Search Results (201)

Despite advancements in deep learning-based Monocular Depth Estimation (MDE), applying these models to video sequences remains challenging due to geometric ambiguities in texture-less regions and temporal instability caused by independent per-frame inference. To address these limitations, we propose STF-Depth, a novel post-processing framework that enhances depth quality by logically fusing heterogeneous information—geometric, semantic, and panoptic—without requiring additional retraining. Our approach introduces a robust RANSAC-based Vanishing Point Estimation to guide Dynamic Depth Gradient Correction for background separation, alongside Adaptive Instance Re-ordering to clarify occlusion relationships. Experimental results on the KITTI, NYU Depth V2, and TartanAir datasets demonstrate that STF-Depth functions as a universal plug-and-play module. Notably, it achieved a 25.7% reduction in Absolute Relative error (AbsRel) and significantly enhanced temporal consistency compared to state-of-the-art backbone models. These findings confirm the framework’s practicality for real-world applications requiring geometric precision and video stability, such as autonomous driving, robotics, and augmented reality (AR). Full article

We propose a composite Lyapunov framework for nonlinear autonomous systems that ensures strict decay through a pair of differential inequalities. The approach yields integral estimates, quantitative convergence rates, vanishing of dissipation measures, convergence to a critical set, and semistability under mild conditions, without relying on invariance principles or compactness assumptions. The framework unifies convergence to points and sets and is illustrated through applications to inertial gradient systems and Primal–Dual gradient flows. Full article

A gap test is a relatively new experimental and numerical test proposed by Bažant and co-workers. The test is designed to show that the effective mode I (opening mode) fracture energy depends strongly on the crack-parallel normal stress. Other typical tests (for instance three-point bending) are not able to capture this influence. The gap test is therefore considered as a breakthrough in fracture mechanics. The authors of this paper tried to recreate numerically the gap test in a concrete specimen using a plasticity-based damage model for concrete, the so-called concrete damaged plasticity in Abaqus software. A specimen in the gap test is in the form of a concrete beam with a notch. At the first stage of the test the specimen is supported with poly-propylene pads. A static scheme of the specimen changes when the gap between the specimen and roller supports vanishes, exactly when the full yield of pads occurs. The gap test revealed the fact that the initial fracture energy varies its value depending on normal stress parallel to a crack propagating from the notch. To calculate the initial fracture energy, the authors of this paper used the size effect law. The results obtained by the authors are comparable with the laboratory and numerical results presented by Nguyen and co-workers. Full article

A Multi-Layer Perceptron (MLP), as the basic structure of neural networks, is an important component of various deep learning models such as CNNs, RNNs, and Transformers. Nevertheless, MLP training faces significant challenges, with a large number of saddle points and local minima in its non-convex optimization space, which can easily lead to gradient vanishing and premature convergence. Compared with traditional heuristic algorithms relying on a population-based parallel search, such as GA, GWO, DE, etc., the Besiege and Conquer Algorithm (BCA) employs a one-spot update strategy that provides a certain level of global optimization capability but exhibits clear limitations in search flexibility. Specifically, it lacks fast detection, fast adaptation, and fast convergence. First, the fixed sinusoidal amplitude limits the accuracy of fast detection in complex regions. Second, the combination of a random location and fixed perturbation range limits the fast adaptation of global convergence. Finally, the lack of a hierarchical adjustment under a single parameter (BCB) hinders the dynamic transition from exploration to exploitation, resulting in slow convergence. To address these limitations, this paper proposes a Flexible Besiege and Conquer Algorithm (FBCA), which improves search flexibility and convergence capability through three new mechanisms: (1) the sine-guided soft asymmetric Gaussian perturbation mechanism enhances local micro-exploration, thereby achieving a fast detection response near the global optimum; (2) the exponentially modulated spiral perturbation mechanism adopts an exponential spiral factor for fast adaptation of global convergence; and (3) the nonlinear cognitive coefficient-driven velocity update mechanism improves the convergence performance, realizing a more balanced exploration–exploitation process. In the IEEE CEC 2017 benchmark function test, FBCA ranked first in the comprehensive comparison with 12 state-of-the-art algorithms, with a win rate of 62% over BCA in 100-dimensional problems. It also achieved the best performance in six MLP optimization problems, showing excellent convergence accuracy and robustness, proving its excellent global optimization ability in complex nonlinear MLP optimization training. It demonstrates its application value and potential in optimizing neural networks and deep learning models. Full article

