Search

Nonstandard miner behavior can have adverse effects on coal mine safety production. Therefore, accurately capturing miner behavior in complex environments is particularly important. In the intelligent mine monitoring system, using visual perception to detect miner behavior is a challenging task due to high behavioral similarity and difficult temporal relationships. In this paper, a new deep learning framework is proposed to construct a coal miner behavior recognition model with a spatio-temporal dual-branch structure and transposed attention representation mechanism. The spatio-temporal dual-branch structure extracts rich spatial semantic information from intrinsic safety video sensor input video sequences while ensuring effective capture of rapidly changing human behavior. Subsequently, considering the discrimination of miner behavior similarity, a merged transposed weighted representation mechanism (TWR) is introduced to guide the model in extracting feature information more strongly related to the classification target, thereby effectively improving the model's ability to classify highly similar behaviors. Experiments were conducted on UCF101, HMDB51, and a self-built miner behavior dataset, achieving significant improvements compared to other state-of-the-art methods. This collaborative structure further creates a more discriminative behavior detection model, contributing to the reliability of miner behavior detection. [ABSTRACT FROM AUTHOR]

Laser vision seam tracking enhances robotic welding by enabling external information acquisition, thus improving the overall intelligence of the welding process. However, camera images captured during welding often suffer from distortion due to strong noises, including arcs, splashes, and smoke, which adversely affect the accuracy and robustness of feature point detection. To mitigate these issues, we propose a feature point extraction algorithm tailored for weld images, utilizing an improved Deeplabv3+ semantic segmentation network combined with EfficientDet. By replacing Deeplabv3+'s backbone with MobileNetV2, we enhance prediction efficiency. The DenseASPP structure and attention mechanism are implemented to focus on laser stripe edge extraction, resulting in cleaner laser stripe images and minimizing noise interference. Subsequently, EfficientDet extracts feature point positions from these cleaned images. Experimental results demonstrate that, across four typical weld types, the average feature point extraction error is maintained below 1 pixel, with over 99% of errors falling below 3 pixels, indicating both high detection accuracy and reliability. [ABSTRACT FROM AUTHOR]

A ratiometric fluorescence probe based on carbon quantum dots with 420 nm emission (bCQDs) and a p-phenylenediamine-derived fluorescence probe with 550 nm emission (yprobe) is constructed for the detection of Mn2+. The presence of Mn2+ results in the enhanced absorption band at 400 nm of yprobe, and the fluorescence of yprobe is significantly enhanced based on the chelation-enhanced fluorescence mechanism. The fluorescence of bCQDs is then quenched based on the inner filtration effect. The ratio (I550/I420) linearly increases with the increase of Mn2+ concentration within 2.00 × 10−7–1.50 × 10−6 M, and the limit of detection is 1.76 × 10−9 M. Given the fluorescence color changing from blue to yellow, the visual sensing of Mn2+ is feasible based on bCQDs/yprobe coupled with RGB value analysis. The practicability of the proposed method has been verified in tap water, lake water, and sparkling water beverage, indicating that bCQDs/yprobe has promising application in Mn2+ monitoring. [ABSTRACT FROM AUTHOR]

With the development of image processing technology, computer vision technology has been widely applied in the field of motion training. How to effectively restore blurred images during motion training and achieve deduplication of motion training images has become a current research hotspot. However, traditional algorithms cannot effectively extract the blurred information in the overlapping areas. The article studied a motion training image blur deduplication algorithm based on visual sensing was studied. The algorithm first performed blur restoration on the input motion training image, and then used the KL (Kullback Leibler) transform to deduplicate the restored image. The results showed that the image blur deduplication method based on visual sensing had a maximum deduplication rate of 88.6%. The visual perception technology proposed in this paper has better results in motion training image processing, which can improve the clarity and visual quality of the image. [ABSTRACT FROM AUTHOR]

A ratiometric fluorescence sensing strategy has been developed for the determination of Cu2+ and glyphosate with high sensitivity and specificity based on OPD (o-phenylenediamine) and glutathione-stabilized gold nanoclusters (GSH-AuNCs). Water-soluble 1.75-nm size GSH-AuNCs with strong red fluorescence and maximum emission wavelength at 682 nm were synthesized using GSH as the template. OPD was oxidized by Cu2+, which produced the bright yellow fluorescence oxidation product 2,3-diaminophenazine (DAP) with a maximum fluorescence emission peak at 570 nm. When glyphosate existed in the system, the chelation between glyphosate and Cu2+ hindered the formation of DAP and reduced the fluorescence intensity of the system at the wavelength of 570 nm. Meanwhile, the fluorescence intensity at the wavelength of 682 nm remained basically stable. It exhibited a good linear relationship towards Cu2+ and glyphosate in water in the range 1.0–10 µM and 0.050–3.0 µg/mL with a detection limit of 0.547 µM and 0.0028 µg/mL, respectively. The method was also used for the semi-quantitative determination of Cu2+ and glyphosate in water by fluorescence color changes visually detected by the naked eyes in the range 1.0–10 µM and 0.30–3.0 µg/mL, respectively. The sensing strategy showed higher sensitivity, more obvious color changes, and better disturbance performance, satisfying with the detection demands of Cu2+ and glyphosate in environmental water samples. The study provides a reliable detection strategy in the environment safety fields. [ABSTRACT FROM AUTHOR]

