Varying Lighting Conditions Research Articles

Development of AI Model for Robotic Vision Inspection of Sheet-Metal Components in Manufacturing

This paper presents an AI-based robotic vision inspection system designed to enhance quality control in the manufacturing of sheet-metal components. The system integrates a sequential AI model with advanced image processing techniques to automate real-time defect detection and classification. Utilizing a high-resolution camera, the system captures images of components on a conveyor, which are pre- processed using grayscale conversion, Gaussian blurring, and Canny edge detection to emphasize structural details. A deep learning model then classifies isolated regions of interest based on normalized, resized images. Feature matching through ORB (Oriented FAST and Rotated BRIEF) enables accurate alignment with reference templates, while automated measurements convert pixel dimensions to physical units, ensuring reliable detection of deviations. With an average accuracy of 88.3%, the system consistently identifies subtle and complex defects, such as scratches and dimensional deviations, under variable lighting and noise conditions. This AI-driven approach reduces the need for manual inspection, minimizes error, and enhances workflow efficiency, representing a major step toward robust, real-time quality assurance solutions in industrial environments.

A versatile real-time vision-led runway localisation system for enhanced autonomy

This paper proposes a solution to the challenging task of autonomously landing Unmanned Aerial Vehicles (UAVs). An onboard computer vision module integrates the vision system with the ground control communication and video server connection. The vision platform performs feature extraction using the Speeded Up Robust Features (SURF), followed by fast Structured Forests edge detection and then smoothing with a Kalman filter for accurate runway sidelines prediction. A thorough evaluation is performed over real-world and simulation environments with respect to accuracy and processing time, in comparison with state-of-the-art edge detection approaches. The vision system is validated over videos with clear and difficult weather conditions, including with fog, varying lighting conditions and crosswind landing. The experiments are performed using data from the X-Plane 11 flight simulator and real flight data from the Uncrewed Low-cost TRAnsport (ULTRA) self-flying cargo UAV. The vision-led system can localise the runway sidelines with a Structured Forests approach with an accuracy approximately 84.4%, outperforming the state-of-the-art approaches and delivering real-time performance. The main contribution of this work consists of the developed vision-led system for runway detection to aid autonomous landing of UAVs using electro-optical cameras. Although implemented with the ULTRA UAV, the vision-led system is applicable to any other UAV.

YOLOv10-based Model for Player and Football Detection

This study presents an advanced YOLOv10n-based method for the automatic detection of football players and balls directly from match videos. We enhance the YOLOv10 architecture with several significant improvements, including additional detection heads, the integration of C2f_faster and C3_faster modules for enhanced processing speed and accuracy, and the inclusion of BotNet modules with self-attention mechanisms for managing complex visual scenes. Further, we incorporate GhostConv modules to reduce computational overhead while maintaining effective feature extraction. These architectural modifications ensure robust detection capabilities in real-time sports environments, addressing challenges such as high-speed movements, frequent occlusions, and variable lighting conditions typical of both indoor and outdoor stadiums. Validation on internet-sourced images from football matches demonstrates the practicality and effectiveness of our model.

Enhanced Rhodophyta biomass monitoring using aggregated residual transformations for improved generalization under varying light conditions

Enhanced Rhodophyta biomass monitoring using aggregated residual transformations for improved generalization under varying light conditions

Morpho-Physiological Adjustment of Swietenia humilis Zucc. Plants to Varied Nutrient and Light Conditions and Their Performance in Nurseries and Fields Under Soils with Different Preparations

To enhance the plantation performance of Swietenia humilis Zucc., a threatened precious woody species from the dry tropics of Latin America, this study examined its morpho-physiological responses to variations in nutrient and light availability. We established a nursery trial with factorial treatments: three levels of fertilization (4, 6, and 8 g L−1 of substrate using a controlled-release fertilizer, CRF, 18-6-12) and two shade intensities (60% and 40%), alongside a full sun treatment. The field performance of nursery-raised plants was evaluated under two site conditions (with and without mechanical soil preparation) over 48 months. In the nursery, S. humilis exhibited diverse morpho-physiological characteristics influenced by the studied factors, with optimal growth observed at 6 g of the CRF and 40% shade. Mechanical soil preparation significantly improved plant survival, reducing mortality risk by 99.16% and increasing survival probability to nearly 75%. Height growth was also enhanced, being 2.5 times greater in the prepared site compared to that in the unprepared one. S. humilis showed acclimatization in the field, producing new foliage with high chlorophyll content. In conclusion, nursery management and soil site preparation influence the field performance of S. humilis. These findings have practical implications for improving the management of S. humilis in plantations across the dry tropics.