The purpose of this research is to develop approximate weak and strong stationary conditions for interval-valued multiobjective optimization problems with vanishing constraints (IVMOPVC) involving nonsmooth functions. In many real-world situations, the exact values of objectives are uncertain or imprecise; hence, interval-valued formulations are used to model such uncertainty more effectively. The proposed approximate weak and strong stationarity conditions provide a robust framework for deriving meaningful optimality results even when the usual constraint and data qualifications fail. We first introduce approximate variants of these qualifications and establish their relationships. Secondly, we establish some approximate KKT type necessary optimality conditions in terms of approximate weak strongly stationary points and approximate strong strongly stationary points to identify type-2 $\mathcal{E}$-quasi weakly Pareto and type-1 $\mathcal{E}$-quasi Pareto solutions of the IVMOPVC. Lastly, we show that the approximate weak and strong strongly stationary conditions are sufficient for optimality under some approximate convexity assumptions. All the outcomes are well illustrated by examples. Full article

The complex structures and diverse object categories in indoor environments pose significant challenges for point cloud semantic segmentation. To address the insufficient capability of extracting local features in complex scenes, this paper proposes a point cloud segmentation network based on neighborhood feature enhancement termed PKA-Net. First, to obtain richer and more discriminative feature representations, we design a local feature encoding module that extracts geometric features, color information, and spatial information from local regions of the point cloud for joint feature encoding. Furthermore, we enhance the hierarchical feature extraction by integrating Kolmogorov–Arnold Networks (KAN) to form the SAPK module, improving the network’s ability to fit complex geometric structures. A residual structure is also adopted to optimize feature propagation and alleviate the problem of gradient vanishing. Finally, we propose the dual attention mechanism C-MSCA, which dynamically selects and strengthens key features through the synergistic action of channel and spatial attention, enhancing the network’s perception of local details and global structure. To evaluate the performance of the proposed PKA-Net, extensive experiments were conducted on the S3DIS dataset. Experimental results demonstrate that PKA-Net improves OA by 2.1%, mAcc by 2.9%, and mIoU by 4% compared to the baseline model. It outperforms other mainstream models, delivering enhanced overall segmentation performance. Full article

The paper investigates the regularization of solutions to nonlinear Volterra integral equations of the first kind, under the assumption that a solution exists and belongs to the space of continuous functions. The kernel of the equation is a differentiable function and vanishes on the diagonal at an interior point of the integration interval. By applying an appropriate differential operator (with respect to x), the Volterra integral equation of the first kind is reduced to a Volterra integral equation of the third kind, equivalent with respect to solvability. The subdomain method is employed by partitioning the integration interval into two subintervals. Within the imposed constraints, a compatibility condition for the solutions is satisfied at the junction point of the partial subintervals. A Lavrentiev-type regularizing operator is constructed that preserves the Volterra structure of the equation. The uniform convergence of the regularized solution to the exact solution is proved, and conditions ensuring the uniqueness of the solution in Hölder space are established. Full article

Aging water pipelines are increasingly prone to structural failure, leakage, and ground subsidence, creating critical risks to urban infrastructure. Closed-circuit television endoscopy is widely used for internal assessment, but it depends on manual interpretation and lacks reliable quantitative defect information. Traditional vanishing point detection techniques, such as the Hough Transform, often fail under practical conditions due to irregular lighting, debris, and deformed pipe surfaces, especially when pipes are water-filled. To overcome these challenges, this study introduces a deep learning-based method that estimates inverse projection parameters from monocular endoscopic images. The proposed approach reconstructs a spatially accurate two-dimensional projection of the pipe interior from a single frame, enabling defect quantification for cracks, scaling, and delamination. This eliminates the need for stereo cameras or additional sensors, providing a robust and cost-effective solution compatible with existing inspection systems. By integrating convolutional neural networks with geometric projection estimation, the framework advances computational intelligence applications in pipeline condition monitoring. Experimental validation demonstrates high accuracy in pose estimation and defect size recovery, confirming the potential of the system for automated, non-disruptive pipeline health evaluation. Full article

Single-image based camera auto-calibration holds significant value for improving perception efficiency in traffic surveillance systems. However, existing approaches face dual challenges: scarcity of real-world datasets and poor adaptability to multi-view scenarios. This paper presents a systematic solution framework. First, we constructed a large-scale synthetic dataset containing 36 highway scenarios using the CARLA 0.9.15 simulation engine, generating approximately 336,000 virtual frames with precise calibration parameters. The dataset achieves statistical consistency with real-world scenes by incorporating diverse view distributions, complex weather conditions, and varied road geometries. Second, we developed DeepCalib, a deep calibration network that explicitly models perspective projection features through the triplet attention mechanism. This network simultaneously achieves road direction vanishing point localization and camera pose estimation using only a single image. Finally, we adopted a progressive learning paradigm: robust pre-training on synthetic data establishes universal feature representations in the first stage, followed by fine-tuning on real-world datasets in the second stage to enhance practical adaptability. Experimental results indicate that DeepCalib attains an average calibration precision of 89.6%. Compared to conventional multi-stage algorithms, our method achieves a single-frame processing speed of 10 frames per second, showing robust adaptability to dynamic calibration tasks across diverse surveillance views. Full article

