Search Results (20,882)

Few-Shot Industrial Anomaly Detection (FSIAD) is an essential yet challenging problem in practical scenarios such as industrial quality inspection. Its objective is to identify previously unseen anomalous regions using only a limited number of normal support images from the same category. Recently, large pre-trained vision-language models (VLMs), such as CLIP, have exhibited remarkable few-shot image-text representation abilities across a range of visual tasks, including anomaly detection. Despite their promise, real-world industrial anomaly datasets often contain noisy labels, which can degrade prompt learning and detection performance. In this paper, we propose AnomalyNLP, a new Noisy-Label Prompt Learning approach designed to tackle the challenge of few-shot anomaly detection. This framework offers a simple and efficient approach that leverages the expressive representations and precise alignment capabilities of VLMs for industrial anomaly detection. First, we design a Noisy-Label Prompt Learning (NLPL) strategy. This strategy utilizes feature learning principles to suppress the influence of noisy samples via Mean Absolute Error (MAE) loss, thereby improving the signal-to-noise ratio and enhancing overall model robustness. Furthermore, we introduce a prompt-driven optimal transport feature purification method to accurately partition datasets into clean and noisy subsets. For both image-level and pixel-level anomaly detection, AnomalyNLP achieves state-of-the-art performance across various few-shot settings on the MVTecAD and VisA public datasets. Qualitative and quantitative results on two datasets demonstrate that our method achieves the largest average AUC improvement over baseline methods across 1-, 2-, and 4-shot settings, with gains of up to 10.60%, 10.11%, and 9.55% in practical anomaly detection scenarios. Full article

Games are considered a suitable and standard benchmark for checking the performance of artificial intelligence-based algorithms in terms of training, evaluating, and comparing the performance of AI agents. In this research, an application of the Intrinsic Curiosity Module (ICM) and the Asynchronous Advantage Actor–Critic (A3C) algorithm is explored using action games. Having been proven successful in several gaming environments, its effectiveness in action games is rarely explored. Providing efficient learning and adaptation facilities, this research aims to assess whether integrating ICM with A3C promotes curiosity-driven explorations and adaptive learning in action games. Using the MAME Toolkit library, we interface with the game environments, preprocess game screens to focus on relevant visual elements, and create diverse game episodes for training. The A3C policy is optimized using the Proximal Policy Optimization (PPO) algorithm with tuned hyperparameters. Comparisons are made with baseline methods, including vanilla A3C, ICM with pixel-based predictions, and state-of-the-art exploration techniques. Additionally, we evaluate the agent’s generalization capability in separate environments. The results demonstrate that ICM and A3C effectively promote curiosity-driven exploration in action games, with the agent learning exploration behaviors without relying solely on external rewards. Notably, we also observed an improved efficiency and learning speed compared to baseline approaches. This research contributes to curiosity-driven exploration in reinforcement learning-based virtual environments and provides insights into the exploration of complex action games. Successfully applying ICM and A3C in action games presents exciting opportunities for adaptive learning and efficient exploration in challenging real-world environments. Full article

This study explores and models the use of MATLAB technology in multimodal learning interactions to address the challenges of teaching and learning statistics in a multinational postgraduate classroom. The term multimodal refers to the deliberate integration of multiple representational and interaction modes, i.e., visual, textual, symbolic, and interactive computational modelling, within a coherent instructional design. MATLAB is utilised as it is a comprehensive tool for enhancing students’ understanding of statistical skills, practical applications, and data analysis—areas where traditional methods often fall short. International postgraduate students were chosen for this study because their diverse educational backgrounds present unique learning challenges. A qualitative case study design was employed, and data collection methods included classroom observations, interviews, and student work analysis. The collected data were analysed and modelled by conceptualising key elements and themes using thematic analysis, with findings verified through data triangulation and expert review. Emerging themes were structured into models that illustrate multimodal teaching and learning interactions. The novelty of this research lies in its contribution to multimodal teaching and learning strategies for multinational students in statistics education. The findings highlight significant challenges international students face, including language and technical barriers, limited prior content knowledge, time constraints, technical difficulties, and a lack of independent thinking. To address these challenges, MATLAB promotes collaborative learning, increases student engagement and discussion, boosts motivation, and develops essential skills. This study suggests that educators integrate multimodal interactions in their teaching strategies to better support multinational students in statistical learning environments. Full article