Light condition during grain-filling stage of main crop strongly influences ratooning ability of low-stubble ratoon rice

Compared with high-stubble ratoon rice (RR), low-stubble RR is superior in yield potential, grain quality, and economic benefit. However, the unstable ratooning ability limits the grain yield of low-stubble RR production. Light condition during the grain-filling stage of main crop (GFMC) may be important for rice ratooning. To elucidate the role of light condition during GFMC in affecting ratooning ability, the key response periods, and their underlying mechanisms, field experiments were conducted using two indica cultivars in 2021 and 2022. To create varied light conditions at the canopy base during GFMC, two planting density treatments combining three nitrogen (N) treatments were established in 2021, and three density treatments combining two N treatments and four shading treatments were established in 2022 for the main crop. Light intensity (LI), light quality as reflected by the ratio of red light/far red light (R/FR), and light transmission ratio (LTR) at the canopy base during GFMC and ratooning ability were dramatically altered by N fertilization but not by planting density. With increased N application, LTR, root bleeding rate, and maximum ratooning rate significantly decreased in 2021. In 2022, low N rate increased LI, R/FR, and maximum ratooning rate by 155.7-241.4%, 47.4-65.3%, and 15.6-27.5%, respectively, but reduced missing hill percentage (proportion of hills without regenerated tillers to the total number of hills) by 30.0-62.1% compared with high N rate. The missing hill percentage was negatively correlated with the indices of light condition, while the maximum ratooning rate was positively correlated with them for both cultivars. Root activity and the ratios of abscisic acid (ABA) to cytokinins (CTK), indole-3-acetic acid (IAA), and IAA+CTK could explain the effect of light condition during GFMC on ratooning ability. Shading experiment confirmed the effect of light condition on ratooning ability and further revealed that only shading during middle and late GFMC affected ratooning ability. These findings provide new insights into the regulation of ratooning ability, which are useful for developing management practices to increase the grain yield and yield stability of low-stubble RR.

SlowFast-TCN: A Deep Learning Approach for Visual Speech Recognition

Visual Speech Recognition (VSR), commonly referred to as automated lip-reading, is an emerging technology that interprets speech by visually analyzing lip movements. A challenge in VSR where visually distinct words produce similar lip movements is known as homopheme problem. Visemes are the basic visual units of speech that are produced by the lip movements and positions. Furthermore, visemes are typically having shorter durations than words. Consequently, there is less temporal information for distinguishing between different viseme classes, leading to increased visual ambiguity during classification. To address this challenge, viseme classification must not only extract lip image spatial features, but also to handle visemes of varying durations and temporal features. Therefore, this study proposed a new deep learning approach SlowFast-TCN. SlowFast network is used as the frontend architecture to extract the spatio-temporal features of the slow and fast pathways. Temporal Convolutional Network (TCN) is used as the backend architecture to learn the features from the frontend to perform the classification. A comparative ablation analysis to dissect each component of the proposed SlowFast-TCN is performed to evaluate the impact of each component. This study utilizes a benchmark dataset, Lip Reading in Wild (LRW), that focuses on English language. Two subsets of the LRW dataset, comprising of homopheme words and unique words, represent the homophemic and non-homophemic dataset, respectively. The proposed approach is evaluated on varying lighting conditions to assess its performance in real-world scenarios. It was found that illumination can significantly affect the visual data. Key performance metrics, such as accuracy and loss are used to evaluate the effectiveness of the proposed approach. The proposed approach outperforms traditional baseline models in accuracy while maintaining competitive execution time. Its dual-pathway architecture effectively captures both long-term dependencies and short-term motions, leading to better performance in both homophemic and non-homophemic datasets. However, it is less robust when dealing with non-ideal lighting scenarios, indicating the need for further enhancements to handle diverse lighting scenarios. Doi: 10.28991/ESJ-2024-08-06-024 Full Text: PDF

Driver inattention detection system using multi-task cascaded convolutional networks

