Visual Research Articles

Skin-Friendly Large Matrix Iontronic Sensing Meta-Fabric for Spasticity Visualization and Rehabilitation Training via Piezo-Ionic Dynamics

Rehabilitation training is believed to be an effectual strategy that can reduce the risk of dysfunction caused by spasticity. However, achieving visualization rehabilitation training for patients remains clinically challenging. Herein, we propose visual rehabilitation training system including iontronic meta-fabrics with skin-friendly and large matrix features, as well as high-resolution image modules for distribution of human muscle tension. Attributed to the dynamic connection and dissociation of the meta-fabric, the fabric exhibits outstanding tactile sensing properties, such as wide tactile sensing range (0 ~ 300kPa) and high-resolution tactile perception (50Pa or 0.058%). Meanwhile, thanks to the differential capillary effect, the meta-fabric exhibits a "hitting three birds with one stone" property (dryness wearing experience, long working time and cooling sensing). Based on this, the fabrics can be integrated with garments and advanced data analysis systems to manufacture a series of large matrix structure (40 × 40, 1600 sensing units) training devices. Significantly, the tunability of piezo-ionic dynamics of the meta-fabric and the programmability of high-resolution imaging modules allow this visualization training strategy extendable to various common disease monitoring. Therefore, we believe that our study overcomes the constraint of standard spasticity rehabilitation training devices in terms of visual display and paves the way for future smart healthcare.

Using three-dimensional virtual imaging of renal masses to improve prediction of robotic-assisted partial nephrectomy Tetrafecta with SPARE score.

To improve the predictability of outcomes in robotic-assisted partial nephrectomy, we utilized three-dimensional virtual imaging for SPARE nephrometry scoring. We compared this method with a conventional two-dimensional scoring system to determine whether 3D virtual images offer enhanced predictive accuracy for Tetrafecta outcomes. We retrospectively collected basic information, demographic data, and perioperative indices from patients who underwent robot-assisted partial nephrectomy for renal masses at the Department of Urology, First Affiliated Hospital of Xi'an Jiaotong University. A three-dimensional visualization system (IPS system, Yorktal) was employed to reconstruct the patients' imaging data using AI-based automatic segmentation, resulting in a three-dimensional visualization model (3DVM). This model was then imported into the virtual surgical planning software (Touch Viewer System, Yorktal) for automatic measurement of the SPARE score. Tetrafecta was defined as a warm ischemic time (WIT) of less than 25min, negative surgical margins, absence of major perioperative complications, and no decline in postoperative renal function. The receiver operating characteristic (ROC) curve was utilized to evaluate the sensitivity and specificity of the SPARE score. A total of 141 patients were included in this study, with a mean age of 55.6 ± 11.14years and a mean tumor size of 3.5 ± 1.2cm. All variables, except for estimated blood loss (EBL) (Coefficient = 0.056, 0.035; P = 0.514, 0.685), showed significant correlation with the SPARE score when comparing CT and 3D virtual models (Tetrafecta: Coefficient = 0.408, 0.56; P < 0.001, < 0.001; WIT: Coefficient = 0.340, 0.237; P < 0.001, 0.007; ΔeGFR: Coefficient = 0.212, 0.257; P = 0.012, 0.002). The area under the curve (AUC) values from the ROC curves indicated that the 3D virtual model group had significantly better performance than the 2D image group for the SPARE score. However, there was no significant difference in the ROC curves for the SPARE complexity category between the two imaging modalities (AUC for SPARE category with 3DVM = 0.658 vs. AUC for SPARE category with CT = 0.643, P = 0.59; AUC for SPARE score with 3DVM = 0.854 vs. AUC for SPARE score with CT = 0.755, P < 0.001). The SPARE score combined with 3DVM has a more accurate predictive ability for Tetrafecta of RAPN compared to the traditional 2D SPARE score.

Video-Wise Just-Noticeable Distortion Prediction Model for Video Compression with a Spatial–Temporal Network

Just-Noticeable Difference (JND) in an image/video refers to the maximum difference that the human visual system cannot perceive, which has been widely applied in perception-guided image/video compression. In this work, we propose a Binary Decision-based Video-Wise Just-Noticeable Difference Prediction Method (BD-VW-JND-PM) with deep learning. Firstly, we model the VW-JND prediction problem as a binary decision process to reduce the inferring complexity. Then, we propose a Perceptually Lossy/Lossless Predictor for Compressed Video (PLLP-CV) to identify whether the distortion can be perceived or not. In the PLLP-CV, a Spatial–Temporal Network-based Perceptually Lossy/Lossless predictor (ST-Network-PLLP) is proposed for key frames by learning the spatial and temporal distortion features, and a threshold-based integration strategy is proposed to obtain the final results. Experimental results evaluated on the VideoSet database show that the mean prediction accuracy of PLLP-CV is about 95.6%, and the mean JND prediction error is 1.46 in QP and 0.74 in Peak-to-Noise Ratio (PSNR), which achieve 15% and 14.9% improvements, respectively.

