Search Results (3,264)

Artificial intelligence-based image processing is critical for sensor fusion and image transformation in mobility systems. Advanced driver assistance functions such as forward monitoring and digital side mirrors are essential for driving safety. Degradation due to raindrops, fog, and high-dynamic range (HDR) imbalance caused by lighting changes impairs visibility and reduces object recognition and distance estimation accuracy. This paper proposes a diffusion framework to enhance visibility under multi-degradation conditions. The denoising diffusion probabilistic model (DDPM) offers more stable training and high-resolution restoration than the generative adversarial networks. The DDPM relies on large-scale paired datasets, which are difficult to obtain in raindrop scenarios. This framework applies the Palette diffusion model, comprising data augmentation and raindrop-removal modules. The data augmentation module generates raindrop image masks and learns inpainting-based raindrop synthesis. Synthetic masks simulate raindrop patterns and HDR imbalance scenarios. The raindrop-removal module reconfigures the Palette architecture for image-to-image translation, incorporating the augmented synthetic dataset for raindrop removal learning. Loss functions and normalization strategies improve restoration stability and removal performance. During inference, the framework operates with a single conditional input, and an efficient sampling strategy is introduced to significantly accelerate the process. In post-processing, tone adjustment and chroma compensation enhance visual consistency. The proposed method preserves fine structural details and outperforms existing approaches in visual quality, improving the robustness of vision systems under adverse conditions. Full article

Background/Objectives: Previous research has demonstrated that the perception of self-motion, as signaled by cervical proprioception, is significantly altered during neck muscle fatigue, while no similar effects are observed when self-motion is signaled by the vestibular system. Given that in typical natural movements, both proprioceptive and vestibular signals are activated simultaneously, this study sought to investigate whether the misperception of motion persists during neck muscle fatigue when both proprioceptive and vestibular stimulation are present. Methods: The study evaluated the gain of the perceptual responses to symmetric yaw sinusoidal head rotations on a stationary trunk during visual target localization tasks across different rotational frequencies. In addition, the final localization error of the visual target was assessed following asymmetric sinusoidal head rotations with differing half-cycle velocities. Results: The findings indicated that even with combined proprioceptive and vestibular stimulation, self-motion perceptual responses under neck muscle fatigue showed a pronounced reduction in the gain at low-frequency stimuli and a notable increase in localization error following asymmetric rotations. Notably, spatial localization error was observed to persist after asymmetric stimulation conditioning in the light. Additionally, even moderate levels of muscle fatigue were found to result in increased self-motion misperception. Conclusions: This study suggests that neck muscle fatigue can disrupt spatial orientation, even when the vestibular system is activated, so that slow movements are inaccurately perceived. This highlights the potential risks associated with neck muscle fatigue in daily activities that demand precise spatial perception. Full article

This study presents a comprehensive assessment of lighting conditions in the Intensive Care Units (ICUs) of two major hospitals in the Democratic Republic of Congo (DRC): Hospital du Cinquantenaire in Kinshasa and Jason Sendwe Hospital in Lubumbashi. A mixed-methods approach was employed, integrating continuous illuminance monitoring with structured staff surveys to evaluate visual comfort in accordance with the EN 12464-1 standard for indoor workplaces. Objective measurements revealed that more than 52.2% of the evaluated ICU workspaces failed to meet the recommended minimum illuminance level of 300 lux. Subjective responses from healthcare professionals indicated that poor lighting significantly reduced job satisfaction by 40%, lowered self-rated task performance by 30%, decreased visual comfort scores from 4.1 to 2.6 (on a 1–5 scale), and increased the prevalence of well-being symptoms (eye fatigue, headaches) by 25–35%. Frequent complaints included eye strain, glare, and discomfort with posture, with these issues often exacerbated during the rainy season due to reduced natural daylight. The study highlights critical deficiencies in current lighting infrastructure and emphasizes the need for urgent improvements in clinical environments. Moreover, inconsistent energy supply to these healthcare settings also impacts the assurance of visual comfort. To address these shortcomings, the study recommends transitioning to energy-efficient LED lighting, enhancing access to natural light, incorporating circadian rhythm-based lighting systems, enabling individual lighting control at workstations, and ensuring a consistent power supply via the integration of solar inverters to the grid supply. These interventions are essential not only for improving healthcare staff performance and safety but also for supporting better patient outcomes. The findings offer actionable insights for hospital administrators and policymakers in the DRC and similar low-resource settings seeking to enhance environmental quality in critical care facilities. Full article

