Accurately estimating the state of equipment plays an important role in ensuring efficient operation Industrial 4.0 systems. This article focuses on monitoring operating and detecting faults beam pumping units under condition heavy noise within Internet Things. On one hand, system designed this uses acceleration sensor, signal which contains considerable that greatly reduces motion estimation accuracy. other complexity indicator diagrams makes it difficult to extract features, limits ability...

Riverbank sand overexploitation is threatening the ecology and shipping safety of rivers. The rapid identification riverbank mining areas from satellite images extremely important for ecological protection management. Image segmentation methods based on AI technology are gradually becoming popular in academia industry. However, traditional neural networks have complex structures numerous parameters, making them unsuitable meeting needs extraction large areas. To improve efficiency, we...

Street trees are of great importance to urban green spaces. Quick and accurate segmentation street from high-resolution remote sensing images is significance in space management. However, traditional methods can easily miss some targets because the different sizes trees. To solve this problem, we propose Double-Branch Multi-Scale Contextual Network (DB-MSC Net), which has two branches a (MSC) block encoder. The MSC combines parallel dilated convolutional layers transformer blocks enhance...

The introduction of an attention mechanism in remote sensing image segmentation improves the accuracy segmentation. In this paper, a novel multi-scale KAN-based linear (MKLA) network MKLANet is developed to promote better result. A hybrid global–local feature decoder designed enhance ability aggregating context and avoiding potential blocking artifacts for extraction local channel adopts MKLA block by bringing merits KAN convolution Mamba like improve handling nonlinear complex function...

This paper proposes a three-dimensional micro-vibration measurement method based on amplification and tracking to achieve vibration of the target. In environments with excessive lighting, which can hinder effective target tracking, first applies magnification algorithm enlarge image. The enlarged image is then processed using stereo matching align left right camera views. Following this, calculates pixel displacement Finally, target's spatial information derived through dimensional...

Volume of change provides a comprehensive and objective reflection land surface transformation, meeting the emerging demand for feature monitoring in era big data. However, existing methods often focus on single dimension, either horizontal or vertical, making it challenging to achieve quantitative volumetric monitoring. Accurate measurements are indispensable many fields, such as open-pit coal mines. Therefore, main content conclusions this paper follows: (1) A method Automatic Control...

Beam pumping units (BPUs) are key equipment in oilfield production. Currently, many fault diagnosis methods for BPUs have been developed, and most of them based on feature or image classification indicator diagrams. However, low-quality monitoring data the limited proportion effective pixels diagram greatly restrict performances these methods. This article proposes an efficient two-step method BPUs. In first step, to overcome impact data, a dynamic time warping-based matching is proposed...

In this paper, measurements from a spectrum awareness system are used to study the application of machine learning methods in dynamic assignment for Land Mobile Radio (LMR). Specifically, deep recurrent neural network is learn time-varying distributions users' traffic, which turn, help determine best and sharing strategies LMR bands. Using RF data, simulations conducted validate evaluate suitability chosen methodology. It shown that approaches have great potential characterizing usage...

Change detection (CD), as a special remote-sensing (RS) segmentation task, faces challenges, including alignment errors and illumination variation, dense small targets, large background intraclass variance in very high-resolution (VHR) images. Recent methods have avoided the misjudgment caused by variation increasing ability of global modeling, but latter two problems still not been fully addressed. In this paper, we propose new CD model called SFCD, which increases feature extraction...

Aiming at the current alumina ceramic ball speckle defect detection algorithm in practical applications, such as low accuracy and slow speed, a fast method of surface defects based on improved YOLOv5 is proposed. The image data first obtained using camera, then processed to standardize size, rotated, added noise, subjected other enhancement operations. annotated annotation software create dataset, which used improve backbone network YOLOv5s MobileNetV3. When parameters are checked, it...

Abstract. Urban lakes serve an indispensable role in maintaining the ecological balance of cities, ensuring flood safety, and providing recreational spaces for tourism. With development human activities economic, extent urban are inevitably influenced. Currently, ability to detect detailed temporal changes lake areas using high resolution data still has limitations. This study proposed a novel method by combining time series Gaofen-1 (GF-1) remote sensing random forest machine learning...

Cloud rendering is an emerging technology in which rendering-heavy applications run on the cloud server and then stream rendered contents to end-user device. High density high scalability of services are crucial support millions users concurrently cost-effectively. However, it still challenging Android OS smoothly with without compromising user experience. This paper presents DroidCloud, first open-source Android\footnoteAndroid a trademark Google LLC. solution focusing scalable design...

Several studies have been proposed to deploy the flow recording (i.e., size counting and sketching algorithms) on programmable switches for high-speed processing, helping network management tasks like scheduling. Although provide a remarkable packet processing speed, they are of compact resources follow restrictive pipeline programming. To fit these limitations, current algorithms either sacrifice accuracy or harm switch throughput. In this paper, we propose InheritSketch further...

As the concept of minimally invasive repair has gradually become mainstream, onlays have attracted more and attention from clinicians. Ceramic resin are two kinds materials that mainly used to make onlays. This article will review physical properties, bonding properties clinical applications materials, in order provide reference for practice.

This paper suggests an enhanced YOLOv5 network segmentation technique to address the issue of low efficiency and accuracy standard detection algorithms increase automation degree scratch ceramic blades. The addition attention module improves performance network. It also increases network's concentration on important features lessens impact unimportant features. To improve ability extract more information from image, image dataset is expanded. By using feature maps different scales for object...

Low-level handcrafted features (e.g., edge and saliency) dominate the design of traditional algorithms, endow themselves effective capability dealing with simple classification problems. However, such excellent properties have not been well explored in popular deep convolutional neural networks (DCNNs). In this paper, we propose a new model, termed Guided Convolutional Networks (GCNs), using low-level to guide training process DCNNs, which can be used following vision tasks. Furthermore,...