Varying Lighting Conditions Research Articles

Mathematical Modeling for Robot 3D Laser Scanning in Complete Darkness Environments to Advance Pipeline Inspection

This paper introduces an autonomous robot designed for in-pipe structural health monitoring of oil/gas pipelines. This system employs a 3D Optical Laser Scanning Technical Vision System (TVS) to continuously scan the internal surface of the pipeline. This paper elaborates on the mathematical methodology of 3D laser surface scanning based on dynamic triangulation. This paper presents the mathematical framework governing the combined kinematics of the Mobile Robot (MR) and TVS. It discusses the custom design of the MR, adjusting it to use of robustized mathematics, and incorporating a laser scanner produced using a 3D printer. Both experimental and theoretical approaches are utilized to illustrate the formation of point clouds during surface scanning. This paper details the application of the simple and robust mathematical algorithm RANSAC for the preliminary processing of the measured point clouds. Furthermore, it contributes two distinct and simplified criteria for detecting defects in pipelines, specifically tailored for computer processing. In conclusion, this paper assesses the effectiveness of the proposed mathematical and physical method through experimental tests conducted under varying light conditions.

Measurement of visual function in infantile nystagmus: a systematic review

Background/aimsRecent work has called into question the ability of visual acuity (VA) to accurately represent changes in visual function in infantile nystagmus (IN). This systematic review investigated factors affecting visual...

Dual-stage color calibration of UAV imagery using multivariate regression and deep learning

Accurate color representation in UAV-based imagery is critical in high-throughput plant phenotyping and precision agriculture for monitoring crop health and identifying nutrient deficiencies or water stress. However, achieving color accuracy is challenging due to varying lighting conditions and flight altitudes in UAV operations. To overcome these challenges, we propose an automated dual-phase approach for consistent color calibration in UAV imagery, integrating traditional color calibration methods with cutting-edge deep learning algorithms. In the first phase, we derived a color calibration matrix (M) for four color boards placed within the imaging area. This involved mapping the observed color values in the UAV images to their true color values, as measured by a colorimeter, using multivariate linear regression (MLR). The individual matrices from each board were then merged into a single, weighted average matrix, which was then applied to the UAV images for color correction, resulting in standardized color outputs. In the second phase, we developed a deep learning model, trained on the MLR-corrected images as a ground truth. This model assigns varying weights for different pixels across the image scene, translating the image from a distorted to a color-corrected image using image-to-image regression without manual intervention. We validated this approach using 184 UAV images of various sizes, ranging from 3969 × 2589 to 15257 × 15098 pixels, captured under a wide array of lighting conditions and altitudes. The results show that our deep learning model significantly outperforms existing methods, with a marginal color error of 1.52 %. This two-phase color correction method greatly improves operational efficiency by removing the necessity for physical color boards in each image and allowing for automated processing after training.

Differential Regulation of Opsin Gene Expression in Response to Internal and External Stimuli.

Determining how internal and external stimuli interact to determine developmental trajectories of traits is a challenge that requires the integration of different subfields of biology. Internal stimuli, such as hormones, control developmental patterns of phenotypic changes, which might be modified by external environmental cues (e.g. plasticity). Thyroid hormone (TH) modulates the timing of opsin gene expression in developing Midas cichlid fish (Amphilophus citrinellus). Moreover, fish reared in red light accelerate this developmental timing compared to fish reared in white light. Hence, we hypothesized that plasticity caused by variation in light conditions has coopted the TH signaling pathway to induce changes in opsin gene expression. We treated Midas cichlids with TH and crossed this treatment with two light conditions, white and red. We observed that not only opsin expression responded similarly to TH and red light but also that, at high TH levels, there is limited capacity for light-induced plasticity. Transcriptomic analysis of the eye showed that genes in the TH pathway were affected by TH, but not by light treatments. Coexpression network analyses further suggested that response to light was independent of the response to TH manipulations. Taken together, our results suggest independent mechanisms mediating development and plasticity during development of opsin gene expression, and that responses to environmental stimuli may vary depending on internal stimuli. This conditional developmental response to external factors depending on internal ones (e.g. hormones) might play a fundamental role in the patterns of phenotypic divergence observed in Midas cichlids and potentially other organisms.