The construction industry is vulnerable to heat-related hazards, necessitating identification of high-risk areas and contributing factors. This study introduces a novel framework that integrates microclimate simulations with geometric modeling, focusing on the often-underestimated role of heat radiation in assessing heat-related hazards in construction environments. By analyzing 2 years of meteorological data from a construction site in College Station, Texas, this research uncovers the inadequacies of the heat index (HI), a widely recognized thermal-physiological model in the US construction sector. Compared with the wet bulb globe temperature (WBGT), the HI displayed notable variations. Specifically, out of 1,719 data points labeled as danger by HI, WBGT recategorized them as low risk (n=62), moderate risk (n=1,264), high risk (n=300), and extreme risk (n=93). These discrepancies are predominantly associated with the influence of heat radiation. Furthermore, this study emphasizes the importance of accounting for the spatially varying nature of heat radiation in construction environments, influenced by factors such as adjacent structure height, surface materials, and shading patterns. The research highlights the need for monitoring site-specific heat radiation and its potential impact on workers' health and safety. Overall, the findings contribute to our understanding of heat-related hazards in construction and offer valuable insights for developing more effective heat-related safety management strategies. [ABSTRACT FROM AUTHOR]

We present TimberTool (TTool v2.1.1), a software designed for woodworking tasks assisted by augmented reality (AR), emphasizing its essential function of the real-time localization of a tool head's poses within camera frames. The localization process, a fundamental aspect of AR-assisted tool operations, enables informed integration with contextual tracking, facilitating the computation of meaningful feedback for guiding users during tasks on the target object. In the context of timber construction, where object pose tracking has been predominantly explored in additive processes, TTool addresses a noticeable gap by focusing on subtractive tasks with manual tools. The proposed methodology utilizes a machine learning (ML) classifier to detect tool heads, offering users the capability to input a global pose and utilizing an automatic pose refiner for final pose detection and model alignment. Notably, TTool boasts adaptability through a customizable platform tailored to specific tool sets, and its open accessibility encourages widespread utilization. To assess the effectiveness of TTool in AR-assisted woodworking, we conducted a preliminary experimental campaign using a set of tools commonly employed in timber carpentry. The findings suggest that TTool can effectively contribute to AR-assisted woodworking tasks by detecting the six-degrees-of-freedom (6DoF) pose of tool heads to a satisfactory level, with a millimetric positional error of 3.9 ± 1 mm with possible large room for improvement and 1.19 ± 0.6° for what concerns the angular accuracy. [ABSTRACT FROM AUTHOR]

In order to form and regulate skin color throughout the entire visible range, chameleons use a non-close-packed arranged guanine nanocrystals in iris cells. Inspired by chameleons, PS@SiO2 particles are embedded particles in poly-(ethylene glycol) phenyl ether acrylate (PEGPEA) to prepare photonic crystal thin films. Photonic crystal films have bright structural color (reflectivity >60%), high tensile strain (ε = 70%), high sensitivity (2.16 nm/%–1), a fast response time (1.9 nm/ms), and a wide tuning range of reflection wavelength (Δλ = 152 nm). The unique core–shell structure of PS@SiO2 microspheres has elasticity and stretchability. Polystyrene (PS) improves the refractive index difference between the nanometer microspheres and the elastomer matrix, making the structure color of the photonic crystal film brighter. Nondense arrays can ignore particle rearrangement under strain, making the mechanical discoloration process of thin films completely reversible. Based on these characteristics, this crystal thin film is used as a wearable sensor for the human body, which can report various actions by outputting different colors during the change process. Meanwhile, simple manufacturing processes have the potential to achieve large-scale production. So the crystal thin film has certain practical value in visual sensing, display, and anticounterfeiting. [ABSTRACT FROM AUTHOR]

Insects exhibit remarkable abilities in navigating complex natural environments, whether it be evading predators, capturing prey, or seeking out con-specifics, all of which rely on their compact yet reliable neural systems. We explore the field of bio-inspired robotic vision systems, focusing on the locust inspired Lobula Giant Movement Detector (LGMD) models. The existing LGMD models are thoroughly evaluated, identifying their common meta-properties that are essential for their functionality. This article reveals a common framework, characterized by layered structures and computational strategies, which is crucial for enhancing the capability of bio-inspired models for diverse applications. The result of this analysis is the Strategic Prototype, which embodies the identified meta-properties. It represents a modular and more flexible method for developing more responsive and adaptable robotic visual systems. The perspective highlights the potential of the Strategic Prototype: LGMD-Universally Prototype (LGMD-UP), the key to re-framing LGMD models and advancing our understanding and implementation of bio-inspired visual systems in robotics. Itmight open upmore flexible and adaptable avenues for research and practical applications. [ABSTRACT FROM AUTHOR]

We present a novel learning framework to solve the transport-and-packing (TAP) problem in 3D. It constitutes a full solution pipeline from partial observations of input objects via RGBD sensing and recognition to final box placement, via robotic motion planning, to arrive at a compact packing in a target container. The technical core of our method is a neural network for TAP, trained via reinforcement learning (RL), to solve the NP-hard combinatorial optimization problem. Our network simultaneously selects an object to pack and determines the final packing location, based on a judicious encoding of the continuously evolving states of partially observed source objects and available spaces in the target container, using separate encoders both enabled with attention mechanisms. The encoded feature vectors are employed to compute the matching scores and feasibility masks of different pairings of box selection and available space configuration for packing strategy optimization. Extensive experiments, including ablation studies and physical packing execution by a real robot (Universal Robot UR5e), are conducted to evaluate our method in terms of its design choices, scalability, generalizability, and comparisons to baselines, including the most recent RL-based TAP solution. We also contribute the first benchmark for TAP which covers a variety of input settings and difficulty levels. [ABSTRACT FROM AUTHOR]