Driver inattention has emerged as a critical concern impacting road safety, resulting in an alarming surge in accidents and fatalities. This research introduces a novel system for detecting inattention, structured across six levels: perception, facial feature extraction, tracking driver face, and driver secondary task using pre-trained deep learning models, inattention detection, risk estimation, and alert. The system is based on image processing captured from two strategically positioned cameras that simultaneously capture the driver’s activities while driving and their facial expressions. The second contribution concerns the driver facial features extraction using multi-task cascaded convolutional networks (MTCNN), and it is comparison with the histogram of gradient (HOG)-based frontal face detector, and haar feature based cascade classifier. The algorithms were compared based on their runtime efficiency, robustness in handling varying lighting conditions, and various head movements. The MTCNN achieves high performance, reaching accuracy levels ranging from 96.4% to 99.5% on two datasets including realistic driving scenarios: the DrivFace dataset and, the driver drowsiness dataset. The comparative analysis sheds light on the strengths and weaknesses of each algorithm, providing valuable insights for selecting the most suitable face detection algorithm to use in our system.

New photochromic spectacle lenses improve glare discomfort and photostress recovery

PurposeTo investigate the influence of a new photochromic spectacle lens on glare discomfort (GD) and photostress recovery time (PRT) in normal healthy subjects. A subject-masked within-subject comparison was performed using the fully activated state of Transitions® Gen STM and clear control. Methods30 participants (M = 19.2 ± 1.3 years) were tested. GD was assessed both physically, by measuring palpebral fissure size, and subjectively, via a 5-point Likert scale. PRT was measured as the amount of time necessary to regain visual function after an intense (∼15 % bleach) 5-second exposure to a broadband (emulated sunlight) light stressor. ResultsGD, measured both physically and subjectively was significantly reduced (p < 0.001) when participants viewed the stimulus through the test lens compared to the clear control lens. PRT was also significantly reduced when using the test lens (p < 0.001). ConclusionsThese data are consistent with past findings but show that they apply to a commonly used photochromic tint. The results highlight the potential of specialized lenses to enhance visual comfort and performance in varying lighting conditions.

6D Pose Estimation of Industrial Parts Based on Point Cloud Geometric Information Prediction for Robotic Grasping

In industrial robotic arm gripping operations within disordered environments, the loss of physical information on the object’s surface is often caused by changes such as varying lighting conditions, weak surface textures, and sensor noise. This leads to inaccurate object detection and pose estimation information. A method for industrial object pose estimation using point cloud data is proposed to improve pose estimation accuracy. During the feature extraction process, both global and local information are captured by integrating the appearance features of RGB images with the geometric features of point clouds. Integrating semantic information with instance features effectively distinguishes instances of similar objects. The fusion of depth information and RGB color channels enriches spatial context and structure. A cross-entropy loss function is employed for multi-class target classification, and a discriminative loss function enables instance segmentation. A novel point cloud registration method is also introduced to address re-projection errors when mapping 3D keypoints to 2D planes. This method utilizes 3D geometric information, extracting edge features using point cloud curvature and normal vectors, and registers them with models to obtain accurate pose information. Experimental results demonstrate that the proposed method is effective and superior on the LineMod and YCB-Video datasets. Finally, objects are grasped by deploying a robotic arm on the grasping platform.

The Influence of Screen Brightness and Saturation on Emotional Perception in Human-Machine Interactions: A Functional Near-Infrared Spectroscopy Study

The prefrontal cortex (PFC) is essential in emotional and cognitive processes, yet the effects of light and color on brain activity and emotional perception remain underexplored. This study examined the influence of screen brightness and saturation on emotional perception in human-machine interfaces under simulated daytime, cloudy, and nighttime conditions, with 29 volunteers participating. Functional near-infrared spectroscopy assessed PFC hemodynamics. Results indicated higher pleasantness for low versus high saturation under both high and low brightness, while high brightness and saturation were more attractive to participants. High-saturation screens activated the orbitofrontal cortex (OFC) and left dorsolateral PFC (dIPFC) under high brightness, whereas low saturation increased left dIPFC activation under low brightness. OFC activation positively correlated with pleasantness, while left dIPFC activation negatively correlated with attractiveness. These findings highlight the neurophysiological impact of light and color on emotional perception, providing insights to optimize interface design and user experience in varying lighting conditions.

