Abstract The integrity of railway track structures is gravely threatened by rail damage, posing significant risks to the safe operation transportation systems. However, existing detection techniques, including computer vision, ultrasound, acoustic emission, and thermal imaging, exhibit inherent limitations in identifying minor defects concealed damage at rail’s underside. To address this challenge, a method based on modal analysis proposed, integrating finite element simulations with...

The collision between sea ice and ship is a serious threat to navigation safety. However, the idea of using vibration data generated from ship-ice collisions obtain physical information about ice, such as its type thickness, novel method. This method overcomes problem traditional techniques (optical cameras satellite remote sensing) being affected by visibility, also provides for assessing load capacity ships. In this paper, we propose new thickness identification based on sensor networks...

Tracking and segmenting small targets in remote sensing videos on edge devices carries significant engineering implications. However, many semi-supervised video object segmentation (S-VOS) methods heavily rely extensive random-access memory (VRAM) resources, making deployment challenging. Our goal is to develop an edge-deployable S-VOS method that can achieve high-precision tracking by selecting a bounding box for the target object. First, tracker introduced pinpoint position of tracked...

The quantitative evaluation of sports motions is a precondition for scientific training and data-driven professional athletic motion analyses. However, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">high accuracy</i> xmlns:xlink="http://www.w3.org/1999/xlink">portability</i> schemes are contradictory it difficult to simultaneously improve both metrics in one device. In this work, we propose an effective gymnastics recognition system...

Aircrafts have been an important object of study in the field multi-scale image detection due to their strategic role. However, aircrafts and key parts from remote sensing images can be a challenge, as often present complex backgrounds obscured conditions. Most today’s datasets consist independent objects lack mixed annotations parts. In this paper, we contribute aircraft dataset (AP-DATA) consisting 7000 that were taken environments Our includes We also information augmentation framework...

The increasing consumption of ducks and chickens in China demands characterizing carcasses domestic birds efficiently. Most existing methods, however, were developed for pigs or cattle. Here, we a noncontact automated weighing method duck hanging on production line. A 2D camera with its facilitating parts recorded the moving To estimate weight carcasses, images acquired dataset modeled by convolution neuron network (CNN). This model was trained evaluated using 10-fold cross-validation....

The recognition of abnormal gait behavior is important in the field motion assessment and disease diagnosis. Currently, primarily recognized by pressure inertial data obtained from wearable sensors. However, drift wearing difficulties for patients have impeded application these Here, we propose a contactless method that captures human pose using monocular camera. A lightweight OpenPose (OP) model generated with Depthwise Separable Convolution to recognize joint points extract their...

To accurately recognize ordinary handwritten Chinese characters, it is necessary to the normative level of these characters. This study proposes methods quantitatively evaluate and characters based on their similarities. Three different types similarities, including correlation coefficient, pixel coincidence degree, cosine similarity, are calculated between printed Song typeface Eight features derived from similarities used verify evaluation performance an artificial neural network character...

Abstract Retraction: [Tao Cheng, Lianjiang Li, Optimization of E‐commerce platform marketing method and comment recognition model based on deep learning intelligent blockchain, IET Software 2023 ( https://doi.org/10.1049/sfw2.12117 )]. The above article from , published online 3 February in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the Editor‐in‐Chief, Hana Chockler, Institution Engineering Technology (the IET) John Sons Ltd. This was as part a...

When well cementing is conducted in horizontal shale gas wells Weiyuan block of the Weiyuan–Changning State Shale Demonstration Area, Sichuan Basin, oil-based drilling fluid not compatible with cement slurry, difficult to be displaced because its high density, and sheath tends broken situations large-scale stimulated reservoir volume fracturing. In view these problems, a series studies were carried out specifically on theoretical basis sealing mechanical parameters, oil-displacement preflush...

Traditional asphalt rejuvenators, like aromatic oil (AO), are known to be effective in improving the low-temperature properties and fatigue performances of aged SBS (styrene-butadiene-styrene) modified (SBSMA) binders mixtures. However, these rejuvenators inevitably compromise their high-temperature deformation resistances because they dilute binder but do not fix damaged structures SBS. In this study, a highly-active chemical called polymerized 4,4-diphenylmethane diisocyanate (PMDI) was...

Periodic sea ice is generated every winter seriously affects the safety of navigation and maritime operations in high latitude areas. However, current monitoring techniques could not detect basic physical parameters an automatic accurate way, which far away from needs marine production navigation. At present, most used methods based on naked eyes equipment are time-consuming labor-expensive. While utilizing satellite remote sensing radar observation station highly limited by accuracy cost....

In this paper, thermo-responsive cement slurry system were being developed, the properties of conventional slurry, compressive strength high temperature sheath, mechanical sheath and thermal tested. Results used simulated by Well-Life Software, Thermo-responsive can meet requirements heavy oil recovery production. Mechanical Tensile fracture energy is larger than cement. The heat absorption capacity that means more needed for unit mass once increasing 1.0 °C, which also indicates own good...

The close packing technology is the key part of high-performance cement slurry system design, which can effectively solve difficulties high density mix proportion in complex well conditions. Through test particle surface properties and size distribution, we select weighting material suitable for packed design; based on compressible model, predict multiparticle density. And also combining with water demand prediction, it comes out series high-density design method software. software,...

ABSTRACT Arrhythmia is a marked symptom of many cardiovascular diseases (CVDs). The accurate and timely detection heart rate can greatly reduce the harm arrhythmia to people. However, it still challenge robustly remotely measure in daily life due changing environmental conditions during measurement, such as varying light intensity, movement people, uncertain distance between sensor In this study, we propose method accurately human within 4.5 meters under different intensities by simply using...

The aircrafts and ships in visible remote sensing (RS) images are of different scales. They difficult to detect as they may be easily obscured by complex weather conditions such snow cloud. Therefore, it is important eliminate the interference order these multi-scale objects accurately. This letter proposes an improved robust stacked feature fusion network RSMNet address this problem from two aspects. First, a dilated convolution used enlarge receptive fields high-resolution improve ability...

We introduce two approaches for combining neural evolution strategy (NES) and proximal policy optimization (PPO): parameter transfer space noise. Parameter is a PPO agent with parameters transferred from NES agent. noise to directly add the agent`s parameters. demonstrate that could benefit both methods through experimental comparison on discrete action environments as well continuous control tasks

Soft gripper's grasping prediction is a crucial task in industrial automation, which requires feedback from multiple modalities, such as vision and touch. However, multimodal fusion challenging due to mismatched data dimensions density. Meanwhile, the soft bending result of force can be monitored by flex sensors, may also helpful for prediction. Here we create dataset fin ray effect gripper various objects while being camera, pressure sensor, sensor. Next, fuse sensory based on deep learning...