Amoxicillin (AMX) is commonly used antibiotic with a broad activity against Gram-negative and Gram-positive bacteria. Therefore, it is crucial access to an instant and real-time method for accurate and correct AMX determination. Here, we have developed a three-dimensional hydrogel modified with Europium -Metal organic frameworks (Eu-MOFs) and carbon dots (CDs) as fluorophores, which allows for ratiometric fluorescence detection of AMX. The Eu-MOFs and CDs as an AMX probe trapped in Carboxymethyl cellulose (CMC) based hydrogel. In the presence of AMX, the fluorescence intensity of CDs@Eu-MOFs, enhanced at 448 nm and decreased at 568 nm under 360 nm excitation, due to energy transfer process. Under optimal conditions, AMX determined in the linear range of 10 μM to 106.7 μM, with a detection limit of 1.17 μM in liquid state. While, in the hydrogel state, the linear range for AMX detection is from 0.3 μM to 3.07 μM, with a detection limit of 0.08 μM. So, through applying hydrogel state, assay sensitivity is increased 32-times compared with conventional using liquid state. Furthermore, the change of the fluorescence color under UV irradiation also applied for colorimetric sensing of AMX at concentration up to 106 μM with detection limit 0.63 μM using smartphone RGB color sensing software. In addition, presented assay indicates strong selectivity for AMX over other biomolecules, salts and also other antibiotics. Also, it successfully detects different quantities of AMX in water and milk samples with excellent sensitivity, precision, and reliability in both liquid and hydrogel states.

In order to obtain high precision image identification results of transmission line ice thickness, an image identification method of transmission line ice thickness based on visual sensing was proposed. The ice-covered images of transmission lines are collected by visual sensors, and we take the non-local self-similarity information of the images as a regularized prior constraint term, which is combined with K-SVD algorithm to establish an image denoising model to remove the noise in the images. The ice-covered regions of transmission lines are segmented in the image and the edge features of ice-covered regions are extracted; we can calculate Ice thickness based on regional pixels, and finally realized the image identification of transmission line ice thickness. After experimental testing, it has been proven that the signal-to-noise ratio of the denoised image using the proposed method is higher than 3.76 dB, which can effectively extract the grayscale features of the icing on the transmission line and obtain accurate identification results of the icing thickness. The deviation between the calculated icing thickness and the actual value is less than 0.05 mm, and the accuracy of icing thickness identification is high, which has strong engineering application value. [ABSTRACT FROM AUTHOR]

Robotic fiber placement systems, which have larger workspaces, better dexterity, and easier installations, are less expensive alternatives to the conventional fiber placement machines. To overcome the disadvantage of poor path accuracy, sensor-based path correction can be implemented online by reference to the edge of a previously deposited tow or a guide line on the tooling surface. In fiber placement operations, the end-effector with a compaction roller, acting like a unicycle on the surface, is subject to the nonholonomic constraints when moving along a predefined path to provide proper guidance and pressure for the tows. If the constraints are ignored in path correction, the corrective motion with a lateral velocity component will result in improper tow guidance that induces undesired and uncontrolled tow steering, leading to deviations and wrinkles. In this paper, the nonholonomic constraints of the robot end point is discussed for the first time in online path correction, and a novel method is proposed under the constraints to achieve online path correction without damaging lay-up quality. The path correction problem is analyzed and simplified using twist decomposition. Then the instantaneous motion of the roller is derived by modified vector pursuit algorithm. Afterwards, an iterative algorithm is proposed based on screw theory and differential kinematics for consecutive external control of correction in robot joint space. The implementation of the method is tentatively explored by applying visual sensing based on directional light source and external control to a robotic fiber placement system. Simulations and experiments have showed that the large initial path deviations can be effectively reduced and the roller motions along the corrected paths satisfy the nonholonomic constraints. • The end point nonholonomic constraints are discussed for the first time in online path correction of robotic fiber placement. • The path correction problem is simplified using twist decomposition. • The desired instantaneous motion is derived by modified vector pursuit algorithm. • An iterative algorithm is proposed for consecutive external control of correction in joint space. • The implementation is explored using visual sensing and external control. [ABSTRACT FROM AUTHOR]

We present TimberTool (TTool v2.1.1), a software designed for woodworking tasks assisted by augmented reality (AR), emphasizing its essential function of the real-time localization of a tool head’s poses within camera frames. The localization process, a fundamental aspect of AR-assisted tool operations, enables informed integration with contextual tracking, facilitating the computation of meaningful feedback for guiding users during tasks on the target object. In the context of timber construction, where object pose tracking has been predominantly explored in additive processes, TTool addresses a noticeable gap by focusing on subtractive tasks with manual tools. The proposed methodology utilizes a machine learning (ML) classifier to detect tool heads, offering users the capability to input a global pose and utilizing an automatic pose refiner for final pose detection and model alignment. Notably, TTool boasts adaptability through a customizable platform tailored to specific tool sets, and its open accessibility encourages widespread utilization. To assess the effectiveness of TTool in AR-assisted woodworking, we conducted a preliminary experimental campaign using a set of tools commonly employed in timber carpentry. The findings suggest that TTool can effectively contribute to AR-assisted woodworking tasks by detecting the six-degrees-of-freedom (6DoF) pose of tool heads to a satisfactory level, with a millimetric positional error of 3.9 ± 1 mm with possible large room for improvement and 1.19 ± 0.6° for what concerns the angular accuracy.