This article proposes an improved Visible Light Communication (VLC) solution that, besides the indoor lighting and data transfer, offers an energy-efficient alternative for modern workspaces. Unlike Light-Fidelity (LiFi), designed for high-speed data communication, VLC primarily targets applications where fast data rates are not essential. The developed prototype ensures reliable communication under variable lighting conditions, addressing low-speed requirements such as test bench monitoring, occupancy detection, remote commands, logging or access control. Although the tested data rate was limited to 100 kb/s with a Bit Error Rate (BER) below 10⁻⁷, the key innovation is the light dimming dynamic adaptation. Therefore, the system self-adjusts the LED duty cycle between 10% and 90%, based on natural or artificial ambient light, to maintain a minimum illuminance of 300 lx at the workspace level. Additionally, this work includes a scalability analysis through simulations conducted in an office scenario with up to six users. The results show that the system can adjust the lighting level and maintain the connectivity according to users’ presence, significantly reducing energy consumption without compromising visual comfort or communication performance. With this light intensity regulation algorithm, the proposed solution demonstrates real potential for implementation in smart indoor environments focused on sustainability and connectivity. Full article

Objectives: Anterior petrosectomy is a challenging neurosurgical procedure requiring precise identification and preservation of multiple critical structures. This technical note explores the feasibility of using real-time near-infrared indocyanine green (NIR-ICG) fluorescence with white light overlay to enhance visualization of the petrous internal carotid artery (ICA) during transpetrosal drilling. We aimed to assess its utility for planning and performing modified Dolenc–Kawase drilling. Methods: We integrated NIR-ICG and white light overlay using a robotic microscope with simultaneous visualization capabilities. This technique was applied to improve neurovascular preservation and skull base landmark identification. Intraoperative video frames and images were captured during an anterior transpetrosal approach for a petroclival meningioma, with technical details, surgical time, and feedback documented. Results: Real-time NIR-ICG with white light overlay successfully identified the posterior genu, horizontal petrosal segment, anterior genu, and superior petrosal sinus. It facilitated precise localization of cochlear landmarks, enabling tailored drilling of the Dolenc–Kawase rhomboid according to patient anatomy and accommodating potential anatomical variants. Conclusions: This approach could enhance intraoperative safety and improve exposure, possibly reducing neurovascular risks without extending operative time. It may serve as a valuable adjunct for complex skull base surgeries. Full article

We present a (Light)weight (S)atellite–(A)erial feature (M)atching framework (LiteSAM) for robust UAV absolute visual localization (AVL) in GPS-denied environments. Existing satellite–aerial matching methods struggle with large appearance variations, texture-scarce regions, and limited efficiency for real-time UAV applications. LiteSAM integrates three key components to address these issues. First, efficient multi-scale feature extraction optimizes representation, reducing inference latency for edge devices. Second, a Token Aggregation–Interaction Transformer (TAIFormer) with a convolutional token mixer (CTM) models inter- and intra-image correlations, enabling robust global–local feature fusion. Third, a MinGRU-based dynamic subpixel refinement module adaptively learns spatial offsets, enhancing subpixel-level matching accuracy and cross-scenario generalization. The experiments show that LiteSAM achieves competitive performance across multiple datasets. On UAV-VisLoc, LiteSAM attains an RMSE@30 of 17.86 m, outperforming state-of-the-art semi-dense methods such as EfficientLoFTR. Its optimized variant, LiteSAM (opt., without dual softmax), delivers inference times of 61.98 ms on standard GPUs and 497.49 ms on NVIDIA Jetson AGX Orin, which are 22.9% and 19.8% faster than EfficientLoFTR (opt.), respectively. With 6.31M parameters, which is 2.4× fewer than EfficientLoFTR’s 15.05M, LiteSAM proves to be suitable for edge deployment. Extensive evaluations on natural image matching and downstream vision tasks confirm its superior accuracy and efficiency for general feature matching. Full article

The timely detection of falls among the elderly remains challenging. Single modality sensing approaches using inertial measurement units (IMUs) or vision-based monitoring systems frequently exhibit high false positives and compromised accuracy under suboptimal operating conditions. We propose a novel bimodal deep learning-based bimodal sensing framework to address the problem, by leveraging a memory-based autoencoder neural network for inertial abnormality detection and an attention-based neural network for visual pose assessment, with late fusion at the decision level. Our experimental evaluation with a custom dataset of simulated falls and routine activities, captured with waist-mounted IMUs and RGB cameras under dim lighting, shows significant performance improvement by the described bimodal late-fusion system, with an F1-score of 97.3% and, most notably, a false-positive rate of 3.6% significantly lower than the 11.3% and 8.9% with IMU-only and vision-only baselines, respectively. These results confirm the robustness of the described fall detection approach and validate its applicability to real-time fall detection under different light settings, including nighttime conditions. Full article