Solar Spectrum Light‐Driven Silicone‐Based Fluidic Actuators

AbstractSoft materials that convert light into mechanical energy can create new untethered strategies for actuating soft robotics. Yet, the available light‐driven materials are often incompatible with standard fabrication in soft robotics and restricted to shapes (e.g., sheets) that have limited capability for 3D deformation; often laser or focused light is required for actuation. Here, to address these challenges, a straightforward method for synthesizing sunlight‐responsive fluidic actuators from off‐the‐shelf silicone precursors capable of expanding in 3D is developed. A liquid phase and activated carbon as photothermal elements are constrained in the elastomer. Solar spectral light triggers a liquid–gas phase transition creating sufficient pressure to overcome the internal elastic stress and actuate the material. The fluidic actuation is characterized under varying light conditions reaching expansion cycle times between ≈20–500 s, strains of 28%, and actuation stress of ≈1.3 MPa in different experiments. The materials were then used to exemplarily drive a mechanical switch, a liquid dispensing soft pump, a valve, and a bending actuator. As the described materials are easy to produce in a 5 min synthesis by standard molding techniques, it is believed that they are a promising opportunity for embodied energy converters in environmentally powered soft robots that respond to sunlight.

An Advanced Approach to Object Detection and Tracking in Robotics and Autonomous Vehicles Using YOLOv8 and LiDAR Data Fusion

Accurately and reliably perceiving the environment is a major challenge in autonomous driving and robotics research. Traditional vision-based methods often suffer from varying lighting conditions, occlusions, and complex environments. This paper addresses these challenges by combining a deep learning-based object detection algorithm, YOLOv8, with LiDAR data fusion technology. The principle of this combination is to merge the advantages of these technologies: YOLOv8 excels in real-time object detection and classification through RGB images, while LiDAR provides accurate distance measurement and 3D spatial information, regardless of lighting conditions. The integration aims to apply the high accuracy and robustness of YOLOv8 in identifying and classifying objects, as well as the depth data provided by LiDAR. This combination enhances the overall environmental perception, which is critical for the reliability and safety of autonomous systems. However, this fusion brings some research challenges, including data calibration between different sensors, filtering ground points from LiDAR point clouds, and managing the computational complexity of processing large datasets. This paper presents a comprehensive approach to address these challenges. Firstly, a simple algorithm is introduced to filter out ground points from LiDAR point clouds, which are essential for accurate object detection, by setting different threshold heights based on the terrain. Secondly, YOLOv8, trained on a customized dataset, is utilized for object detection in images, generating 2D bounding boxes around detected objects. Thirdly, a calibration algorithm is developed to transform 3D LiDAR coordinates to image pixel coordinates, which are vital for correlating LiDAR data with image-based object detection results. Fourthly, a method for clustering different objects based on the fused data is proposed, followed by an object tracking algorithm to compute the 3D poses of objects and their relative distances from a robot. The Agilex Scout Mini robot, equipped with Velodyne 16-channel LiDAR and an Intel D435 camera, is employed for data collection and experimentation. Finally, the experimental results validate the effectiveness of the proposed algorithms and methods.

Energy-Efficient Design of Seabed Substrate Detection Model Leveraging CNN-SVM Architecture and Sonar Data

This study introduces an innovative seabed substrate detection model that harnesses the complementary strengths of Convolutional Neural Networks (CNNs) and Support Vector Machines (SVMs) to analyze sonar data with a focus on energy efficiency. The model addresses the challenges of underwater sensing and imaging, including variable lighting conditions, backscattering effects, and acoustic sensor limitations, while minimizing energy consumption. By leveraging advanced machine learning techniques, the proposed model aims to enhance seabed classification accuracy, a crucial aspect for marine operations, ecological studies, and energy-intensive underwater applications.The introduced ShuffleNet-DSE architecture demonstrates significant improvements in both accuracy and stability for seabed sediment image classification, while maintaining energy-efficient performance. This robust tool offers a valuable asset for underwater exploration, research, and monitoring efforts, especially in environments where energy resources are limited.