Microplastics (MPs) have received much more attention as a novel breed of pollutant due to their tiny size and difficulty to degrade in natural conditions. MPs are produced from primary sources that are purposefully engineered to be small in magnitude and subsequently discharged into the earth after usage, like resin pellets seen in microplastic factory spills. MPs can also be produced by secondary sources such as the breakdown of macro debris and breakdown of particles, or dust released during the wear and tear of artificial garments, tires, and brake pads. Since rivers meander through municipalities and cities, as well as transporting effluent from plastic-related enterprises and other sewer pollutants into them, which automatically causes MP's contamination in a river to intimately tie to the land environment. Current reports suggest that the amount of plastic trash produced in the upstream watershed can be positively correlated to the amount of plastic waste in the river. While there are currently several sensing approaches for MPs today, there are still several restrictions such as lengthy sensing times, an elevated false sensing rate, and costly sensing apparatus that make detecting MPss in natural environments difficult. Direct, quick, effective, and precise sensing techniques for MPs are direly required to enhance environmental conditions. This study focused on the current analysis methods for the detection of MPs. Direct detection methods for MP pollution in water and sedimentary environments are also outlined. Finally, the toxicological impacts of MPs on aquatic life and other living organisms are discussed.

Multifunctionality is important to the development of next-generation actuators and intelligent robots. However, current multi-functional actuating systems are achieved based on the integration of diverse functional units with complex design, especially lacking in multi-mode sensing and displaying functions. Herein, a light-driven actuator integrated with self-powered/visual dual-mode sensing functions and rewritable display function is proposed. The actuator demonstrates a bending curvature of 0.93 cm-1 under near-infrared light irradiation. Meanwhile, by embedding a pencil-drawn graphite generator and thermochromic materials, the actuator also provides two independent sensing functions. First, owing to the photo-thermoelectric effect of graphite, the actuator spontaneously outputs a self-powered voltage (Seebeck coefficient: 23 μV K-1), which can reflect the deformation trend of actuator. Second, color changes occur on the actuator during deformation, which provide a visual sensing due to the thermochromic property. Furthermore, the actuator can be utilized as a rewritable display, owing to the integrated color-memorizing component. Intelligent robots, switches, and smart homes are further demonstrated as applications. All of them can spontaneously provide self-powered and visual sensing signals to demonstrate the working states of actuating systems, accompanied by rewritable displays on the actuators. This study will open a new direction for self-powered devices, multi-functional actuators, and intelligent robots. [ABSTRACT FROM AUTHOR]

Due to the oscillation behavior of the molten pool, the dynamic change of the molten pool profile has a certain fluctuation. Based on the visual sensing detection method, the oscillation behavior of molten pool in pulsed micro-plasma arc welding of 0.1 mm thick ultra-thin stainless steel was studied, and the influence of different pulse parameters of welding current on the oscillation frequency and amplitude of molten pool was discussed. The results show that the size of molten pool fluctuates slightly during the action of base current and peak current, which is caused by molten pool oscillation. The oscillation frequency of the molten pool increases with the increase of the pulse current frequency in a certain range, while the oscillation amplitude of the molten pool decreases with the increase of the pulse current frequency. The oscillation frequency of the molten pool can reflect the quality of the weld seam forming. [ABSTRACT FROM AUTHOR]

Photoelectrochemical (PEC) is a highly sensitive and fast analytical method that can be used at low concentrations, while photoelectrochromic is a simple and low-cost method primarily utilized for high concentration detection. Therefore, we have developed a dual-signal output sensing platform based on both PEC and photoelectrochromism for rapid and sensitive OTA detection. The sensing platform is divided into signal generation (SG) region and signal output (SO) region, which modified with WO 3 /BiVO 4 photoactive nanocomposites and polyaniline (PANI), respectively. By irradiating the SG region, photogenerated electrons are generated and injected into the SO region through the conductive pathway, resulting in a decrease in surface blue polyaniline and a change to green. The smart device can accurately measure the RGB-Green values, enabling the construction of a photochromic visual sensing platform. After immobilizing the OTA aptamer in the SG region, a linear correlation was observed between the concentration of OTA and the RGB-Green value within the range of 20 ng/L ∼250 μg/L. The detection limit was determined to be 8.33 ng/L (S/N = 3). Furthermore, for a more sensitive OTA detection, a PEC sensing platform was developed utilizing the SG region as a photoanode, exhibiting a linear correlation in the range of 2 pg/L∼300 μg/L with a detection limit of 0.8 pg/L (S/N = 3). The detection of these two modes under the requirement of the international standard for the maximum limit realizes the sensitive OTA detection. The RGB-Green is verified to PEC signal, which improves the detection accuracy. The sensing platform has several advantages and is suitable for various application scenarios. [Display omitted] • The dual-mode sensing platform for OTA detection in diverse scenarios. • Color signal reading for field detection without the bulky signal reading instruments. • The PEC sensing mode for OTA has a wide linear range and a low detection limit. • Prepared self-powered POCT sensing platform based on excellent WO 3 -related materials. [ABSTRACT FROM AUTHOR]