Background/Objectives: Vestibular schwannoma (VS) is a benign cerebellopontine angle tumor causing audiological and vestibular symptoms. This pilot study aimed to describe the application of video Head Impulse Test (vHIT) in the diagnosis and follow-up of patients with unilateral VS treated surgically. The objective was to describe a detailed interpretation of vHIT—not only numerical parameters such as gain and corrective saccades, but also a visual analysis of vHIT curves. Methods: The results were presented in four cases for better understanding and more straightforward explanation. The patients underwent surgery through the middle cranial fossa and translabyrinthine approach. In each patient, vHIT examinations were performed preoperatively and at one month, three months, and one year after the surgery. Results: Before treatment, vestibular loss features varied within the presented cases. Findings of vestibulo-ocular reflex deficiency were most pronounced in the lateral semicircular canals. After the surgery, severe signs of acute labyrinth denervation were found during the first follow-up visit. Over time, features indicating central compensation became more pronounced, despite a decrease in gain in subsequent vHIT examinations. Conclusions: Detailed analysis of vHIT curves is crucial to analyze vestibulo-ocular reflex in patients with VS. Our preliminary data confirms that vHIT examination can be helpful in the postoperative follow-up assessment and compensation evaluation. Full article

Strabismus, or squint or deviating eyes, is defined as misalignment of the eyes when fixating on an object and is a common problem in ophthalmology. Palsy of the third, fourth or sixth cranial nerve is one of the leading underlying causes for paralytic strabismus, often requiring surgery. However, uncertainty regarding factors influencing surgical success remains. Background/Objectives: The purpose of this study is to review the outcome and influencing factors of strabismus surgery in patients with cranial nerve palsy. Methods: A retrospective study of 57 patients with third cranial nerve (CN3) palsy, fourth cranial nerve (CN4) palsy, sixth cranial nerve (CN6) palsy or combined nerve palsy who underwent strabismus surgery between October 2009 and December 2023 was conducted. Analyzed data included demographic details, type of surgical intervention, etiology of nerve palsy, pre- and postoperative angle of deviation (AOD), vertical deviation (VD), best-corrected visual acuity (BCVA), and refractive error. Results: Mean age was 41.29 ± 23.14 years with a mean follow-up of 10.8 ± 15.38 months. 30 patients (52.63%) had CN6 palsy, 12 patients (21.05%) had CN3 palsy, eight patients (14.04%) had CN4 palsy and seven patients (12.28%) had combined nerve palsy. Brain neoplasm was the most common cause of nerve palsy (33.33%). Mean preoperative AOD improved from 17.54° ± 10.68 to 7.13° ± 8.93 and from 17.21° ± 9.58 to 7.49° ± 9.75 for near and distance, respectively (p < 0.001). Changes in VD, refractive error, and BCVA were not statistically significant. Conclusions: Age, gender, preoperative AOD, subtype and etiology of nerve palsy had no significant influence on surgical outcomes, which are satisfactory in patients with cranial nerve palsy (80.7%). Full article

Background: In this study, hydroponic experiments were conducted to examine the roles of sulphur (S), calcium (Ca), and nitric oxide (NO) in alleviating salt stress (20 mM NaCl) in tomato (Solanum lycopersicum L.) seedlings. Methods: Analyses included Na⁺/K⁺ contents, inorganic nitrogen (nitrate, nitrite, ammonium), nitrogen- and ammonium-assimilating enzymes (NR, NiR, GS, GOGAT), sulphur-assimilating enzymes (ATPS, OASTL), protein content, ROS (O₂^∙−, H₂O₂), and in vivo NO visualization were conducted. Results: We observed that salt stress increased Na⁺, reduced K⁺, disrupted nitrogen and sulphur metabolism, elevated ROS, and decreased NO, causing oxidative stress and reduced enzymatic activity. Supplementation with potassium sulphate (40 µM), calcium chloride (30 µM), and sodium nitroprusside (SNP; 40 µM) mitigated these effects, enhancing enzymatic activities, restoring Na⁺/K⁺ balance, improving protein content, and lowering ROS. The protective role of NO was confirmed using inhibitors L-NAME (500 µM) and cPTIO (100 µM), which reversed SNP’s benefits and aggravated stress damage. Conclusion: Overall, S, Ca, and NO were found to synergistically improve salt stress tolerance by modulating ion homeostasis, nitrogen and sulphur metabolism, and oxidative balance, offering nutrient- and signal-based strategies to enhance tomato resilience under salinity. Full article