Developing WasteSAM: A novel approach for accurate construction waste image segmentation to facilitate efficient recycling

The escalating volume of construction activities and resultant waste generation underscores the imperative for developing sophisticated segmentation models to facilitate efficient sorting and recycling processes. This study introduces WasteSAM, an enhanced iteration of the segment anything model (SAM), specifically tailored to address the intricate complexities inherent in construction waste imagery. Drawing upon a comprehensive dataset comprising over 15,000 masks representing five distinct categories of construction materials, WasteSAM exhibits notably superior segmentation capabilities. Quantitative analysis demonstrates significant performance improvements, with WasteSAM outperforming the original SAM model by an average of 23.9% in dice similarity coefficient and 30.0% in normalized surface distance metrics. The integration of stereo-image techniques in refining the training dataset has facilitated WasteSAM in more accurately discerning the three-dimensional structure of waste materials, thereby augmenting the precision of waste classification. Noteworthy is the model’s adeptness in handling intricate textures and patterns across diverse imaging modalities, including varying lighting conditions and complex object interactions. While showing promising results, this study also highlights the need for high-quality, diverse datasets that reflect real-world construction site complexities, rather than merely larger datasets.

Multi-Task Water Quality Colorimetric Detection Method Based on Deep Learning.

The colorimetric method, due to its rapid and low-cost characteristics, demonstrates a wide range of application prospects in on-site water quality testing. Current research on colorimetric detection using deep learning algorithms predominantly focuses on single-target classification. To address this limitation, we propose a multi-task water quality colorimetric detection method based on YOLOv8n, leveraging deep learning techniques to achieve a fully automated process of "image input and result output". Initially, we constructed a dataset that encompasses colorimetric sensor data under varying lighting conditions to enhance model generalization. Subsequently, to effectively improve detection accuracy while reducing model parameters and computational load, we implemented several improvements to the deep learning algorithm, including the MGFF (Multi-Scale Grouped Feature Fusion) module, the LSKA-SPPF (Large Separable Kernel Attention-Spatial Pyramid Pooling-Fast) module, and the GNDCDH (Group Norm Detail Convolution Detection Head). Experimental results demonstrate that the optimized deep learning algorithm excels in precision (96.4%), recall (96.2%), and mAP50 (98.3), significantly outperforming other mainstream models. Furthermore, compared to YOLOv8n, the parameter count and computational load were reduced by 25.8% and 25.6%, respectively. Additionally, precision improved by 2.8%, recall increased by 3.5%, mAP50 enhanced by 2%, and mAP95 rose by 1.9%. These results affirm the substantial potential of our proposed method for rapid on-site water quality detection, offering new technological insights for future water quality monitoring.

ALFusion: Adaptive fusion for infrared and visible images under complex lighting conditions

In the task of infrared and visible image fusion, source images often exhibit complex and variable characteristics due to scene illumination. To address the challenges posed by complex lighting conditions and enhance the quality of fused images, we develop the ALFusion method, which combines dynamic convolution and Transformer for infrared and visible image fusion. The core idea is to utilize the adaptability of dynamic convolution coupled with the superior long-range modeling capabilities of the Transformer to design a hybrid feature extractor. This extractor dynamically and comprehensively captures features from source images under various lighting conditions. To enhance the effectiveness of mixed features, we integrate an elaborate multi-scale attention enhancement module into the skip connections of the U-Net architecture. In this module, we use convolutional kernels of various sizes to expand the receptive field and incorporate an attention mechanism to enhance and highlight the combined use of dynamic convolution and the Transformer. Considering the performance in advanced visual tasks, an illumination detection network is integrated into the loss function, strategically balancing the pixel fusion ratio and optimizing visual impacts under varied lighting conditions, leveraging information from different source images. The fusion results indicate that our method consistently delivers superior background contrast and enhanced texture and structural features across diverse lighting conditions. Ablation and complementary experiments have been conducted, further affirming the effectiveness of our proposed method and highlighting its potential in advanced visual tasks.