Eye-Inspired Single-Pixel Imaging with Lateral Inhibition and Variable Resolution for Special Unmanned Vehicle Applications in Tunnel Inspection

This study presents a cutting-edge imaging technique for special unmanned vehicles (UAVs) designed to enhance tunnel inspection capabilities. This technique integrates ghost imaging inspired by the human visual system with lateral inhibition and variable resolution to improve environmental perception in challenging conditions, such as poor lighting and dust. By emulating the high-resolution foveal vision of the human eye, this method significantly enhances the efficiency and quality of image reconstruction for fine targets within the region of interest (ROI). This method utilizes non-uniform speckle patterns coupled with lateral inhibition to augment optical nonlinearity, leading to superior image quality and contrast. Lateral inhibition effectively suppresses background noise, thereby improving the imaging efficiency and substantially increasing the signal-to-noise ratio (SNR) in noisy environments. Extensive indoor experiments and field tests in actual tunnel settings validated the performance of this method. Variable-resolution sampling reduced the number of samples required by 50%, enhancing the reconstruction efficiency without compromising image quality. Field tests demonstrated the system’s ability to successfully image fine targets, such as cables, under dim and dusty conditions, achieving SNRs from 13.5 dB at 10% sampling to 27.7 dB at full sampling. The results underscore the potential of this technique for enhancing environmental perception in special unmanned vehicles, especially in GPS-denied environments with poor lighting and dust.

Artificial Intelligence-Based Methodologies for Early Diagnostic Precision and Personalized Therapeutic Strategies in Neuro-Ophthalmic and Neurodegenerative Pathologies

Advancements in neuroimaging, particularly diffusion magnetic resonance imaging (MRI) techniques and molecular imaging with positron emission tomography (PET), have significantly enhanced the early detection of biomarkers in neurodegenerative and neuro-ophthalmic disorders. These include Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, neuromyelitis optica, and myelin oligodendrocyte glycoprotein antibody disease. This review highlights the transformative role of advanced diffusion MRI techniques—Neurite Orientation Dispersion and Density Imaging and Diffusion Kurtosis Imaging—in identifying subtle microstructural changes in the brain and visual pathways that precede clinical symptoms. When integrated with artificial intelligence (AI) algorithms, these techniques achieve unprecedented diagnostic precision, facilitating early detection of neurodegeneration and inflammation. Additionally, next-generation PET tracers targeting misfolded proteins, such as tau and alpha-synuclein, along with inflammatory markers, enhance the visualization and quantification of pathological processes in vivo. Deep learning models, including convolutional neural networks and multimodal transformers, further improve diagnostic accuracy by integrating multimodal imaging data and predicting disease progression. Despite challenges such as technical variability, data privacy concerns, and regulatory barriers, the potential of AI-enhanced neuroimaging to revolutionize early diagnosis and personalized treatment in neurodegenerative and neuro-ophthalmic disorders is immense. This review underscores the importance of ongoing efforts to validate, standardize, and implement these technologies to maximize their clinical impact.

Advanced Monocular Outdoor Pose Estimation in Autonomous Systems: Leveraging Optical Flow, Depth Estimation, and Semantic Segmentation with Dynamic Object Removal

