Search Results (3,315)

Infrared and visible image fusion (IVIF) fully integrates the complementary features of different modal images, and the fused image provides a more comprehensive and objective interpretation of the scene compared to each source image, thus attracting extensive attention in the field of computer vision in recent years. However, current fusion methods usually center their attention on the extraction of prominent features, falling short of adequately safeguarding subtle and diminutive structures. To address this problem, we propose an end-to-end unsupervised IVIF method (ESFuse), which effectively enhances fine edges and small structures. In particular, we introduce a two-branch head interpreter to extract features from source images of different modalities. Subsequently, these features are fed into the edge refinement module with the detail injection module (DIM) to obtain the edge detection results of the source image, improving the network’s ability to capture and retain complex details as well as global information. Finally, we implemented a multiscale feature reconstruction module to obtain the final fusion results by combining the output of the DIM with the output of the head interpreter. Extensive IVIF fusion experiments on existing publicly available datasets show that the proposed ESFuse outperforms the state-of-the-art(SOTA) methods in both subjective vision and objective evaluation, and our fusion results perform well in semantic segmentation, target detection, pose estimation and depth estimation tasks. The source code has been availabled. Full article

Petunia stands as the top-selling bedding plant in the U.S., and improved lighting control in greenhouses holds the potential to reduce crop production time and optimize crop quality. This study investigated the impact of four distinct daily light integral (DLI) conditions with and without supplemental far-red (FR) radiation on the growth of petunia liners and subsequent development of finish plants. Two experiments were conducted in spring (9 April to 18 June 2021) and winter (28 October 2021 to 6 January 2022). Petunia cuttings were rooted in a common environment and then transferred to four greenhouse sections with different DLI treatments: 6, 9, 12, and 15 mol·m⁻²·d⁻¹ for four weeks. Within each DLI condition, half of the plants were exposed to 28 μmol·m⁻²·s⁻¹ supplemental FR radiation for 16 h daily (equivalent to 1.61 mol·m⁻²·d⁻¹ light integral). The number of flower buds and open flowers were tracked daily. Representative liners were destructively harvested and evaluated after four weeks of lighting treatments. The remaining plants were transplanted and moved to a common DLI condition of 15 mol·m⁻²·d⁻¹ for an additional three weeks before being destructively harvested and evaluated as finish plants. The primary finding reveals the promoting effect of DLI on flowering, branching, morphology, and biomass accumulation of petunia liners, with many effects persisting into the finish stage. A threshold DLI of 9 mol·m⁻²·d⁻¹ was identified, as lower DLI (6 mol·m⁻²·d⁻¹) resulted in extensive stem elongation, rendering the plants unmarketable. Higher DLI levels were found to be optimal in terms of flowering and morphology. Supplemental FR accelerated flowering by up to three days in the summer experiment and up to 12 days in the winter experiment. However, FR had limited impact on the number of flower buds and open flowers, branching, and shoot and root weight of the finish plants. Interactions between DLI and FR were observed on some parameters, whereby FR effects were more pronounced under lower DLI. Overall, both higher DLI and supplemental FR exhibited beneficial effects, but DLI had a more pronounced effect. Thus, DLI during petunia liner production appears more important than adding FR. This study well simulated the commercial propagation and production of petunia plants, providing practical insights for decision-making regarding lighting strategies. Full article

Background/Objectives: The tibial nerve, commonly misnamed the “posterior tibial nerve”, branches into four key nerves: the medial plantar, lateral plantar, inferior calcaneal (Baxter’s nerve), and medial calcaneal branches. These nerves are integral to both the sensory and motor functions of the foot. Approximately 15% of adults with foot issues experience heel pain, frequently stemming from neural origins, such as tarsal tunnel syndrome (TTS). TTS diagnosis remains challenging due to a high false negative rate in neurophysiological studies. This study aims to improve the understanding and diagnosis of distal tarsal tunnel pathology to enable more effective treatments, including platelet-rich plasma, hydrodissections, radiofrequencies, and prolotherapy. Methods: Ultrasound-guided techniques were employed to examine the distal tarsal tunnel using the Heimkes triangle for optimal probe placement. Results: The results indicate that the tunnel consists of two chambers separated by the interfascicular septum, housing the medial, lateral plantar, and inferior calcaneal nerves. Successful interventions depend on precise visualization and patient positioning. This study emphasizes the importance of avoiding the calcaneus periosteum to reduce discomfort. Conclusions: Standardizing nerve involvement classification in TTS is difficult without robust neurophysiological studies. The accurate targeting of nerve branches is essential for effective treatment. Full article