Design and Implementation of an AI-Based Robotic Arm for Strawberry Harvesting

This study presents the design and implementation of a wire-driven, multi-joint robotic arm equipped with a cutting and gripping mechanism for harvesting delicate strawberries, with the goal of reducing labor and costs. The arm is mounted on a lifting mechanism and linked to a laterally movable module, which is affixed to the tube cultivation shelf. The trained deep learning model can instantly detect strawberries, identify optimal picking points, and estimate the contour area of fruit while the mobile platform is in motion. A two-stage fuzzy logic control (2s-FLC) method is employed to adjust the length of the arm and bending angle, enabling the end of the arm to approach the fruit picking position. The experimental results indicate a 90% accuracy in fruit detection, an 82% success rate in harvesting, and an average picking time of 6.5 s per strawberry, reduced to 5 s without arm recovery time. The performance of the proposed system in harvesting strawberries of different sizes under varying lighting conditions is also statistically analyzed and evaluated in this paper.

Nitrogen at the crossroads of light: Integration of light signalling and plant nitrogen metabolism.

Plants have developed complex mechanisms to perceive, transduce, and respond to environmental signals, such as light, which are essential for acquiring and allocating resources, including nitrogen (N). This review delves into the complex interaction between light signals and N metabolism, emphasising light-mediated regulation of N uptake and assimilation. Firstly, we discuss the details of light-mediated regulation of N uptake and assimilation, focusing on the light-responsive activity of nitrate reductase (NR) and nitrate transporters. Secondly, we discuss the influence of light on N-dependent developmental plasticity, elucidating how N availability regulates crucial developmental transitions such as flowering time, shoot branching, and root growth, as well as how light modulates these processes. Additionally, we discuss the molecular interaction between light and N signalling, focusing on photoreceptors and transcription factors such as HY5, which are necessary for N uptake and assimilation under varying light conditions. A recent understanding of the nitrate signalling and perception of low N is also highlighted. The insilico transcriptome analysis suggests a reprogramming of N signalling genes by shade and identifies NLP7, bZIP1, CPK30, CBL1, LBD37, LBD38 and HRS1 as crucial molecular regulators integrating light-regulated N metabolism.

Attention enhanced machine instinctive vision with human-inspired saliency detection

Salient object detection (SOD) enables machines to recognize and accurately segment visually prominent regions in images. Despite recent advancements, existing approaches often lack progressive fusion of low and high-level features, effective multi-scale feature handling, and precise boundary detection. Moreover, the robustness of these models under varied lighting conditions remains a concern. To overcome these challenges, we present Attention Enhanced Machine Instinctive Vision framework for SOD. The proposed framework leverages the strategy of Multi-stage Feature Refinement with Optimal Attentions-Driven Framework (MFRNet). The multi-level features are extracted from six stages of the EfficientNet-B7 backbone. This provides effective feature fusions of low and high-level details across various scales at the later stage of the framework. We introduce the Spatial-optimized Feature Attention (SOFA) module, which refines spatial features from three initial-stage feature maps. The extracted multi-scale features from the backbone are passed from the convolution feature transformation and spatial attention mechanisms to refine the low-level information. The SOFA module concatenates and upsamples these refined features, producing a comprehensive spatial representation of various levels. Moreover, the proposed Context-Aware Channel Refinement (CACR) module integrates dilated convolutions with optimized dilation rates followed by channel attention to capture multi-scale contextual information from the mature three layers. Furthermore, our progressive feature fusion strategy combines high-level semantic information and low-level spatial details through multiple residual connections, ensuring robust feature representation and effective gradient backpropagation. To enhance robustness, we train our network with augmented data featuring low and high brightness adjustments, improving its ability to handle diverse lighting conditions. Extensive experiments on four benchmark datasets — ECSSD, HKU-IS, DUTS, and PASCAL-S — validate the proposed framework’s effectiveness, demonstrating superior performance compared to existing SOTA methods in the domain. Code, qualitative results, and trained weights will be available at the link: https://github.com/habib1402/MFRNet-SOD.

Identification and Expression Analysis of TCP Transcription Factors Under Abiotic Stress in Phoebe bournei.