Autonomous technologies have revolutionized transportation, military operations, and space exploration, necessitating precise localization in environments where traditional GPS-based systems are unreliable or unavailable. While widespread for outdoor localization, GPS systems face limitations in obstructed environments such as dense urban areas, forests, and indoor spaces. Moreover, GPS reliance introduces vulnerabilities to signal disruptions, which can lead to significant operational failures. Hence, developing alternative localization techniques that do not depend on external signals is essential, showing a critical need for robust, GPS-independent localization solutions adaptable to different applications, ranging from Earth-based autonomous vehicles to robotic missions on Mars. This paper addresses these challenges using Visual odometry (VO) to estimate a camera’s pose by analyzing captured image sequences in GPS-denied areas tailored for autonomous vehicles (AVs), where safety and real-time decision-making are paramount. Extensive research has been dedicated to pose estimation using LiDAR or stereo cameras, which, despite their accuracy, are constrained by weight, cost, and complexity. In contrast, monocular vision is practical and cost-effective, making it a popular choice for drones, cars, and autonomous vehicles. However, robust and reliable monocular pose estimation models remain underexplored. This research aims to fill this gap by developing a novel adaptive framework for outdoor pose estimation and safe navigation using enhanced visual odometry systems with monocular cameras, especially for applications where deploying additional sensors is not feasible due to cost or physical constraints. This framework is designed to be adaptable across different vehicles and platforms, ensuring accurate and reliable pose estimation. We integrate advanced control theory to provide safety guarantees for motion control, ensuring that the AV can react safely to the imminent hazards and unknown trajectories of nearby traffic agents. The focus is on creating an AI-driven model(s) that meets the performance standards of multi-sensor systems while leveraging the inherent advantages of monocular vision. This research uses state-of-the-art machine learning techniques to advance visual odometry’s technical capabilities and ensure its adaptability across different platforms, cameras, and environments. By merging cutting-edge visual odometry techniques with robust control theory, our approach enhances both the safety and performance of AVs in complex traffic situations, directly addressing the challenge of safe and adaptive navigation. Experimental results on the KITTI odometry dataset demonstrate a significant improvement in pose estimation accuracy, offering a cost-effective and robust solution for real-world applications.

DE-RGBD SLAM: Enhancing Static Feature Point Selection in RGB-D Visual SLAM Using Depth Information

Abstract Feature point extraction plays a key role in visual simultaneous localization and mapping (SLAM) systems. And it remains a major challenge to accurately select static feature points in a complex dynamic environment. To address this issue, this paper proposes an RGB-D SLAM method, referred to as DE-RGBD SLAM, which optimizes feature selection by integrating depth information and effectively utilizes depth data and multi-view geometric information to achieve localization and navigation for mobile robots in dynamic environments. Firstly, the method analyzes prominent feature regions in the image based on colour and depth information captured by an RGB-D camera. It sets adaptive FAST corner detection thresholds according to the grayscale information of these regions while masking other areas. Next, the method obtains in-depth information on the detected feature points in the current frame. It combines their pixel coordinates in the image coordinate system to determine the presence of redundant feature points. Notably, the method can detect some dynamic feature points between consecutive frames. Subsequently, in the camera coordinate system, the method compares the depth information of feature points in the depth image with the epipolar depth estimates derived from the essential matrix to determine whether the features are static and eliminate dynamic feature points. This approach significantly enhances the reliability of static feature points. Finally, the accuracy and robustness of the proposed method are validated through experiments conducted on the public TUM dataset and real-world scenarios compared to state-of-the-art visual SLAM systems.

An Enhanced Cycle Generative Adversarial Network Approach for Nighttime Pineapple Detection of Automated Harvesting Robots

Nighttime pineapple detection for automated harvesting robots is a significant challenge in intelligent agriculture. As a crucial component of robotic vision systems, accurate fruit detection is essential for round-the-clock operations. The study compared advanced end-to-end style transfer models, including U-GAT-IT, SCTNet, and CycleGAN, finding that CycleGAN produced relatively good-quality images but had issues such as the inadequate restoration of nighttime details, color distortion, and artifacts. Therefore, this study further proposed an enhanced CycleGAN approach to address limited nighttime datasets and poor visibility, combining style transfer with small-sample object detection. The improved model features a novel generator structure with ResNeXtBlocks, an optimized upsampling module, and a hyperparameter optimization strategy. This approach achieves a 29.7% reduction in FID score compared to the original CycleGAN. When applied to YOLOv7-based detection, this method significantly outperforms existing approaches, improving precision, recall, average precision, and F1 score by 13.34%, 45.11%, 56.52%, and 30.52%, respectively. These results demonstrate the effectiveness of our enhanced CycleGAN in expanding limited nighttime datasets and supporting efficient automated harvesting in low-light conditions, contributing to the development of more versatile agricultural robots capable of continuous operation.