Many amines with high toxicity always cause a serious threat to the ecological environment and human health; thus, their detection is important. Herein, a dual-mode colorimetric and ratiometric fluorescent sensor based on cellulose for detecting amines has been constructed by a new strategy. This sensor is made of a "negative response" indicator (Lum-MDI-CA) and a "positive response" indicator (perylene tetracarboxylic acid, PTCA). Lum-MDI-CA was obtained by attaching luminol onto cellulose chains, which emitted blue fluorescence and was quenched upon contact with amines. A possible mechanism of fluorescence quenching phenomenon is proposed by the intramolecular charge transfer (ICT) of Lum-MDI-CA. Subsequently, by simply mixing Lum-MDI-CA with PTCA, a dual-mode fluorescence sensor was designed for visual detection and classification of amines. When adding ammonia (NH 3), morpholine (MOR), benzylamine (BNZ), diethylamine (DEA), and triethylamine (TEA), respectively, the dual-mode sensor showed visible different color changes under both UV light and daylight. In addition, owing to the excellent processibility and formability of cellulose acetate backbone, the prepared sensor can be easily processed into different material forms, including inks, coatings, films, and fibers, which still exhibit excellent fluorescence emission. Such sensors based on cellulose fluorescent materials are of great value in anti-counterfeiting and information encryption. [Display omitted] [ABSTRACT FROM AUTHOR]

This study presents a novel method for producing polymer dots (PDs) and carbon polymer dots (CPDs) from biomass waste through an environmentally friendly process. This process involves room-temperature oxidation/polymerization and Schiff base reactions, which do not require external energy, making it both energy-efficient and eco-friendly. The CPDs developed in this study show outstanding Fe3+ detection capabilities, with a wide linear range (0–10 μM) and a low detection limit of 0.34 µM. Recovery rates in real sample tests range from 97 % to 103.2 %, with relative standard deviations between 0.63 % and 2.14 %. These findings highlight the method's sustainability, cost-effectiveness, and accuracy. Moreover, the CPDs exhibit reversible fluorescence when activated by ascorbic acid (AA), paving the way for their use in creating fluorescent inks and films with polyvinyl alcohol (PVA). This suggests significant potential for applications in visual sensing and detection. • Biomass material Jacaranda petals were applied to prepare CPDs. • CPDs were synthesized at room-temperature, eliminating the need for external energy sources. • CPDs were applied to the Fe3+ detection, whose limit of detection was 0.34 μM. • CPDs can be used for fluorescent inks, and by combining with PVA, they can achieve visual sensing and detection. [ABSTRACT FROM AUTHOR]

Water-soluble smart materials with multi-stimuli-responsiveness and ultra-long room-temperature phosphorescence (RTP) have garnered broad attention. Herein, a water-soluble terpyridine zinc complex (MeO-Tpy-Zn-OAc), featuring a simple donor-π-acceptor (D-π-A) structure is presented, which responds to a variety of stimuli, including changes in solvents, pH, temperature, and the addition of amino acids. Notably, MeO-Tpy-Zn-OAc functions as a fluorescence probe, capable of visually and selectively discriminating aspartate or histidine among other common amino acids in water. Additionally, when incorporated into polyvinyl alcohol (PVA) to form the composite MeO-Tpy-Zn-OAc@PVA, the material exhibits reversible writing, photochromism, and a prolonged RTP with a 14 s afterglow. These unique properties enable the composite to be utilized in potential applications such as secure data encryption and inkless printing., (© 2024 Wiley‐VCH GmbH.)

There is a need for the effective weld seam profile extraction method to realize the automatic and intelligent welding process with thick steel plates based on laser vision sensing. In this paper, a method was proposed to identify the variable weld seam profiles from the strong arc background based on intensity mutation and density feature detection for the thick plate welding process with T-joints. First, an improved Canny algorithm was used to magnify the weld seam profile and restrain interference. Next, an intensity mutation detection method was proposed to further strengthen the weld seam profile because there exists intensity mutation in the local region surrounding the weld seam profile. Finally, an algorithm based on the band-width and density feature scanning method was proposed to further eliminate the interference data after the strengthened image was binarized with the Otsu algorithm. The weld seam profile was identified as clusters with their spatial scale features after the nearest neighbor clustering was dealt with. The results show that this method can identify over 95% of the weld seam profiles from the arc interference background whose area is about 20% of the image. It provides valuable reference for promoting the automatic and intelligent welding process with different joints.

Developing sensitive, nondestructive and real-time methods for assessing meat freshness is vital for ensuring food safety. Traditional monochromatic fluorescent probes for monitoring meat freshness markers like biogenic amines often face challenges in sensitivity, interference, visualization and portability for practical applications. Herein, we constructed a new Ru-Flu@COF fluorescence sensor, by combining a green fluorescent dye-encapsulated covalent organic framework (Flu@COF) with a red fluorescent compound (Ru (bpy) 3 2+), exhibiting a dual inverse ratiometric fluorescence signal towards essential biogenic amines His and NH 3 ·H 2 O, thus achieving a rich color gradient and spoilage-sensitive color discrimination. The detection linear range for His and NH 3 ·H 2 O was 0.6–200 μM and 0.5–60 mg/L, with limits of detection of 0.5 μM and 0.2 mg/L, respectively. Furthermore, Ru-Flu@COF gel labels together with a smartphone enable portable, in situ and naked-eye monitoring freshness of pork, fish and shrimp, where the freshness level can be visually identified through the gel label color from red (fresh) to orange (less fresh), yellow (spoilage) and green (severe spoilage). The proposed smart label has the advantages of high sensitivity, excellent visual discrimination and portability, and it can be a promising platform for in situ nondestructive monitoring freshness of various kinds of meat for food safety inspection in the future. [Display omitted] • A dual-response ratiometric fluorescence sensor was built to monitor meat freshness. • Portable COF-based hydrogel allowed for gas-phase sensing of volatile amines in meat. • Multicolor fluorescence response was rapidly detected and quantified via smartphone. • Real-time, in-situ and non-destructive fast screening of meat freshness was realized. [ABSTRACT FROM AUTHOR]