Comparative effects of heat stress on photosynthesis and oxidative stress in Halophila ovalis and Thalassia hemprichii under different light conditions

This study investigated the physiological responses of two tropical seagrass species, Halophila ovalis and Thalassia hemprichii, to heat stress under varying light conditions in a controlled 5-day experiment. The experimental design included four treatments: control, saturating light, heat stress under sub-saturating light, and heat stress under saturating light (combined stress). We assessed various parameters, including chlorophyll fluorescence, levels of reactive oxygen species (ROS), antioxidant enzyme activities, and growth rates. In H. ovalis, heat stress resulted in a significant reduction in the maximum quantum yield of photosystem II (Fv/Fm) regardless of the light condition. However, the effects of heat stress on the effective quantum yield of photosystem II (ɸPSII) were more pronounced under saturating light conditions. In T. hemprichii, saturating irradiance exacerbated the heat stress effects on Fv/Fm and ɸPSII, although the overall photoinhibition was less severe than in H. ovalis. Heat stress led to ROS accumulation in H. ovalis and reduced the activity of superoxide dismutase (SOD) and ascorbate peroxidase in the sub-saturating light condition. Conversely, T. hemprichii exhibited elevated SOD activity under saturating light. Heat stress suppressed the growth of both seagrass species, regardless of the light environment. The Biomarker Response Index indicated that H. ovalis displayed severe effects in the heat stress treatment under both light conditions, while T. hemprichii exhibited moderate effects in sub-saturating light and major effects in saturating light conditions. However, the Effect Addition Index revealed an antagonistic interaction between heat stress and high light in both seagrass species. This study underscores the intricate responses of seagrasses, emphasizing the importance of considering both local and global stressors when assessing their vulnerability.

Automated blood volume estimation in surgical drains for clinical decision support.

Monitoring Jackson Pratt and Hemovac drains plays a crucial role in assessing a patient's recovery and identifying potential postoperative complications. Accurate and regular monitoring of the blood volume in the drain is essential for making decisions about patient care. However, transferring blood to a measuring cup and recording it is a challenging task for both patients and doctors, exposing them to bloodborne pathogens such as the human immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus (HCV). To automate the recording process with a non-contact approach, we propose an innovative approach that utilizes deep learning techniques to detect a drain in a photograph, compute the blood level in the drain, estimate the blood volume, and display the results on both web and mobile interfaces. Our system employs semantic segmentation on images taken with mobile phones to effectively isolate the blood-filled portion of the drain from the rest of the image and compute the blood volume. These results are then sent to mobile and web applications for convenient access. To validate the accuracy and effectiveness of our system, we collected the Drain Dataset, which consists of 1,004 images taken under various background and lighting conditions. With an average error rate of less than 5% in milliliters, our proposed approach achieves highly accurate blood level detection and estimation, as demonstrated by our trials on this dataset. The system also exhibits robustness to variations in lighting conditions and drain shapes, ensuring its applicability in different clinical scenarios. The proposed automated blood volume estimation system can significantly reduce the time and effort required for manual measurements, enabling healthcare professionals to focus on other critical tasks. The dataset and annotations are available at: https://www.kaggle.com/datasets/ayenahin/liquid-volume-detection-from-drain-images and the code for the web application is available at https://github.com/itsjustaplant/AwesomeProject.git.

Optimizing the viola-jones algorithm for robust face recognition in variable lighting and orientation conditions

Optimizing the viola-jones algorithm for robust face recognition in variable lighting and orientation conditions

Fabrication of cobalt-doped ZnS thin films by successive ionic layer adsorption and reaction for UV–visible photodetectors on flexible substrate