Hakka traditional vernacular dwellings embody regionally specific climatic adaptation strategies. This study takes the Meizhou Guangludi enclosed house as a case study to evaluate its climate adaptability with longevity and passive survivability factors of the Hakka three-hall enclosed house under subtropical climatic conditions. A mixed research method is employed, integrating visualized parametric modeling analysis and on-site measurement comparisons to quantify wind, temperature, solar radiation/illuminance, and humidity, along with human comfort zone limits and building environment. The results reveal that nature erosion in the Guangludi enclosed house is the most pronounced during winter and spring, particularly on exterior walls below 2.8 m. Key issues include bulging, spalling, molding, and fractured purlins caused by wind-driven rain, exacerbated by low wind speeds and limited solar exposure, especially at test spots like the E8–E10 and N1–N16 southeast and southern walls below 1.5 m. Fungal growth and plant intrusion are severe where surrounding trees and fengshui forests restrict wind flow and lighting. In terms of passive survivability, the Guangludi enclosed house has strong thermal insulation and buffering, aided by the Huatai mound; however, humidity and day illuminance deficiencies persist in the interstitial spaces between lateral rooms and the central hall. To address these issues, this study proposes strategies such as adding ventilation shafts and flexible partitions, optimizing patio dimensions and window-to-wall ratios, retaining the spatial layout and Fengshui pond to enhance wind airflow, and reinforcing the identified easily eroded spots with waterproofing, antimicrobial coatings, and extended eaves. Through parametric simulation and empirical validation, this study presents a climate-responsive retrofit framework that supports the sustainability and conservation of the subtropical Hakka enclosed house. Full article

Color grading of boiled shrimp is a critical factor influencing market price, yet the process is usually conducted visually by buyers such as middlemen and processing plants. This subjective practice raises concerns about accuracy, impartiality, and fairness, often resulting in disputes with farmers. To address this issue, this study proposes a standardized and automated grading approach based on image processing and artificial intelligence. The method requires only a photograph of boiled shrimp placed alongside a color grading ruler. The grading process involves two stages: segmentation of shrimp and ruler regions in the image, followed by color comparison. For segmentation, deep learning models based on Mask R-CNN with a Feature Pyramid Network backbone were employed. Four model configurations were tested, using ResNet and ResNeXt backbones with and without a Boundary Loss function. Results show that the ResNet + Boundary Loss model achieved the highest segmentation performance, with IoU scores of 91.2% for shrimp and 87.8% for the color ruler. In the grading step, color similarity was evaluated in the CIELAB color space by computing Euclidean distances in the L (lightness) and a (red–green) channels, which align closely with human perception of shrimp coloration. The system achieved grading accuracy comparable to human experts, with a mean absolute error of 1.2, demonstrating its potential to provide consistent, objective, and transparent shrimp quality assessment. Full article

This study presents the design and evaluation of AquaSignalNet, a deep learning-based system for recognizing underwater visual commands to enable the autonomous navigation of a Remotely Operated Vehicle (ROV). The system is built on a ResNet50 V2 architecture and trained with a custom dataset, UVSRD, comprising 33,800 labeled images across 12 gesture classes, including directional commands, speed values, and vertical motion instructions. The model was deployed on a Raspberry Pi 4 integrated with a TIVA C microcontroller for real-time motor control, a PID-based depth control loop, and an MPU9250 sensor for orientation tracking. Experiments were conducted in a controlled pool environment using printed signal cards to define two autonomous trajectories. In the first trajectory, the system achieved 90% success, correctly interpreting a mixed sequence of turns, ascents, and speed changes. In the second, more complex trajectory, involving a rectangular inspection loop and multi-layer navigation, the system achieved 85% success, with failures mainly due to misclassification resulting from lighting variability near the water surface. Unlike conventional approaches that rely on QR codes or artificial markers, AquaSignalNet employs markerless visual cues, offering a flexible alternative for underwater inspection, exploration, and logistical operations. The results demonstrate the system’s viability for real-time gesture-based control. Full article

The accurate assessment of Chinese bayberry (Myrica rubra) maturity is critical for intelligent harvesting. This study proposes a novel cascaded framework combining instance segmentation and multi-feature regression for accurate maturity detection. First, a lightweight SOLOv2-Light network is employed to segment each fruit individually, which significantly reduces computational costs with only a marginal drop in accuracy. Then, a multi-feature extraction network is developed to fuse deep semantic, color (LAB space), and multi-scale texture features, enhanced by a channel attention mechanism for adaptive weighting. The maturity ground truth is defined using the a*/b* ratio measured by a colorimeter, which correlates strongly with anthocyanin accumulation and visual ripeness. Experimental results demonstrated that the proposed method achieves a mask mAP of 0.788 on the instance segmentation task, outperforming Mask R-CNN and YOLACT. For maturity prediction, a mean absolute error of 3.946% is attained, which is a significant improvement over the baseline. When the data are discretized into three maturity categories, the overall accuracy reaches 95.51%, surpassing YOLOX-s and Faster R-CNN by a considerable margin while reducing processing time by approximately 46%. The modular design facilitates easy adaptation to new varieties. This research provides a robust and efficient solution for in-field bayberry maturity detection, offering substantial value for the development of automated harvesting systems. Full article