Display-field communication (DFC) is an imperceptible screen-to-camera technology that embeds and recovers data from the frequency domain of an image frame. Conventional DFC requires a reference frame for each data frame to estimate the channel, a method that, while reliable, is not bandwidth-efficient. Similarly, iterative DFC requires the transmission of pilot symbols for channel estimation. In this paper, we propose an implicit DFC (iDFC) scheme that eliminates the need for reference frames by estimating them using the first-order statistics of the received image. The system employs discrete Fourier-transform-based subcarrier mapping and adds data directly to the frequency coefficients of the host image. At the receiver, statistical estimation enables blind channel equalization without sacrificing the data rate. The simulation results show that iDFC achieves an achievable data rate (ADR) of up to $1.52\times {10}^5$ bps, a significant enhancement of approximately 97% and 11% compared to conventional and iterative DFC schemes, respectively. Furthermore, the analysis reveals a critical trade-off between communication robustness and visual imperceptibility; allocating 70% of signal power to the image maintains high visual quality but results in a symbol error rate (SER) floor of $1.5\times {10}^{-1}$, whereas allocating only 10% improves the SER to below ${10}^{-2}$ at the cost of visible artifacts. The findings also identify QPSK as the optimal modulation order that maximizes the data rate, showing that higher-order schemes can be detrimental due to system impairments such as signal clipping. The proposed iDFC scheme presents a more efficient and robust solution for high-capacity DFC applications by balancing the competing demands of data throughput and visual fidelity. Full article

This study explores the influence of the urban soundscape on pedestrian perception and walking experience in the historic and lower parts of Béjaïa, Algeria. More precisely, the analysis investigates how variations in soundscape configuration relate to perceived comfort, safety, and walking pleasantness across five morphologically distinct urban zones. A mixed-method approach combining quantitative tools (LAeq acoustic measurements) and qualitative methods (soundwalks, sound diaries, and mental maps) was applied in accordance with ISO 12913. The study involved 50 participants for the sound diaries and 58 for the soundwalks. Results show that natural and social sounds enhance perceived comfort and safety, while mechanical noise is associated with discomfort and avoidance behaviors. In the morning, moderate to strong correlations were observed between sound comfort and visual perception (ρ = 0.58, p = 0.001, 95% CI [0.27; 0.80]), as well as between sound comfort and walking pleasantness (ρ = 0.40, p = 0.033, 95% CI [0.05; 0.67]). The study highlights the need to integrate soundscape considerations into urban planning and heritage conservation strategies. Full article

Collision avoidance technology serves as a critical enabler for autonomous navigation of unmanned surface vehicles (USVs). To address the limitations of incomplete environmental perception and inefficient decision-making for collision avoidance in USVs, this paper proposes an autonomous collision avoidance method based on deep reinforcement learning. To overcome the restricted field of view of USV perception systems, visual assistance from an unmanned aerial vehicle (UAV) is introduced. Perception data acquired by the UAV are utilized to construct a high-dimensional state space that characterizes the distribution and motion trends of obstacles, while a low-dimensional state space is established using the USV’s own state information, together forming a hierarchical state space structure. Furthermore, to enhance navigation efficiency and mitigate the sparse-reward problem, this paper draws on the trajectory evaluation concept of the dynamic window approach (DWA) to design a set of process rewards. These are integrated with COLREGs-compliant rewards, collision penalties, and arrival rewards to construct a multi-dimensional reward function system. To validate the superiority of the proposed method, collision avoidance experiments are conducted across various scenarios. The results demonstrate that the proposed method enables USVs to achieve more efficient autonomous collision avoidance, indicating strong potential for engineering applications. Full article

Converting tower-mounted videos from perspective to orthographic view is beneficial for their integration with maps and remote sensing images and can provide a clearer and more real-time data source for earth observation. This paper addresses the issue of low geometric accuracy in orthographic video generation by proposing a method that incorporates 3D GIS view matching. Firstly, a geometric alignment model between video frames and 3D GIS views is established through camera parameter mapping. Then, feature point detection and matching algorithms are employed to associate image coordinates with corresponding 3D spatial coordinates. Finally, an orthographic video map is generated based on the color point cloud. The results show that (1) for tower-based video, a 3D GIS constructed from publicly available DEMs and high-resolution remote sensing imagery can meet the spatialization needs of large-scale tower-mounted video data. (2) The feature point matching algorithm based on deep learning effectively achieves accurate matching between video frames and 3D GIS views. (3) Compared with the traditional method, such as the camera parameters method, the orthographic video map generated by this method has advantages in terms of geometric mapping accuracy and visualization effect. In the mountainous area, the RMSE of the control points is reduced from 137.70 m to 7.72 m. In the flat area, it is reduced from 13.52 m to 8.10 m. The proposed method can provide a near-real-time orthographic video map for smart cities, natural resource monitoring, emergency rescue, and other fields. Full article