The TCP gene family encodes plant transcription factors crucial for regulating growth and development. While TCP genes have been identified in various species, they have not been studied in Phoebe bournei (Hemsl.). This study identified 29 TCP genes in the P. bournei genome, categorizing them into Class I (PCF) and Class II (CYC/TB1 and CIN). We conducted analyses on the PbTCP gene at both the protein level (physicochemical properties) and the gene sequence level (subcellular localization, chromosomal distribution, phylogenetic relationships, conserved motifs, and gene structure). Most P. bournei TCP genes are localized in the nucleus, except PbTCP9 in the mitochondria and PbTCP8 in both the chloroplast and nucleus. Chromosomal mapping showed 29 TCP genes unevenly distributed across 10 chromosomes, except chromosome 8 and 9. We also analyzed the promoter cis-regulatory elements, which are mainly involved in plant growth and development and hormone responses. Notably, most PbTCP transcription factors respond highly to light. Further analysis revealed three subfamily genes expressed in five P. bournei tissues: leaves, root bark, root xylem, stem xylem, and stem bark, with predominant PCF genes. Using qRT-PCR, we examined six representative genes-PbTCP16, PbTCP23, PbTCP7, PbTCP29, PbTCP14, and PbTCP15-under stress conditions such as high temperature, drought, light exposure, and dark. PbTCP14 and PbTCP15 showed significantly higher expression under heat, drought, light and dark stress. We hypothesize that TCP transcription factors play a key role in growth under varying light conditions, possibly mediated by auxin hormones. This work provides insights into the TCP gene family's functional characteristics and stress resistance regulation in P. bournei.

Unique structural attributes of the PSI-NDH supercomplex in Physcomitrium patens.

Cyclic electron transport around photosystem I (PSI) is essential for the protection of the photosynthetic apparatus in plants under diverse light conditions. This process is primarily mediated by Proton Gradient Regulation 5 protein/Proton Gradient Regulation 5-like photosynthetic phenotype 1 protein (PGR5/PGRL1) and NADH dehydrogenase-like complex (NDH). In angiosperms, NDH interacts with two PSI complexes through distinct monomeric antennae, LHCA5 and LHCA6, which is crucial for its higher stability under variable light conditions. This interaction represents an advanced evolutionary stage and offers limited insight into the origin of the PSI-NDH supercomplex in evolutionarily older organisms. In contrast, the moss Physcomitrium patens (Pp), which retains the lhca5 gene but lacks the lhca6, offers a glimpse into an earlier evolutionary stage of the PSI-NDH supercomplex. Here we present structural evidence of the Pp PSI-NDH supercomplex formation by single particle electron microscopy, demonstrating the unique ability of Pp to bind a single PSI in two different configurations. One configuration closely resembles the angiosperm model, whereas the other exhibits a novel PSI orientation, rotated clockwise. This structural flexibility in Pp is presumably enabled by the variable incorporation of LHCA5 within PSI and is indicative of an early evolutionary adaptation that allowed for greater diversity at the PSI-NDH interface. Our findings suggest that this variability was reduced as the structural complexity of the NDH complex increased in vascular plants, primarily angiosperms. This study not only clarifies the evolutionary development of PSI-NDH supercomplexes but also highlights the dynamic nature of the adaptive mechanisms of plant photosynthesis.

Light-driven differences in bacterial networks and organic matter decomposition: Insights from an analysis of the harmful cyanobacterium Microcystis aeruginosa PCC 7806

Freshwater systems are critical yet often underestimated components of global carbon cycling, functioning both as carbon sinks and sources. Cyanobacteria play a key role in this cycle by capturing atmospheric carbon dioxide through photosynthesis. The captured carbon is either released back into the atmosphere or sequestered in sediments following organismal decay. This study examines the pivotal role of cyanobacteria, specifically Microcystis aeruginosa PCC 7806, in the biogeochemical cycling of carbon in freshwater ecosystems, with a focus on how light influences the degradation of cyanobacteria-derived organic matter. Using a combination of 16S rDNA sequencing and excitation-emission matrix coupled with parallel factor (EEM-PARAFAC) analysis, we conducted a 50-day experiment to investigate the dynamics of dissolved organic matter (DOM) and lysate organic matter (LOM) derived from M. aeruginosa PCC 7806 under light and dark conditions. Our results demonstrate that light significantly impacts bacterial community composition, gene functionality, and the decomposition of organic matter. The findings emphasize the crucial role of light in facilitating microbial adaptation, stabilizing microbial networks and driving organic substrate transformation. These insights underscore the influence of light on microbial community dynamics and organic matter degradation, revealing shifts in microbial populations under varying light conditions. This suggests a strong link between photochemical processes and microbial activity, with significant ecological implications.