A highly sensitive flexible sweat sensor MnO 2 -TMB@PAM aggregates dual-detection routes of visualization and UV–vis spectrometer, MnO 2 NPs were in-situ synthesized in PAM as the tandem enzyme-like catalyzes, and TMB was modified as the colorimetric indicator. [Display omitted] • In-situ MnO 2 NPs within PAM offer high stability and contact area of reactors, leading to improved catalytic efficiency. • BSA-MnO 2 NPs exhibited dual enzyme-like activities, enabling glucose detection through the catalysis of a single nanozyme. • The flexibility and transmittance patch enable rapid visual detection, performed as a potential candidate for POCT devices. The emergence of flexible, painless sensors for biomolecule measurement represents a growing trend in biomedical research. In this context, a highly sensitive flexible sweat sensor, incorporating MnO 2 -TMB@PAM aggregates employing dual-detection routes via visualization and UV–vis spectrometry, has been developed. The sensor utilizes bovine serum albumin (BSA) templated MnO 2 nanoparticles (NPs) as a catalyst for glucose detection. Unlike natural enzymes, this nanozyme is not only cost-effective and stable but also exhibits tandem enzyme-like activities, which facilitates quick and straightforward glucose detection. MnO 2 NPs are synthesized in situ on the surface of polyacrylamide (PAM), ensuring high stability and increasing the contact area of reactors, thereby enhancing the catalytic efficiency of MnO 2 NPs. Additionally, 3,3′,5,5′-tetramethylbenzidine (TMB) is modified on the hydrogel to create an integrated sensing patch (referred to as MnO 2 -TMB@PAM). This patch enables fast and convenient sensing routes, positioning it as a potential candidate for point-of-care testing (POCT) devices. Due to the excellent flexibility and transmittance of PAM, the sensing patch allows for rapid colorimetric analysis of biomolecules such as glucose and dopamine via UV–vis spectrophotometry, as well as visual detection with the naked eyes. Consequently, a high-sensitive, selective, and visually active flexible sensor has been established based on colorimetric and visual methods, facilitating efficient sensing of biomarkers from in vitro biological fluids. [ABSTRACT FROM AUTHOR]

Wireless visual sensor networks have been adopted in different contexts to provide visual information in a more flexible and distributed way, supporting the development of different innovative applications. Although visual data may be central for a considerable set of applications in areas such as Smart Cities, Industry 4.0, and Vehicular Networks, the actual visual data quality may be not easily determined since it may be associated with many factors that depend on the characteristics of the considered application scenario. This entails several aspects from the quality of captured images (sharpness, definition, resolution) to the characteristics of the networks such as employed hardware, power consumption, and networking efficiency. In order to better support quality analysis and performance comparisons among different wireless visual sensor networks, which could be valuable in many monitoring scenarios, this article surveys this area with special concern on assessment mechanisms and quality metrics. In this context, a novel classification approach is proposed to better categorize the diverse applicable metrics for quality assessment of visual monitoring procedures. Hence, this article yields a practical guide for analyzing different visual sensor network implementations, allowing fairer evaluations and comparisons among a variety of research works. Critical analysis are also performed regarding the relevance and usage of the proposed categories and identified quality metrics. Finally, promising open issues and research directions are discussed in order to guide new developments in this research field. [ABSTRACT FROM AUTHOR]

Grasping is critical for intelligent robots to accomplish sophisticated tasks. Even with multimodal sensor fusion, accurately and reliably estimating grasp poses for complex-shaped objects remains a challenge. In this paper, we design a vision-based grasping platform for a more general case, that is, grasping a variety of objects by a simple parallel gripper with the grasp detection model consuming RGB sensing or depth sensing. Focusing on the grasp pose estimation part, we propose a deep grasp detector that uses a densely connected Feature Pyramid Network (FPN) feature extractor and multiple two-stage detection units to achieve dense grasp pose predictions. Specifically, for the feature extractor, the fusion of different layer feature maps can increase both the model’s capacity to detect the various size grasp areas and the accuracy of the regressed grasp positions. For each of the two-stage detection unit, the first stage generates horizontal candidate grasp areas, while the second stage refines them to predict the rotated grasp poses. We train and validate our grasp pose estimation algorithm on the Cornell Grasp Dataset and the Jacquard Dataset. The model achieves the detection accuracy of 93.3% and 89.6%, respectively. We further design real-world grasp experiments to verify the effectiveness of our vision-based robotic grasping system. The real scenario trials validate that the system is capable of grasping unseen objects, in particular, achieving robust and accurate grasp pose detection and gripper opening width measurement based on depth sensing only. [ABSTRACT FROM AUTHOR]

In the monsoon season, sudden flood events occur frequently in urban areas, which hamper the social and economic activities and may threaten the infrastructure and lives. The use of an efficient large-scale waterlogging sensing and information system can provide valuable near real-time disaster information to facilitate disaster management and enhance awareness of the general public to alleviate losses during and after flood disasters. Therefore, in this study, a visual sensing approach driven by deep neural networks and information and communication technology was developed to provide an end-to-end mechanism to realize waterlogging sensing and event-location mapping. The use of a deep sensing system in the monsoon season in Taiwan was demonstrated, and waterlogging events were predicted on the island-wide scale. The system could sense approximately 2379 vision sources through an internet of video things framework and transmit the event-location information in 5 min. The proposed approach can sense waterlogging events at a national scale and provide an efficient and highly scalable alternative to conventional waterlogging sensing methods.