In this study, the extended (3 + 1)-dimensional Jimbo–Miwa equation, which has not been previously studied using Lie symmetry techniques, is the focus. We derive new symmetry reductions and exact invariant solutions, including lump and rogue wave structures. Additionally, precise solitary wave solutions of the extended (3 + 1)-dimensional Jimbo–Miwa equation using the multivariate generalized exponential rational integral function technique (MGERIF) are studied. The extended (3 + 1)-dimensional Jimbo–Miwa equation is crucial for studying nonlinear processes in optical communication, fluid dynamics, materials science, geophysics, and quantum mechanics. The multivariate generalized exponential rational integral function approach offers advantages in addressing challenges involving exponential, hyperbolic, and trigonometric functions formulated based on the generalized exponential rational function method. The solutions provided by MGERIF have numerous applications in various fields, including mathematical physics, condensed matter physics, nonlinear optics, plasma physics, and other nonlinear physical equations. The graphical features of the generated solutions are examined using 3D surface graphs and contour plots, with theoretical derivations. This visual technique enhances our understanding of the identified answers and facilitates a more profound discussion of their practical applications in real-world scenarios. We employ the MGERIF approach to develop a technique for addressing integrable systems, providing a valuable framework for examining nonlinear phenomena across various physical contexts. This study’s outcomes enhance both nonlinear dynamical processes and solitary wave theory. Full article

Accurate differentiation between Coilia brachygnathus and Coilia nasus is imperative for the effective management of fisheries, the conservation of aquatic ecosystems, and the mitigation of commercial fraud. Current morphological identification remains challenging due to their high morphological similarity—particularly for processed samples—while conventional molecular methods often lack the speed or specificity required for field applications or high-throughput screening. In this study, a novel integrated approach was developed and validated, combining TaqMan quantitative real-time PCR (qPCR). for precise genotyping of C. brachygnathus and C. nasus with Recombinase Polymerase Amplification coupled with Lateral Flow Dipstick (RPA-LFD) for rapid on-site screening. First, species-specific RPA-LFD assays were designed to target the mitochondrial COI gene sequence. This enabled visual detection within 10 min at 37 °C, with a sensitivity of 10² copies/μL, and required no complex equipment. A dual TaqMan MGB qPCR assay was further developed by validating stable differentiating SNPs (chr21:3798155, C/T) between C. brachygnathus and C. nasus, using FAM/VIC dual-labeled MGB probes. Results showed that this assay could distinguish the two species in a single tube: for C. brachygnathus, Ct values in the FAM channel were significantly earlier than those in the VIC channel (ΔCt ≥ 1), with a FAM detection limit of 125 copies/reaction; for C. nasus, only VIC channel amplification was observed, with a detection limit as low as 12.5 copies/reaction. Validation with 171 known tissue samples demonstrated 100% concordance with expected species identities. This integrated approach effectively combines the high accuracy and quantitative capacity of TaqMan qPCR for confirmatory laboratory genotyping with the speed, simplicity, and portability of RPA-LFD for initial field or point-of-need screening. This reliable, efficient, and user-friendly technique provides a powerful tool for resource management, biodiversity monitoring, and ensuring the authenticity of high-quality C. brachygnathus and C. nasus. Full article