Against the global backdrop of energy transition, the precise assessment of urban rooftop photovoltaic (PV) system capacity is recognized as crucial for optimizing the energy structure and enhancing the sustainable utilization efficiency of spatial resources. Publicly available aerial imagery is characterized by non-orthorectified issues; direct utilization is known to lead to geometric distortions in rooftop PV and errors in capacity prediction. To address this, a dual-optimization framework is proposed in this study, integrating monocular vision-based 3D reconstruction with a lightweight linear model. Leveraging the orthogonal characteristics of building structures, camera self-calibration and 3D reconstruction are achieved through geometric constraints imposed by vanishing points. Scale distortion is suppressed via the incorporation of a multi-dimensional geometric constraint error control strategy. Concurrently, a linear capacity-area model is constructed, thereby simplifying the complexity inherent in traditional multi-parameter fitting. Utilizing drone oblique photography and Google Earth public imagery, 3D reconstruction was performed for 20 PV-equipped buildings in Wuhan City. Two buildings possessing high-precision field survey data were selected as typical experimental subjects for validation. The results demonstrate that the 3D reconstruction method reduced the mean absolute percentage error (MAPE)—used here as an estimator of measurement uncertainty—of PV area identification from 10.58% (achieved by the 2D method) to 3.47%, while the coefficient of determination (R²) for the capacity model reached 0.9548. These results suggest that this methodology can provide effective technical support for low-cost, high-precision urban rooftop PV resource surveys. It has the potential to significantly enhance the reliability of energy planning data, thereby contributing to the efficient development of urban spatial resources and the achievement of sustainable energy transition goals. Full article

With the vigorous development of deep learning technology, lane detection tasks have achieved phased results. However, existing lane detection models do not consider the unique geometric and visual features of lanes when dealing with some challenging scenarios, resulting in many difficulties and limitations. To this end, we propose a lane detection network based on full-dimensional convolutional Transformer (OTVLD-Net) to improve the adaptability of the model under extreme road conditions and better handle complex lane topology. In order to extract richer contextual features, we designed ODVT-Net, which uses full-dimensional dynamic convolution combined with improved feature flip fusion layer and non-local network layer, and aggregates lane symmetry features by utilizing the horizontal symmetry of lanes. A feature weight generation mechanism based on Transformer is designed, and a cross-attention mechanism between feature maps and lane requests is added in the decoding stage to enable the network to aggregate global feature information. At the same time, a vanishing point detection module is introduced, and a joint weighted loss function is designed to be trained in coordination with the lane detection task to improve the generalization ability of the lane detection model. Experimental results on the OpenLane and CurveLanes datasets show that the detection effect of the OTVLD-Net model has reached the current advanced level. In particular, the accuracy on the OpenLane dataset is 6.4% higher than the F1 score of the second-ranked model, and the average performance in different challenging scenarios is also improved by 8.9%. At the same time, when ResNet-18 is used as the template feature extraction network, the model achieves a speed of 103FPS and a computing power of 14.2 GFlops, achieving good performance while ensuring real-time performance. Full article

Convolutional neural networks (CNNs) and their improved models (like DenseNet-121) have achieved significant results in image classification tasks. However, the performance of these models is still constrained by issues such as hyperparameter optimization and gradient vanishing and exploding. Owing to their unique exploration and exploitation capabilities, evolutionary algorithms offer new avenues for addressing these problems. Simultaneously, to prevent these algorithms from falling into a local optimum during the search process, this study designs a novel interpolation algorithm. To achieve better image classification performance, thus enhancing classification accuracy and boosting model stability, this paper utilizes a hybrid algorithm based on the horned lizard algorithm with quadratic interpolation and the giant armadillo optimization with Newton interpolation (HGAO) to optimize the hyperparameters of DenseNet-121. It is applied to five datasets spanning different domains. The learning rate and dropout rate have notable impacts on the outcomes of the DenseNet-121 model, which are chosen as the hyperparameters to be optimized. Experiments are conducted using the HGAO algorithm on five image datasets and compared with nine state-of-the-art algorithms. The performance of the model is evaluated based on accuracy, precision, recall, and F1-score metrics. The experimental results reveal that the combination of hyperparameters becomes more reasonable after optimization with the HGAO algorithm, thus providing a crucial improvement. In the comparative experiments, the accuracy of the image classification on the training set increased by up to 0.5%, with a maximum reduction in loss of 0.018. On the test set, the accuracy rose by 0.5%, and the loss decreased by 54 points. The HGAO algorithm provides an effective solution for optimizing the DenseNet-121 model. The designed method boosts classification accuracy and model stability, which also dramatically augments hyperparameter optimization effects and resolves gradient difficulties. Full article

