-organization, is introduced and computational power system investigated...

Spiking neural P systems (SN systems) are a class of parallel and distributed spiking network models, which inspired from the way biological neurons communicating by means spikes. White hole rules, abstracted observation information rejection, were recently introduced into SN systems, neuron consumes its complete contents when it fires. In this work, with white proposed, in each has only rules. The computational power general bounded obtained. Specifically, is achieved constructive that i)...

In special industrial fields such as electric submersible pump (ESP) wells, named entity recognition (NER) often suffers from low accuracy and incomplete due to the scarcity of high-quality corpora prevalence rare words nested entities. To address these issues, this study introduces a character-level convolutional neural network (char-CNN) into Flat-Lattice Transformer (FLAT) model constructs matching rules for ESP well domain, forming char-CNN-FLAT-CRF model. This achieves NER in...

Abstract In order for the offshore drilling platform to operate properly, workers need perform regular maintenance on equipment, but complex working environment exposes hazards. During inspection and maintenance, use of personal protective equipment (PPE) such as helmets workwear can effectively reduce probability worker injuries. Existing PPE detection methods are mostly construction sites only detect whether worn or not. This paper proposes a high-precision high-speed method based object...

Summary Emotion recognition is challenging for understanding people and enhances human–computer interaction experiences, which contributes to the harmonious running of smart health care other services. In this paper, several kinds speech features such as Mel frequency cepstrum coefficient, pitch, formant were extracted combined in different ways reflect relationship between feature fusions emotion performance. addition, we explored two methods, namely, support vector machine (SVM) deep...

The integration of oilfield multidisciplinary ontology is increasingly important for the growth Semantic Web. However, current methods encounter performance bottlenecks either in storing data and searching information when processing large amounts data. To overcome these challenges, we propose a domain-ontology process based on Neo4j graph database. In this paper, focus storage retrieval ontology. We have designed mapping rules from files to regulate database, which can greatly reduce...

Maximizing benefits from a cloud cluster with minimum computational costs is challenging. An accurate prediction to workload important maximize resources usage in the environment. In this paper, we propose an approach using recurrent neural networks (RNN) realize prediction, where CPU and RAM metrics are used evaluate performance of proposed approach. order obtain optimized parameter set, orthogonal experimental design conducted find most influential parameters RNN. The experiments Google...

Electric submersible pumps (ESPs) are crucial equipment in offshore oilfield production. Due to their complex structure and the variable geological environments which they work, ESPs prone a wide range of faults. Existing fault diagnosis models for ESP wells face several issues, including high subjective dependence, large sample data requirements, poor adaptability different environments. These issues lead relatively low accuracy well diagnosis. To address these challenges, this paper...

Unmanned aerial vehicles (UAVs) are a key driver of the low-altitude economy, where precise localization is critical for autonomous flight and complex task execution. However, conventional global positioning system (GPS) methods suffer from signal instability degraded accuracy in dense urban areas. This paper proposes lightweight fine-grained visual UAV algorithm (FIM-JFF) suitable electromagnetic environments. FIM-JFF integrates both shallow image features to leverage contextual information...

The Fullbore Formation Micro Imager (FMI) represents a proficient method for examining subterranean oil and gas deposits. Despite its effectiveness, due to the inherent configuration of borehole logging apparatus, micro-resistivity imaging tool cannot achieve complete coverage. This limitation manifests as blank regions on resulting images, thus posing challenge obtaining comprehensive analysis. In order ensure accuracy subsequent interpretation, it is necessary fill these strips....

Food image recognition is increasingly important for e-health applications. But this a challenging topic due to the diversity of food, and color, light, view angles' effect on food image. Based empirical experimental explorations, we propose use SIFT(Scale Invariant Feature Transform) Gabor descriptors as features KMeans algorithm feature clustering. We also pervasive cloud computing paradigm improve performance heavy requirement large number concurrent requests. Evaluations show that...

There is a symmetrical relationship between safety management and production efficiency of an offshore drilling platform. The development artificial intelligence makes people pay more attention to intelligent security management. It extremely important reinforce workplace by monitoring protective equipment wearing using intelligence, such as helmets workwear uniforms. working environment the platforms particularly complex due small-scale subjects, flexible human postures, oil gas pipeline...

Accurate fault identification is essential for geological interpretation and reservoir exploitation. However, the unclear noisy composition of seismic data makes it difficult to identify complete structure using conventional methods. Thus, we have developed an attentional U-shaped network (EAResU-net) based on enhanced feature fusion automated end-to-end 3D data. EAResU-net uses mechanism reduce semantic gap between encoder decoder improve representation features in combination with residual...

People can post their comments on public events the Internet, such as ideas, emotions, and attitudes that affect others. Online opinions may stability or security of country because speed convenience information disseminating Internet. This article proposes opinion cellular automata for situation deduction to predict possible trending events. In automata, online users are represented by cells, eigenvalues calculated from user's historical comment data. The cell neighbors form a space, whose...

An approach is proposed for medical three-dimensional reconstruction in view of the characteristics modern image this paper. The based on traditional Marching Cubes algorithm combines with seed-occupying to expand isosurface avoiding detection empty cubes and replaces cell edge interpolation by midpoint selection DMC method. Finally, generated directly polygons which are ordered isopoints. Because uses instead triangles without resorting pre-defined cases, it reduces number faces greatly...