Background: Macular edema (ME) secondary to retinal vein occlusion (RVO) is a significant cause of vision impairment. Many patients show suboptimal responses to anti-vascular endothelial growth factor (anti-VEGF) monotherapy, prompting the exploration of alternative treatments. Faricimab is a bispecific antibody that targets VEGF-A and angiopoietin-2. We report 9-month real-world outcomes of switching to faricimab in therapy-resistant RVO-associated ME. Methods: In this retrospective study at a single tertiary center, patients with persistent or recurrent ME despite prior treatments (ranibizumab, aflibercept, or dexamethasone implant) were switched to faricimab. All eyes received a loading phase of four monthly faricimab injections, followed by a treat-and-extend regimen individualized per response. Key outcomes included best-corrected visual acuity (BCVA, logMAR), the central subfield thickness (CST, μm), and the intraretinal fluid (IRF) status on optical coherence tomography, assessed from the baseline (month 0, mo0) through the loading phase (mo1–mo3) and at month 9 (mo9). Results: Nineteen eyes (19 patients, mean age 64.8 years) were analyzed. The median BCVA improved from 0.20 to 0.00 logMAR by month 3 (p < 0.01) and was maintained at month 9. The median CST decreased from 325 μm at the baseline to 285 μm at month 3 (p < 0.01) and remained at 285 μm at month 9. IRF was present in 100% of eyes at the baseline, 26% at month 3, and 26% at month 9 (p < 0.01 for the baseline vs. month 9). Among eyes previously on anti-VEGF therapy (n = 14), the median treatment interval increased from 45.50 days at the baseline to 56.50 days at month 9 (p = 0.01; δ = 0.86). No intraocular inflammation or other adverse events were observed in this cohort over nine months. Conclusions: In this retrospective series, switching to faricimab was associated with improvements in vision and retinal anatomy that were maintained over 9 months; injection intervals were extended in a subset of eyes. These exploratory findings warrant confirmation in larger, controlled studies to define long-term effectiveness, safety, and dosing strategies. Full article

Background: Fulminant idiopathic intracranial hypertension (IIH) is a rare and vision-threatening variant of IIH, characterized by rapid visual deterioration and a high risk of irreversible blindness. Urgent intervention is required to prevent permanent optic nerve damage. Cerebral transverse venous stenting (CTVS) has emerged as an effective treatment for medically refractory IIH, but data on its use in fulminant cases remain limited. Methods: A retrospective consecutive cohort study was conducted at a tertiary center and included all patients with fulminant IIH diagnosed by modified Dandy criteria, with bilateral transverse sinus stenosis > 50% and a trans-stenotic pressure gradient ≥ 8 mmHg on venography. Before stenting, patients received high-dose acetazolamide (up to 3000 mg/day) and IV methylprednisolone (1000 mg/day × 3). Neuro-ophthalmic assessment included BCVA, Ishihara color vision, pupillary exam, disc edema grading, Humphrey visual fields, and optical coherence tomography (OCT). Follow-up occurred at baseline (admission), 1 week, 1 month, 3 months, and 12 months. Results: Five young female patients underwent successful CTVS without peri- or post-procedural complications. Significant improvement in headache and stabilization or recovery of visual function were observed in all patients. OCT revealed early retinal nerve fiber layer thinning within one week, preceding clinical resolution of papilledema. Conclusions: Emergent CTVS appears to be a safe and effective vision-preserving procedure in fulminant IIH, offering rapid intracranial pressure reduction and early neuro-ophthalmologic improvement. OCT may serve as a useful early predictor of treatment success, supporting its role in post-procedural monitoring. Larger prospective studies are warranted. Full article

Introduction: The management of neovascular age-related macular degeneration (nAMD) is constrained by diminishing therapeutic options for retina specialists and their patients when the disease reaches its end stages. Methods: Clinical insights emerge from two case narratives in which patients benefitted from discontinuation of anti-VEGF therapy. Results: Long-term management of nAMD with intravitreal injections of agents targeting vascular endothelial growth factor (VEGF) is crucial for slowing progression of the disease and is generally well-tolerated. However, vision often declines as the disease progresses over time, even with treatment. This article presents strategies for aligning therapeutic goals with their expected visual outcome when an eye has reached end-stage disease. It addresses considerations for how and when to stop treatment when vision becomes limited, taking into consideration the visual status of the fellow eye and incorporating input from low vision specialists who can better assess best-corrected visual acuity (BCVA) and optimize the visual function of patients. We also acknowledge the potential benefits of switching either the dose or the agent that targets VEGF to alter the long-term visual outcome of treatment. Finally, we discuss the importance of taking into consideration related manifestations of the disease, such as macular scarring, geographic atrophy, or other retinal or optic nerve diseases which may limit vision and thus the utility of continued nAMD treatment. Conclusions: Building a strong patient–physician relationship is essential for navigating the shared decision-making process of when to stop treatment for nAMD. Full article