To solve the current problem of poor weld formation due to groove width variation in swing arc narrow gap welding, an infrared passive visual sensing detection approach was developed in this work to measure groove width under intense welding interferences. This approach, called global pattern recognition, includes self-adaptive positioning of the ROI window, equal division thresholding and in situ dynamic clustering algorithms. Accordingly, the self-adaptive positioning method filters several of the nearest values of the arc's highest point of the vertical coordinate and groove's same-side edge position to determine the origin coordinates of the ROI window; the equal division thresholding algorithm then divides and processes the ROI window image to extract the groove edge and forms a raw data distribution of groove width in the data window. The in situ dynamic clustering algorithm dynamically classifies the preprocessed data in situ and finally detects the value of the groove width from the remaining true data. Experimental results show that the equal division thresholding algorithm can effectively reduce the influences of arc light and welding fume on the extraction of the groove edge. The in situ dynamic clustering algorithm can avoid disturbances from simulated welding spatters with diameters less than 2.19 mm, thus realizing the high-precision detection of the actual groove width and demonstrating stronger environmental adaptability of the proposed global pattern recognition approach. [ABSTRACT FROM AUTHOR]

Traffic lights detection and recognition (TLDR) is one of the necessary abilities of multi-type intelligent mobile platforms such as drones. Although previous TLDR methods have strong robustness in their recognition results, the feasibility of deployment of these methods is limited by their large model size and high requirements of computing power. In this paper, a novel lightweight TLDR method is proposed to improve its feasibility to be deployed on mobile platforms. The proposed method is a two-stage approach. In the detection stage, a novel lightweight YOLOv5s model is constructed to locate and extract the region of interest (ROI). In the recognition stage, the HSV color space is employed along with an extended twin support vector machines (TWSVMs) model to achieve the recognition of multi-type traffic lights including the arrow shapes. The dataset, collected in naturalistic driving experiments with an instrument vehicle, is utilized to train, verify, and evaluate the proposed method. The results suggest that compared with the previous YOLOv5s-based TLDR methods, the model size of the proposed lightweight TLDR method is reduced by 73.3%, and the computing power consumption of it is reduced by 79.21%. Meanwhile, the satisfied reasoning speed and recognition robustness are also achieved. The feasibility of the proposed method to be deployed on mobile platforms is verified with the Nvidia Jetson NANO platform.

Mechanoluminescent (ML) fibers and textiles enable stress visualization without auxiliary power, showing great potential in wearable electronics, machine vision, and human-computer interaction. However, traditional ML devices suffer from inefficient stress transfer in soft-rigid material systems, leading to low luminescence brightness and short cycle life. Here, we propose a tendon-inspired scale-bridging mechanics transfer mechanism for ML composites, which employs molecular-scale copolymerized cross-linking and nanoscale inorganic nanoparticles as hierarchical stress transfer sites. This strategy effectively reduces the dissipation of stress in molecular chain segments and alleviates local stress concentration, increases luminescence by 9 times, and extends cycle life to more than 10,000 times. Furthermore, a scalable (kilometer-scale) anti-Plateau-Rayleigh instability manufacturing technology is developed for thermoset ML fibers, compatible with various existing textile techniques. We also demonstrate its system-level applications in motion capture, underwater interaction, etc. , providing a feasible strategy for the next generation of smart visual textiles.

As a flexible passive 3D sensing means, unsupervised learning of depth from monocular videos is becoming an important research topic. It utilizes the photometric errors between the target view and the synthesized views from its adjacent source views as the loss instead of the difference from the ground truth. Occlusion and scene dynamics in real-world scenes still adversely affect the learning, despite significant progress made recently. In this paper, we show that deliberately manipulating photometric errors can efficiently deal with these difficulties better. We first propose an outlier masking technique that considers the occluded or dynamic pixels as statistical outliers in the photometric error map. With the outlier masking, the network learns the depth of objects that move in the opposite direction to the camera more accurately. To the best of our knowledge, such cases have not been seriously considered in the previous works, even though they pose a high risk in applications like autonomous driving. We also propose an efficient weighted multi-scale scheme to reduce the artifacts in the predicted depth maps. Extensive experiments on the KITTI dataset and additional experiments on the Cityscapes dataset have verified the proposed approach’s effectiveness on depth or ego-motion estimation. Furthermore, for the first time, we evaluate the predicted depth on the regions of dynamic objects and static background separately for both supervised and unsupervised methods. The evaluation further verifies the effectiveness of our proposed technical approach and provides some interesting observations that might inspire future research in this direction. [ABSTRACT FROM AUTHOR]