Photodetectors are widely employed in research and innovation, necessitating swift response times, precision, and stability. However, their application in commercial and industrial settings is hindered by limited flexibility and a constrained capacity to respond across a broad spectrum. In this study, we adopted a facile, cost-effective, and room temperature Successive Ionic Layer Adsorption and Reaction (SILAR) technique to craft thin films of Cobalt-doped ZnS on a flexible paper substrate. The Cobalt doping, with varying wt.% (0, 0.5, 1, and 1.5) exhibiting excellent responsivity to both UV and visible spectra. Zinc Sulfide (ZnS), renowned for its wide bandgap, serves as a versatile semiconductor, making it an ideal candidate for photodetectors. The introduction of Cobalt broadened the responsivity of ZnS, covering a wide spectrum, including UV and visible regions. Structural confirmation of the material was achieved through XRD and Raman analyses. Scanning Electron Microscopy revealed the aggregated spherical nanostructure of ZnS on the cellulose substrate. Verification of Cobalt presence in the ZnS thin film and assessment of elemental oxidation states were conducted using XPS. Temporal response and I–V characteristics of the device were evaluated under varying light conditions, specifically at wavelengths of 325 nm and 700 nm at different intensities. Responsivity calculations yielded values of 22.5 mA/W for UV illumination (light intensity 0.6 mW/cm2) and 37.6 mA/W (light intensity 0.4 mW/cm2) for visible light, with corresponding response times recorded as 1.45 s and 2.13 s, respectively. In addition to performance assessments, bending cycle studies were conducted, demonstrating the excellent flexibility of the device up to 800 cycles. The Cobalt-doped ZnS device, as fabricated, emerges as a promising material for energy harvesting.

Enhancement-Mode Carbon Nanotube Optoelectronic Synaptic Transistors with Large and Controllable Threshold Voltage Modulation Window for Broadband Flexible Vision Systems.

The development of large-scale integration of optoelectronic neuromorphic devices with ultralow power consumption and broadband responses is essential for high-performance bionics vision systems. In this work, we developed a strategy to construct large-scale (40 × 30) enhancement-mode carbon nanotube optoelectronic synaptic transistors with ultralow power consumption (33.9 aJ per pulse) and broadband responses (from 365 to 620 nm) using low-work function yttrium (Y)-gate electrodes and the mixture of eco-friendly photosensitive Ag2S quantum dots (QDs) and ionic liquids (ILs)-cross-linking-poly(4-vinylphenol) (PVP) (ILs-c-PVP) as the dielectric layers. Solution-processable carbon nanotube thin-film transistors (TFTs) showed enhancement-mode characteristics with the wide and controllable threshold voltage window (-1 V∼0 V) owing to use of the low-work-function Y-gate electrodes. It is noted that carbon nanotube optoelectronic synaptic transistors exhibited high on/off ratios (>106), small hysteresis and low operating voltage (≤2 V), and enhancement mode even under the illumination of ultraviolet (UV, 365 nm), blue (450 nm), and green (550 nm) to red (620 nm) pulse lights when introducing eco-friendly Ag2S QDs in dielectric layers, demonstrating that they have the strong fault-tolerant ability for the threshold voltage drifts caused by various manufacturing scenarios. Furthermore, some important bionic functions including a high paired pulse facilitation index (PPF index, up to 290%), learning and memory function with the long duration (200 s), and rapid recovery (2 s). Pavlov's dog experiment (retention time up to 20 min) and visual memory forgetting experiments (the duration of high current for 180 s) are also demonstrated. Significantly, the optoelectronic synaptic transistors can be used to simulate the adaptive process of vision in varying light conditions, and we demonstrated the dynamic transition of light adaptation to dark adaptation based on light-induced conditional behavior. This work undoubtedly provides valuable insights for the future development of artificial vision systems.

Residual spatial fusion network for RGB-thermal semantic segmentation

Semantic segmentation plays an important role in widespread applications such as autonomous driving and robotic sensing. This work focuses on RGB-Thermal (RGB-T) semantic segmentation, since RGB images are easily affected by lighting conditions, e.g., darkness, and thermal images are robust to the night scenario as a compensating modality. However, existing works either simply fuse RGB-T images or adopt the encoder with the same backbone for the two modalities, neglecting the semantic difference under varying lighting conditions. Therefore, we present a Residual Spatial Fusion Network (RSFNet) for RGB-T semantic segmentation. Specifically, RSFNet employs an asymmetric encoder to learn the compensating features of RGB and thermal images. To fuse the dual-modality features, it generates pseudo-labels by adopting the saliency detection skill to supervise feature learning, and develops the Residual Spatial Fusion (RSF) module with structural re-parameterization to learn more promising features by spatially fusing the cross-modality features. RSF adopts a hierarchical feature fusion to aggregate multi-level features, and applies the spatial weights with the residual connection to adaptively control the multi-spectral feature fusion by the confidence gate. Extensive experiments were carried out on two benchmarks, i.e., MFNet database and PST900 database. The results have shown the state-of-the-art segmentation performance of our method, which strikes a good balance between accuracy and speed.