During the image compositing process, there may be inconsistencies in tone and illumination between the foreground and background, leading to poor visual quality and low realism in the composite images. To address these issues, image harmonization techniques can be employed. This paper proposes an image harmonization method based on multi-scale and global feature guidance (MSGF). In general, images captured in different scenes may exhibit inconsistencies in lighting after composition. The goal of image harmonization is to adjust the foreground illumination to match that of the background. Traditional methods often attempt to blend pixels directly, which can result in unrealistic outcomes. The proposed approach combines multi-scale feature extraction with global feature guidance, forming the MSGF framework. The experiment was conducted on the iHarmony4 dataset. Comparative experiments showed that MSGF achieved the best performance on three subset indicators, including HCOCO. Ablation studies demonstrated the effectiveness of the proposed module. Efficiency evaluation results indicated that it took 0.01s and had 20.9 million parameters, outperforming comparative methods and effectively achieving high-quality image harmonization. Full article

In the rapid digital era, artificial intelligence (AI) tools have progressively arisen to shape the education environment. In this context, large language models (LLMs) (i.e., ChatGPT vs. 4.0 and Gemini vs. 2.5) have emerged as powerful applications for academic inclusion. This paper investigated how using and trusting LLMs can impact the academic success and quality of life (QoL) of visually impaired university students. Quantitative research was conducted, obtaining data from 385 visually impaired university students through a structured survey design. Partial Least Squares Structural Equation Modelling (PLS-SEM) was implemented to test the study hypotheses. The findings revealed that trust in LLMs can significantly predict LLM usage, which in turn can improve QoL. While LLM usage failed to directly support the academic success of disabled students, but its impact was mediated through QoL, suggesting that enhancements in well-being can contribute to higher academic success. The results highlighted the importance of promoting trust in AI applications, along with developing an accessible, inclusive, and student-centred digital environment. The study offers practical contributions for educators and policymakers, shedding light on the importance of LLM applications for both the QoL and academic success of visually impaired university students. Full article

Can brightness illusions modulate ocular accommodation? Previous studies have shown that brightness illusions can influence pupil size as if caused by actual luminance increases. However, their effects on other ocular responses—such as accommodative or focusing dynamics—remain largely unexplored. This study investigates the influence of brightness illusions, under two ambient lighting conditions, on accommodative and pupillary dynamics (physiological responses), and on perceived brightness and visual comfort (subjective responses). Thirty-two young adults with healthy vision viewed four stimulus types (blue bright and non-bright, yellow bright and non-bright) under low- and high-contrast ambient lighting while ocular responses were recorded using a WAM-5500 open-field autorefractor. Brightness and comfort were rated after each session. The results showed that high ambient contrast (mesopic) and brightness illusions increased accommodative variability, while yellow stimuli elicited a greater lag under photopic condition. Pupil size decreased only under mesopic lighting. Perceived brightness was enhanced by brightness illusions and blue color, whereas visual comfort decreased for bright illusions, especially under low light. These findings suggest that ambient lighting and visual stimulus properties modulate both physiological and subjective responses, highlighting the need for dynamic accommodative assessment and visually ergonomic display design to reduce visual fatigue during digital device use. Full article

Vision is a critical component of daily life, and its loss significantly hinders an individual’s ability to navigate, particularly when using public transportation systems. To address this challenge, this paper introduces a novel approach for accurately identifying bus route numbers and destinations, designed to assist visually impaired individuals in navigating urban transit networks. Our system integrates object detection, image enhancement, and Optical Character Recognition (OCR) technologies to achieve reliable and precise recognition of bus information. We employ a custom-trained You Only Look Once version 8 (YOLOv8) model to isolate the front portion of buses as the region of interest (ROI), effectively eliminating irrelevant text and advertisements that often lead to errors. To further enhance accuracy, we utilize the Enhanced Super-Resolution Generative Adversarial Network (ESRGAN) to improve image resolution, significantly boosting the confidence of the OCR process. Additionally, a post-processing step involving a pre-defined list of bus routes and the Levenshtein algorithm corrects potential errors in text recognition, ensuring reliable identification of bus numbers and destinations. Tested on a dataset of 120 images featuring diverse bus routes and challenging conditions such as poor lighting, reflections, and motion blur, our system achieved an accuracy rate of 95%. This performance surpasses existing methods and demonstrates the system’s potential for real-world application. By providing a robust and adaptable solution, our work aims to enhance public transit accessibility, empowering visually impaired individuals to navigate cities with greater independence and confidence. Full article