Computer Vision System for Multi-Robot Construction Waste Management: Integrating Cloud and Edge Computing

Sorting is an important construction waste management tool to increase recycling rates and reduce pollution. Previous studies have used robots to improve the efficiency of construction waste recycling. However, in large construction sites, it is difficult for a single robot to accomplish the task quickly, and multiple robots working together are a better option. Most construction waste recycling robotic systems are developed based on a client-server framework, which means that all robots need to be continuously connected to their respective cloud servers. Such systems are low in robustness in complex environments and waste a lot of computational resources. Therefore, in this paper, we propose a pixel-level automatic construction waste recognition platform with high robustness and low computational resource requirements by combining multiple computer vision technologies with edge computing and cloud computing platforms. Experiments show that the computing platform proposed in this study can achieve a recognition speed of 23.3 fps and a recognition accuracy of 90.81% at the edge computing platform without the help of network and cloud servers. This is 23 times faster than the algorithm used in previous research. Meanwhile, the computing platform proposed in this study achieves 93.2% instance segmentation accuracy on the cloud server side. Notably, this system allows multiple robots to operate simultaneously at the same construction site using only a single server without compromising efficiency, which significantly reduces costs and promotes the adoption of automated construction waste recycling robots.

Bio-inspired snapshot polarization-hyperspectral camera

The mantis shrimp is recognized to have one of the most powerful vision systems in nature, with up to 16 color-perceiving channels and the perception of linear and circular polarization detection. Inspired by its biostructure, we developed a snapshot polarization-hyperspectral camera (pHScam) to detect linear polarization in four directions and spectral signature in 21 bands of incident light, resulting in a 4D polar-spectral hypercube, represented as H(x,y,l,S→). We introduced two bio-mimetic encoding mechanisms, viz., two-tier color encoding and pixelwise mosaic polarization encoding, to efficiently compress the original 4D datacube into a 2D mosaic raw pattern. The full-resolution recovery is solved by convex optimization algorithm, utilizing known prior sparsity and experimental system calibrations.

Adltformer Team-Training with Detr: Enhancing Cattle Detection in Non-Ideal Lighting Conditions Through Adaptive Image Enhancement

This study proposes an image enhancement detection technique based on Adltformer (Adaptive dynamic learning transformer) team-training with Detr (Detection transformer) to improve model accuracy in suboptimal conditions, addressing the challenge of detecting cattle in real pastures under complex lighting conditions—including backlighting, non-uniform lighting, and low light. This often results in the loss of image details and structural information, color distortion, and noise artifacts, thereby compromising the visual quality of captured images and reducing model accuracy. To train the Adltformer enhancement model, the day-to-night image synthesis (DTN-Synthesis) algorithm generates low-light image pairs that are precisely aligned with normal light images and include controlled noise levels. The Adltformer and Detr team-training (AT-Detr) method is employed to preprocess the low-light cattle dataset for image enhancement, ensuring that the enhanced images are more compatible with the requirements of machine vision systems. The experimental results demonstrate that the AT-Detr algorithm achieves superior detection accuracy, with comparable runtime and model complexity, reaching 97.5% accuracy under challenging illumination conditions, outperforming both Detr alone and sequential image enhancement followed by Detr. This approach provides both theoretical justification and practical applicability for detecting cattle under challenging conditions in real-world farming environments.

Design of a computer vision system for accurate measurement of mango skin defects by surface image slicing

Abstract External features such as color, size, weight, and defect area are among the critical parameters for grading mangoes for export purposes. This paper describes a novel computer vision system (CVS) for the accurate measurement of mango skin defects. A novel algorithm called mango skin image slicing (MSIS) was developed to generate mango slices, each of which includes points of pseudo-equidistance from the camera to accurately calculate the defect area. Based on images captured from 360-degree views, defects on the entire mango surface can be detected. For the first time, a strategy for preventing multiple considerations of the same defects from 360-degree images was introduced based on MSIS. The experimental results revealed that recalculations of the same defects were avoided, and negligible minor defects were excluded, complying with the guidance of the mango exporting firms. The measurement accuracy was evaluated with artificial defects of 100 mm2, yielding a mean error of 6.0 ± 1.4 mm2, which was suitable for defect area measurement considering the mango grading standard. The CVS throughput of approximately 644 mangoes per hour was also a reasonable trade-off for the achieved accuracy and great potential of extending the CVS for more sophisticated classification of non-negligible defects based on the advantages of mango images captured at closer camera–object distances.