Evasive Spectre attacks have used additional nop or memory delay instructions to make effective hardware performance counter based detectors with lower attack detection successful rate. Interestingly, the detection performance gets worse under different workloads. For example, the attack detection successful rate is only 59.8% for realistic applications, while it is much lower 27.52% for memory stress test. Therefore, this paper proposes a two-stage smart detector T-Smade designed for evasive Spectre attacks (e.g., evasive Spectre nop and evasive Spectre memory) under various workloads. T-Smade uses the first-stage detector to identify the type of workloads and then selects the appropriate second-stage detector, which uses four hardware performance counter events to characterize the high cache miss rate and low branch miss rate of Spectre attacks. More importantly, the second stage detector adds one dimension of reusing cache miss rate and branch miss rate to exploit the characteristics of various workloads to detect evasive Spectre attacks effectively. Furthermore, to achieve the good generalization for more unseen evasive Spectre attacks, the proposed classification detector T-Smade is trained by the raw data of Spectre attacks and non-attacks in different workloads using simple Multi-Layer Perception models. The comprehensive results demonstrate that T-Smade makes the average attack detection successful rate of evasive Spectre nop under different workload return from 27.52% to 95.42%, and that of evasive Spectre memory from 59.8% up to 100%. Full article

A convolutional neural network (CNN) has been developed to forecast the occurrence of low-visibility events or hazes in the Paris area. It has been trained and validated using multi-decadal daily regional maps of many meteorological and hydrological variables alongside surface visibility observations. The strategy is to make the machine learn from available historical data to recognize various regional weather and hydrological regimes associated with low-visibility events. To better preserve the characteristic spatial information of input features in training, two branched architectures have recently been developed. These architectures process input features firstly through several branched CNNs with different kernel sizes to better preserve patterns with certain characteristic spatial scales. The outputs from the first part of the network are then processed by the second part, a deep non-branched CNN, to further deliver predictions. The CNNs with new architectures have been trained using data from 1975 to 2019 in a two-class (haze versus non-haze) classification mode as well as a regression mode that directly predicts the value of surface visibility. The predictions of regression have also been used to perform the two-class classification forecast using the same definition in the classification mode. This latter procedure is found to deliver a much better performance in making class-based forecasts than the direct classification machine does, primarily by reducing false alarm predictions. The branched architectures have improved the performance of the networks in the validation and also in an evaluation using the data from 2021 to 2023 that have not been used in the training and validation. Specifically, in the latter evaluation, branched machines captured 70% of the observed low-visibility events during the three-year period at Charles de Gaulle Airport. Among those predicted low-visibility events by the machines, 74% of them are true cases based on observation. Full article

Rice spikelet size, spikelet length and spikelet width, are very important traits directly related to a rice crop’s yield. The accurate measurement of these parameters is quite significant in research such as breeding, yield evaluation and variety improvement for rice crops. Traditional measurement methods still mainly rely on manual labor, which is time-consuming, labor-intensive and error-prone. In this study, a novel method, dubbed the “SSM-Method”, based on convolutional neural network and traditional image processing technology has been developed for the efficient and precise measurement of rice spikelet size parameters on rice panicle structures. Firstly, primary branch images of rice panicles were collected at the same height to build an image database. The spikelet detection model using convolutional neural network was then established for spikelet recognition and localization. Subsequently, the calibration value was obtained through traditional image processing technology. Finally, the “SSM-Method” integrated with a spikelet detection model and calibration value was developed for the automatic measurement of spikelet sizes. The performance of the developed SSM-Method was evaluated through testing 60 primary branch images. The test results showed that the root mean square error (RMSE) of spikelet length for two rice varieties (Huahang15 and Qingyang) were 0.26 mm and 0.30 mm, respectively, while the corresponding RMSE of spikelet width was 0.27 mm and 0.31 mm, respectively. The proposed algorithm can provide an effective, convenient and low-cost tool for yield evaluation and breeding research. Full article