Advances in Real-Time 3D Reconstruction for Medical Endoscopy.

This contribution is intended to provide researchers with a comprehensive overview of the current state-of-the-art concerning real-time 3D reconstruction methods suitable for medical endoscopy. Over the past decade, there have been various technological advancements in computational power and an increased research effort in many computer vision fields such as autonomous driving, robotics, and unmanned aerial vehicles. Some of these advancements can also be adapted to the field of medical endoscopy while coping with challenges such as featureless surfaces, varying lighting conditions, and deformable structures. To provide a comprehensive overview, a logical division of monocular, binocular, trinocular, and multiocular methods is performed and also active and passive methods are distinguished. Within these categories, we consider both flexible and non-flexible endoscopes to cover the state-of-the-art as fully as possible. The relevant error metrics to compare the publications presented here are discussed, and the choice of when to choose a GPU rather than an FPGA for camera-based 3D reconstruction is debated. We elaborate on the good practice of using datasets and provide a direct comparison of the presented work. It is important to note that in addition to medical publications, publications evaluated on the KITTI and Middlebury datasets are also considered to include related methods that may be suited for medical 3D reconstruction.

A Face-Mask Detection Approach based on YOLO V3 applied for a New Collected Dataset

In the wake of the COVID-19 pandemic, the necessity of face-mask detection systems has become paramount to enforce safety measures and regulations. This paper presents a novel approach for face-mask detection utilizing the YOLOv3 architecture applied to a newly collected dataset. The proposed system aims to accurately detect the presence or absence of face masks in real-time scenarios. The dataset used in this research is meticulously curated to encompass diverse environmental conditions, facial expressions, and variations in mask types and orientations. Each image in the dataset is annotated with bounding boxes indicating the regions of faces and masks, facilitating supervised learning for the YOLOv3 model. Our methodology involves fine-tuning the pre-trained YOLOv3 model on the collected dataset to specialize in face-mask detection. The model is trained using a combination of techniques such as data augmentation, transfer learning, and hyperparameter optimization to enhance its performance. Furthermore, to address challenges such as occlusions, varying lighting conditions, and diverse facial orientations, we incorporate techniques like multi-scale training and post-processing algorithms. These techniques aid in improving the robustness and generalization capability of the model, making it suitable for deployment in real-world scenarios. Keywords: Face Mask Detection, YOLO V3, Computer Vision, Deep Learning, Object Detection, Dataset Collection, COVID-19, Image Processing, Real-time Monitoring.

Hand Gesture Recognition System Using Deep Learning

Hand Gesture Recognition Systems have undergone significant advancements, ushering in a new era of human-computer interaction. This paper offers a thorough examination of the current state of the art in hand gesture recognition, addressing both the notable progress achieved and the persistent challenges. By leveraging state-of-the-art technologies such as computer vision and deep learning, the paper explores the methodologies employed in data collection, preprocessing, and the implementation of various algorithms. The research delves into the complexities of popular hand gesture datasets, emphasizing their role in training and testing models. A critical analysis of different algorithms and models, including Hidden Markov Models, Support Vector Machines, and Neural Networks, is presented. The paper scrutinizes their strengths and limitations, providing insights into the delicate balance between accuracy and real-time processing. Furthermore, it investigates the diverse applications of hand gesture recognition, spanning from enriching human-computer interaction to its pivotal role in virtual reality, gaming, and robotics. Despite these advancements, challenges persist, such as occlusion, varying lighting conditions, and the imperative for real-time processing. The hardware utilized in hand gesture recognition systems, including depth sensors, RGB-D cameras, and wearable devices, is examined. Evaluation metrics, such as accuracy, precision, recall, and the F1 score, are employed to evaluate system performance.

Real Time Recognition of Underwater Images using Deep Learning

This study focuses on the development of an effective image recognition system tailored for underwater environments. The objective is to create a robust and efficient artificial neural network capable of accurately categorizing underwater images amidst challenges like noise, distortion, and variable lighting conditions. The system's primary function is to classify images into predefined categories, such as distinct species of marine organisms, underwater objects, or environmental conditions.