An efficient no‐reference image quality analysis based on statistical perceptual features

AbstractIt is well known that image quality needs to be measured with human perception in many computer vision applications. However, these approaches are expensive and require more time for image quality analysis. Therefore, this paper proposes a robust and computationally efficient objective‐mathematical model based on statistical perceptual features. The structural and textural features are computed using the modified regularized heaviside local binary pattern (RH‐LBP) approach and the concept of entropy. The higher‐order probability coefficients of images are considered to extract features that are highly correlated to the human visual system features. Further, the additivity property of Renyi entropy is used to show the randomness of the information combining two terms: One extracts the images spatial intensity changes, and therefore their texture qualities, and the other attain structural details. The features in the proposed approach are jointly optimized to achieve better robustness, monotonicity and match human assessments on image quality, while minimizing the computational complexity and run time. Experiments are conducted with three synthetically distorted datasets, KonIQ‐10K, BIQ2021, and LIVE (wild), and two intentionally distorted datasets, TID2013 and CSIQ and are used to evaluate performance index. The proposed method offers competitive performance compared with state‐of‐the‐art methods.

Refining Flash Visual Evoked Potential Analysis in Rats: A Novel Approach Using Bilateral Epidural Electrodes.

Visual evoked potentials (VEPs) are electrical signals generated at the visual cortex following visual stimulation. Flash VEPs (fVEPs) are produced by global retinal stimulation and are considered an objective measure of the integrity of the entire visual pathway. However, fVEP measurements are highly sensitive to external variables, making relative comparisons of the fVEP waveforms between the two eyes in the same individual challenging. We used the rodent non-arteritic anterior ischemic optic neuropathy (rNAION) model to induce unilateral ischemic optic neuropathy. The severity of optic disc edema was measured with spectral-domain optical coherence tomography, and visual acuity was measured using a virtual optokinetic system. We developed a procedure utilizing implanted bilateral epidural electrodes and derived a mathematical formula to accurately estimate functional differences between the optic nerves. Immunohistology was performed to quantify retinal ganglion cell (RGC) survival using stereology. Compared to subcutaneous methods, the new approach significantly improves the signal-to-noise ratio and is more repeatable when comparing the two eyes. The derived formula accounts for asymmetry in the afferent inputs to the visual cortex. Visual function calculated using the formula correlates strongly with other recognized metrics of visual function, including RGC survival and visual acuity. We have developed a repeatable and accurate method to calculate the relative visual function of diseased optic nerves compared with a contralateral control eye. Our novel method improves fVEP measurement sensitivity and accuracy in rodent preclinical trials, reducing the number of animals needed to achieve statistical significance.

Systemic Effects of Molar and Incisor Biting on Walking Direction With and Without Visual Feedback

Gait stability and walking direction control are conventionally attributed to coordination among somatosensory, visual, and vestibular systems. Recent evidence of functional interdependence between masticatory and neuromuscular systems indicates that the stomatognathic system is neurologically integrated with various body systems relevant to movement planning and execution. This study investigated the effects of unilateral molar biting and incisor biting on walking with and without visual feedback. A cohort of 31 healthy young adults aged 21 to 30 years (average age of 23.93 ± 1.89) participated in this study. Three types of errors in walking direction (angle error, position error, and curve error) were computed. Our findings indicate that, in right-handed individuals, irrespective of visual feedback, unilateral biting caused systematic deviations toward the biting side from initiation to termination of walking. The consistent deviation in walking, particularly during unilateral right biting conditions in right-handed individuals, may indicate a complex interplay between masticatory function and gait control mechanism, potentially influenced by handedness and motor lateralization within the cortex. This study establishes a foundation for future research exploring the interrelation between bite location, visual feedback, and motor control in diverse populations. This research may provide insight for more efficient interventions for gait-related disorders.

Observation on the efficacy of cataract surgery combined with ICL implantation in the contralateral eye for patients with high myopia complicated with monocular cataract