Using multi-view images of forest plots to reconstruct dense point clouds and extract individual tree parameters enables rapid, high-precision, and cost-effective forest plot surveys. However, images captured at close range face challenges in forest reconstruction, such as unclear canopy reconstruction, prolonged reconstruction times, insufficient accuracy, and issues with tree duplication. To address these challenges, this paper introduces a new image dataset creation process that enhances both the efficiency and quality of image acquisition. Additionally, a block-matching-based multi-view reconstruction algorithm, Forest Multi-View Reconstruction with Enhanced Confidence-Guided Dynamic Domain Propagation (CDP-MVS), is proposed. The CDP-MVS algorithm addresses the issue of canopy and sky mixing in reconstructed point clouds by segmenting the sky in the depth maps and setting its depth value to zero. Furthermore, the algorithm introduces a confidence calculation method that comprehensively evaluates multiple aspects. Moreover, CDP-MVS employs a decentralized dynamic domain propagation sampling strategy, guiding the propagation of the dynamic domain through newly defined confidence measures. Finally, this paper compares the reconstruction results and individual tree parameters of the CDP-MVS, ACMMP, and PatchMatchNet algorithms using self-collected data. Visualization results show that, compared to the other two algorithms, CDP-MVS produces the least sky noise in tree reconstructions, with the clearest and most detailed canopy branches and trunk sections. In terms of parameter metrics, CDP-MVS achieved 100% accuracy in reconstructing tree quantities across the four plots, effectively avoiding tree duplication. The accuracy of breast diameter extraction values of point clouds reconstructed by CDPMVS reached 96.27%, 90%, 90.64%, and 93.62%, respectively, in the four sample plots. The positional deviation of reconstructed trees, compared to ACMMP, was reduced by 0.37 m, 0.07 m, 0.18 m and 0.33 m, with the average distance deviation across the four plots converging within 0.25 m. In terms of reconstruction efficiency, CDP-MVS completed the reconstruction of the four plots in 1.8 to 3.1 h, reducing the average reconstruction time per plot by six minutes compared to ACMMP and by two to three times compared to PatchMatchNet. Finally, the differences in tree height accuracy among the point clouds reconstructed by the different algorithms were minimal. The experimental results demonstrate that CDP-MVS, as a multi-view reconstruction algorithm tailored for forest reconstruction, shows promising application potential and can provide valuable support for forestry surveys. Full article

In the transition to fully automated driving, the interaction between drivers and vehicles is crucial as drivers’ emotions directly influence their behavior, thereby impacting traffic safety. Currently, relying solely on a backbone based on a convolutional neural network (CNN) to extract single RGB modal facial features makes it difficult to capture enough semantic information. To address this issue, this paper proposes a Dual-branch Multimodal Fusion Network (DMFNet). DMFNet extracts semantic features from visible–infrared (RGB-IR) image pairs effectively capturing complementary information between two modalities and achieving a more accurate understanding of the drivers’ emotional state at a global level. However, the accuracy of facial recognition is significantly affected by variations in the drivers’ head posture and light environment. Thus, we further propose a U-Shape Reconstruction Network (URNet) to focus on enhancing and reconstructing the detailed features of RGB modes. Additionally, we design a Detail Enhancement Block (DEB) embedded in a U-shaped reconstruction network for high-frequency filtering. Compared with the original driver emotion recognition model, our method improved the accuracy by 18.77% on the DEFE++ dataset, proving the superiority of the proposed method. Full article

All 32 Brazilian species of Parodia Speg (Cactaceae) occurring in Rio Grande do Sul State are considered threatened, according to the IUCN criteria. Until 2021, Parodia rechensis (CR) was known by only two small populations. However, a new population with over 400 individuals was discovered in 2021, prompting the study of its reproductive biology as a way to promote its conservation. Anthesis, breeding system, and natural pollination were studied in the field. The breeding system was studied by applying controlled pollination treatments to plants excluded from pollinators (bagged). Germination features were studied at the Seed Bank of the Porto Alegre Botanical Garden under controlled temperatures (20, 25, and 30 °C). The anthesis is diurnal and lasts for up to four days. The flowers offer pollen as the sole resource to the pollinators. The study species is unable to set fruit and seed without the agency of pollinators and has self-incompatible (unable to set fruit and seeds when pollinated with pollen of the same individual) characteristics that can considerably restrict its reproduction. Native bees of Halictidae and Apidae (Hymenoptera) are the main pollinators, with a smaller contribution of Melyridae (Coleoptera) and Syrphidae (Diptera). Natural fruit set is moderate (≤64%, per individual), but the species presents vegetative growth, producing several branches from the mother plant. Seeds showed the optimum germination rate at 20 °C and an inhibition of 75% in germinability at 30 °C. Our findings suggest the need to manage the species’ habitat to guarantee the permanency of the plants and healthy populations of pollinators as well. Our findings raise concerns about the germination and establishment of new individuals in the context of rising temperatures caused by climate change. Suggestions for the possible management of the extant populations are made. Full article