Evaluating error sources to improve precision in the co-registration of underwater 3D models

Change detection is an essential and widely used approach for investigating ecosystem dynamics. Multi-temporal 3D models increasingly underpin photogrammetry-based analyses of change for many ecologically relevant attributes. To detect change, it is necessary to accurately align 3D models collected at different times using a process referred to as co-registration. However, achieving precise co-registration is difficult in underwater habitats due to practical challenges intrinsic to surveying them. These include a lack of accurate georeferencing information, variable light, turbidity and weather conditions, and diving restrictions dictated by the diver's pressure exposure over time. Here we present an efficient co-registration workflow for 3D models that directly addresses these challenges, derived from underwater structure-from-motion methods. To test our approach, we used 3D models from across a wide range of coral reef habitats covering all those that one may encounter in shallow reefs (15 m depth and above). We then identified and empirically estimated four key sources of error: co-registration, 3D processing, image acquisition, and reference and scaling features (RSF) placement, and quantified their relative contributions to the overall error. Our proposed co-registration workflow had a mean precision of 1.37 ± 16.55 mm. Image acquisition and RSF placement errors contributed the most to the total workflow error (37% and 53%, respectively), while the contribution of co-registration and 3D processing errors was minimal (3% and 7%, respectively). As a result of our analysis, we provide ‘good practice’ guidelines to reduce errors associated with photogrammetric workflows and to facilitate efficient and reliable detection of 3D change in complex underwater ecosystems.

A Hybrid Image Augmentation Technique for User- and Environment-Independent Hand Gesture Recognition Based on Deep Learning

This research stems from the increasing use of hand gestures in various applications, such as sign language recognition to electronic device control. The focus is the importance of accuracy and robustness in recognizing hand gestures to avoid misinterpretation and instruction errors. However, many experiments on hand gesture recognition are conducted in limited laboratory environments, which do not fully reflect the everyday use of hand gestures. Therefore, the importance of an ideal background in hand gesture recognition, involving only the signer without any distracting background, is highlighted. In the real world, the use of hand gestures involves various unique environmental conditions, including differences in background colors, varying lighting conditions, and different hand gesture positions. However, the datasets available to train hand gesture recognition models often lack sufficient variability, thereby hindering the development of accurate and adaptable systems. This research aims to develop a robust hand gesture recognition model capable of operating effectively in diverse real-world environments. By leveraging deep learning-based image augmentation techniques, the study seeks to enhance the accuracy of hand gesture recognition by simulating various environmental conditions. Through data duplication and augmentation methods, including background, geometric, and lighting adjustments, the diversity of the primary dataset is expanded to improve the effectiveness of model training. It is important to note that the utilization of the green screen technique, combined with geometric and lighting augmentation, significantly contributes to the model’s ability to recognize hand gestures accurately. The research results show a significant improvement in accuracy, especially with implementing the proposed green screen technique, underscoring its effectiveness in adapting to various environmental contexts. Additionally, the study emphasizes the importance of adjusting augmentation techniques to the dataset’s characteristics for optimal performance. These findings provide valuable insights into the practical application of hand gesture recognition technology and pave the way for further research in tailoring techniques to datasets with varying complexities and environmental variations.

Emotion Symphony: Deep Learning Journey Through Human Facial Emotions

Abstract: In today's technologically advanced world, understanding human emotions holds paramount importance in various domains ranging from human-computer interaction to mental health diagnosis. Emotions are primarily conveyed through facial expressions, making facial emotion recognition a crucial area of research. This project, "Emotion Symphony," embarks on a deep learning journey aimed at developing an efficient system for recognizing and interpreting human facial emotions. The project employs state-of-the-art deep learning techniques to extract meaningful features from facial images and map them to corresponding emotional states. Leveraging convolutional neural networks (CNNs) and recurrent neural networks (RNNs), the system is designed to analyse subtle nuances in facial expressions, capturing both spatial and temporal dependencies for accurate emotion recognition. To train the model, extensive datasets comprising diverse facial expressions are utilized, ensuring robustness and generalization of the proposed system. Additionally, data augmentation techniques are employed to enhance the model's ability to handle variations in lighting conditions, poses, and facial expressions.