Objective: To assess the surgical outcomes of cataract surgery combined with phakic posterior chamber intraocular lens (ICL) implantation in the contralateral eye for patients with high myopia and monocular cataract. Methods: This is a retrospective case series study. Clinical data were collected from 10 patients (20 eyes) with high myopia and monocular cataract who underwent phacoemulsification cataract extraction and intraocular lens implantation in the cataract eye (cataract surgery eye), and implantable Collamer lens (ICL) implantation in the contralateral eye (ICL implantation eye) at the Ophthalmology Department of the Army Characteristic Medical Center from June 2021 to December 2023. Among them, there were 4 males and 6 females, with an average age of (35.60±7.62) years. Uncorrected visual acuity was measured using the international standard visual acuity chart, and spherical equivalent (SE) and best corrected visual acuity (BCVA) were obtained through comprehensive optometry. Visual acuity was recorded in the form of logarithm of the minimum angle of resolution (logMAR). Objective visual quality parameters were obtained using a visual quality analysis system, and subjective visual quality was obtained through questionnaire surveys. Postoperative accommodative function and defocus curve were also examined. Various indicators before and 5 months after surgery were compared between the two eyes. Results: The BCVA in the cataract surgery eye significantly improved postoperatively (P<0.05), while no significant change was observed in the ICL implantation eye (P>0.05). Postoperative UCVA for the cataract surgery eye was 0.12±0.12 for distance, 0.18±0.18 for intermediate, and 0.28±0.20 for near vision. For the ICL implantation eye, these values were 0.10±0.11, 0.00 (0.00, 0.03), and 0.01 (0.00, 0.06), respectively. Objective scatter indices postoperatively were 2.20±1.82 for the cataract surgery eye and 0.90 (0.48, 1.90) for the ICL implantation eye. All patients became spectacle-independent postoperatively, with 6 experiencing halo phenomena that did not affect their daily life. The average patient satisfaction score was 9.5. The ICL implantation eye showed better accommodative amplitude and facility than the cataract surgery eye, with statistically significant differences (both P<0.05). The defocus curve indicated good binocular visual acuity, and no severe intraoperative or postoperative complications were reported. Conclusions: For patients with high myopia and monocular cataract, the combination of cataract surgery and ICL implantation in the contralateral eye effectively enhances postoperative visual acuity and quality, preserves natural accommodation, establishes binocular vision balance, and facilitates spectacle independence. The procedure is safe, reliable, and improves the quality of life and work.

Insect Ocelli: Ecology, Physiology, and Morphology of the Accessory Visual System

Insect Ocelli: Ecology, Physiology, and Morphology of the Accessory Visual System

Clinical Physiology of the Central Parts of the Visual System

Clinical Physiology of the Central Parts of the Visual System

New Plum Detection in Complex Environments Based on Improved YOLOv8n

To address the challenge of accurately detecting new plums amidst trunk and leaf occlusion and fruit overlap, this study presents a novel target detection model, YOLOv8n-CRS. A specialized dataset, specifically designed for new plums, was created under real orchard conditions, with the advanced YOLOv8n model serving as the base network. Initially, the CA attention mechanism was introduced to the backbone network to improve the model’s ability to extract crucial features of new plums. Subsequently, the RFB module was incorporated into the neck layer to leverage multiscale information, mitigating inaccuracies caused by fruit overlap and thereby enhancing detection performance. Finally, the original CIOU loss function was replaced with the SIOU loss function to further enhance the model’s detection accuracy. Test results show that the YOLOv8n-CRS model achieved a recall rate of 88.9%, with average precision scores of mAP@0.5 and mAP@0.5:0.95 recorded at 96.1% and 87.1%, respectively. The model’s F1 score reached 90.0%, and it delivered a real-time detection speed of 88.5 frames per second. Compared to the YOLOv8n model, the YOLOv8n-CRS exhibited a 2.2-percentage-point improvement in recall rate, alongside increases of 0.7 percentage points and 1.2 percentage points in mAP@0.5 and mAP@0.5:0.95, respectively. In comparison to the Faster R-CNN, YOLOv4, YOLOv5s, and YOLOv7 models, the YOLOv8n-CRS model features the smallest size of 6.9 MB. This streamlined design meets the demands for real-time identification of new plums in intricate orchard settings, providing strong technical backing for the visual perception systems of advanced plum-picking robots.

A Declarative Modeling Framework for Intuitive Multiple Criteria Decision Analysis in a Visual Semantic Urban Planning Environment

The magnitude and multiformity of recorded and readily available urban data from independent sensors and mobile user devices have paved new possibilities for city environment planning and evolution. In previous works, a visual semantic decision support system for urban planning was presented, and the capability of machine learning approaches, ranging from random forests to graph-based convolutional neural networks, to infer preferable directions for future development was explored, extrapolating upon previous opted locations and selected alternatives for publicly commercial services. In this work, the anterior decision-making process leading to establishment choices is addressed with the same sets of criteria and samples within the same environment. A Declarative Modeling shell is proposed, and Multiple Criteria Decision Analysis processes are adopted to encompass the DM’s/DMs’ rationale, solely relying on methodologies close to human intuition. To this end, outranking representatives from the PROMETHEE family as well as weighted sum approaches are employed, fueled by the interpretation of the declarative description of decision parameters on behalf of the DM(s), exploring the ability to achieve classifications in a straight synthetic manner, i.e., in the absence of previous knowledge, thus exhibiting the potential of decision analysis methodologies, enhanced by Declarative Modeling, to be used as powerful intuitive tools in similar paradigm contexts.