Wire plus arc additive manufacturing (WAAM) has been demonstrated to be a powerful technique to produce large-scale metal parts with low cost. However, techniques to achieve accurate geometry control and high process stability have not yet been perfectly developed. Although implementing vision sensing and closed-loop control can contribute to promoting the levels of process automation and stability, it is difficult to markedly improve the geometry precision of parts by only performing the current layer detection due to a large detection lag with vision-based sensors. To deal with this issue, this study proposes a novel strategy of introducing the previous layer information into the current deposition height to increase the response speed of the control system. The previous and current layer heights are monitored by a passive vision sensor. The height features are extracted by image processing algorithms mainly including edge detection, threshold division, and line fitting. Deviations in deposition height are automatically compensated via controlling the wire feed speed based on a PID controller. A helpful software interface is implemented in the Visual C++ environment to study the automatic detection and control system. In comparison to the closed-loop control using only the current layer detection, the deposition height of thin-walled parts can be excellently controlled by the proposed control system using the visual inspection of previous and current layers, significantly increasing the process stability and achieving accurate height control in WAAM. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Fuzhou University is the property of Journal of Fuzhou University, Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

A simple method was developed in this work for facile and visual detection of S2− using a paper‐based fluorescence (FL) sensor coated with CdTe quantum dots (QDs) by headspace sampling. With the addition of hydrochloric acid, the target S2− in the liquid phase would transform to H2S, which was released to headspace and quenched the FL of CdTe QDs in a linear manner through a gas–solid reaction, with any possible liquid‐phase interference avoided. The regular quenching caused by S2− in analyte solution with increased concentration could be easily observed by the naked eye, and the limit of detection (LOD) for this method was 0.13 μM and 0.93 μM for FL and visual sensing, respectively, comparable or not to that by other sensing probes. A relative standard deviation of 1.2% was accomplished from seven replicated measurements, implying the high reproducibility, and the recovery for the spiked water samples ranging from 94 to 103%, and illustrating the satisfactory reliability of this method. Moreover, the preparation of this paper sensor was facile and did not require any complicated or time‐consuming procedures for additional modification or functionalization as for other probes previously reported. [ABSTRACT FROM AUTHOR]

Generic person tracking is a basic task in visual surveillance by using camera as sensors. Many deep learning-based trackers have obtained outstanding performance. Among them, trackers based on Siamese networks have drawn great attention and are promising. These training methods are competitive but training data can be more effectively augmented to improve their person tracking performance. Many other trackers use only one layer to extract semantic features, likely hindering their discriminative learning. In this paper, we propose an enhanced discriminative model prediction method with efficient data augmentation and robust feature fusion. Specifically, we propose to implement an effective data augmentation strategy (e.g., color jitter and motion blur) to unleash the greater potential of original training data. We also adopt a multi-layer feature fusion to obtain a more discriminative feature map. Thus, the proposed tracker can discriminate an object in complicated scenarios in real time. We conduct extensive experiments on two datasets, i.e., VOT2018 and UAV123. Objective evaluation on VOT2018 demonstrates that with its expected average overlap value of 0.430, it outperforms a state-of-the-art tracker by 4.88%. On UAV123, it does so by 4.5% in success rate and 4.4% in precision rate. In addition, our further experimental results reveal that our algorithm can reach a speed that is high enough to meet the real-time tracking requirement when camera are used as sensors. [ABSTRACT FROM AUTHOR]

Wireless sensor networks comprising nodes equipped with cameras have become common in many scenarios, providing valuable visual data for some relevant services such as localization, tracking, patterns identification and emergencies detection. In this context, algorithms and optimization approaches have been designed to perform different types of quality assessment or performance enhancement tasks, addressing challenging issues such as networking, compression, availability, reliability, security, energy efficiency and virtually any subject related to the operational challenges of those networks. However, the dynamics of coverage failures have not been properly modelled in visual sensor networks, resulting in unrealistic perceptions when optimizing or assessing quality in most visual sensing scenarios. Particularly, the Field of View of visual sensors will be affected by occlusion caused by obstacles in the monitored field, which may turn such sensors inadequate for the expected monitoring services of the considered network. Therefore, this article proposes a mathematical model to assess occlusion caused by mobile obstacles such as vehicles on a road or forklifts in an industrial plant, aiming at the selection of the visual sensor nodes that will not have their coverage significantly restricted by those obstacles. Doing so, the proposed model can be exploited by any optimization or quality assessment approach in wireless visual sensor networks, providing a preprocessing method when selecting visual nodes.

A cost-effective, facile, and sensitive fluorescence sensing strategy for Pb2+ ion detection has been developed based on the fluorescence resonance energy transfer (FRET) between carbon quantum dots (CQDs) and Au nanoparticles (NPs). Glutathione (GSH)-synthesized CQDs acted as both the fluorescence donor and the sorbent to extract Pb2+ ions from the solution via Pb-GSH complexes. Pb2+ ions on CQDs reacted with –SH groups on AuNPs to generate sandwich-type Au-PdS-CQDs, leading to a dramatic decrease in the fluorescence of the CQDs. To expand the potential applications of this strategy, we constructed a sensing strategy using self-organized TiO2 nanotube arrays (TiNTs). The high aspect ratio and transparency for light emitted from the CQDs enabled the TiNTs to serve as a sensitive solid visual platform for the highly selective detection of Pb2+ ions with a detection limit as low as 4.1 × 10−8 mg mL−1. More importantly, the long observation length combined with a small volume enabled a sample acquisition volume of only 2.1 × 10−3 μL, which is smaller than the traditional fluorescence analysis in solution and on commercially available test paper, thus endowing this visual platform with the potential for use in single-cell diagnostics. [ABSTRACT FROM AUTHOR]