Early and accurate identification of maize diseases is crucial for ensuring sustainable agricultural development. However, existing maize disease identification models face challenges including high inter-class similarity, intra-class variability, and limited capability in identifying early-stage symptoms. To address these limitations, we proposed DSTANet (decomposed spatial token aggregation network), a lightweight and high-performance model for maize leaf disease identification. In this study, we constructed a comprehensive maize leaf image dataset comprising six common disease types and healthy samples, with early and late stages of northern leaf blight and eyespot specifically differentiated. DSTANet employed MobileViT as the backbone architecture, combining the advantages of CNNs for local feature extraction with transformers for global feature modeling. To enhance lesion localization and mitigate interference from complex field backgrounds, DSFM (decomposed spatial fusion module) was introduced. Additionally, the MSTA (multi-scale token aggregator) was designed to leverage hidden-layer feature channels more effectively, improving information flow and preventing gradient vanishing. Experimental results showed that DSTANet achieved an accuracy of 96.11%, precision of 96.17%, recall of 96.11%, and F1-score of 96.14%. With only 1.9M parameters, 0.6 GFLOPs (floating point operations), and an inference speed of 170 images per second, the model meets real-time deployment requirements on edge devices. This study provided a novel and practical approach for fine-grained and early-stage maize disease identification, offering technical support for smart agriculture and precision crop management. Full article

Efficient autonomous exploration in unknown obstacle cluttered environments with interior obstacles remains a challenging task for mobile robots. In this work, we present a novel exploration process for a non-holonomic agent exploring 2D spaces using onboard LiDAR sensing. The proposed method generates velocity commands based on the calculation of the solution of an elliptic Partial Differential Equation with Dirichlet boundary conditions. While solving Laplace’s equation yields collision-free motion towards the free space boundary, the agent may become trapped in regions distant from free frontiers, where the potential field becomes almost flat, and consequently the agent’s velocity nullifies as the gradient vanishes. To address this, we solve a Poisson equation, introducing a source point on the free explored boundary which is located at the closest point from the agent and attracts it towards unexplored regions. The source values are determined by an exponential function based on the shortest path of a Hybrid Visibility Graph, a graph that models the explored space and connects obstacle regions via minimum-length edges. The computational process we apply is based on the Walking on Sphere algorithm, a method that employs Brownian motion and Monte Carlo Integration and ensures efficient calculation. We validate the approach using a real-world platform; an AmigoBot equipped with a LiDAR sensor, controlled via a ROS-MATLAB interface. Experimental results demonstrate that the proposed method provides smooth and deadlock-free navigation in complex, cluttered environments, highlighting its potential for robust autonomous exploration in unknown indoor spaces. Full article

Regarding the distortion correction problem of large field of view wide-angle cameras commonly used in railway visual inspection systems, this paper proposes a novel online calibration method for non-specially made cooperative calibration objects. Based on the radial distortion divisor model, first, the spatial coordinates of natural spatial landmark points are constructed according to the known track gauge value between two parallel rails and the spacing value between sleepers. By using the image coordinate relationships corresponding to these spatial coordinates, the coordinates of the distortion center point are solved according to the radial distortion fundamental matrix. Then, a constraint equation is constructed based on the collinear constraint of vanishing points in railway images, and the Levenberg–Marquardt algorithm is used to found the radial distortion coefficients. Moreover, the distortion coefficients and the coordinates of the distortion center are re-optimized according to the least squares method (LSM) between points and the fitted straight line. Finally, based on the above, the distortion correction is carried out for the distorted railway images captured by the camera. The experimental results show that the above method can efficiently and accurately perform online distortion correction for large field of view wide-angle cameras used in railway inspection without the participation of specially made cooperative calibration objects. The whole method is simple and easy to implement, with high correction accuracy, and is suitable for the rapid distortion correction of camera images in railway online visual inspection. Full article