Waxy corn is a special type of maize primarily consumed as a fresh vegetable by humans. Nitrogen (N) plays an essential role in regulating the growth progression, maturation, yield, and quality of waxy maize. A reasonable N application rate is vital for boosting the accumulation of both N and carbon (C) in the grains, thereby synergistically enhancing the grain quality. However, the impact of varying N levels on the dynamic changes in N metabolism, carbohydrate formation, and anthocyanin synthesis in purple waxy corn kernels, as well as the regulatory relationships among these processes, remains unclear. To explore the effects of varying N application rates on the N metabolism, carbohydrate formation, and anthocyanin synthesis in kernels during grain filling, a two-year field experiment was carried out using the purple waxy maize variety Jinnuo20 (JN20). This study examined the different N levels, specifically 0 (N0), 120 (N1), 240 (N2), and 360 (N3) kg N ha⁻¹. The results of the analysis revealed that, for nearly all traits measured, the N application rate of N2 was the most suitable. Compared to the N0 treatment, the accumulation and content of anthocyanins, total nitrogen, soluble sugars, amylopectin, and C/N ratio in grains increased by an average of 35.62%, 11.49%, 12.84%, 23.74%, 13.00%, and 1.87% under N2 treatment over five filling stages within two years, respectively, while the harmful compound nitrite content only increased by an average of 30.2%. Correspondingly, the activities of related enzymes also significantly increased and were maintained under N2 treatment compared to N0 treatment. Regression and correlation analysis results revealed that the amount of anthocyanin accumulation was highly positively correlated with the activities of phenylalanine ammonia-lyase (PAL) and flavanone 3-hydroxylase (F3H), but negatively correlated with anthocyanidin synthase (ANS) and UDP-glycose: flavonoid-3-O-glycosyltransferase (UFGT) activity, nitrate reductase (NR), and glutamine synthetase (GS) showed significant positive correlations with the total nitrogen content and lysine content, and a significant negative correlation with nitrite, while soluble sugars were negatively with ADP-glucose pyrophosphorylase (AGPase) activity, and amylopectin content was positively correlated with the activities of soluble starch synthase (SSS), starch branching enzyme (SBE), and starch debranching enzyme (SDBE), respectively. Furthermore, there were positive or negative correlations among the detected traits. Hence, a reasonable N application rate improves purple waxy corn kernel nutritional quality by regulating N metabolism, as well as carbohydrate and anthocyanin biosynthesis. Full article

The access of a high proportion of photovoltaic (PV) will change the energy structure of the distribution network (DN), resulting in a series of safety operation risks. This paper proposes a two-stage PV hosting capacity (PVHC) calculation model to assess the maximum PVHC, considering the uncertainty and active management (AM). Firstly, we employ a robust optimization model to characterize the uncertainty of sources and loads in DN with PV and analyze the worst-case scenarios for PVHC. Subsequently, we construct a PVHC calculation model that takes into account AM, and convert the model into a mixed-integer second-order cone (MISOC) model using linearization techniques. Finally, we apply “heuristic optimization + CPLEX solver” to solve the model and introduce overvoltage and overcurrent indices to analyze the safety of the DN under PV limit access. Case studies are carried out on the IEEE 33-bus system and a practical case. Results show that (1) only the uncertainty that reduces the load or increases the output efficiency will affect PVHC; (2) for DN limited by overvoltage, AM can better improve PVHC; however, for DN limited by maximum transmission power, the effect of AM is low; (3) for most DN, SVC can improve PVHC, but the effect is modest. And network reconfiguration can significantly increase PVHC on the system with poor branch network, even reaching 150% of the original PVHC. Full article

The field of multi-source remote sensing observation is becoming increasingly dynamic through the integration of various remote sensing data sources. However, existing deep learning methods face challenges in differentiating between internal and external relationships and capturing fine spatial features. These models often struggle to effectively capture comprehensive information across remote sensing data bands, and they have inherent differences in the size, structure, and physical properties of different remote sensing datasets. To address these challenges, this paper proposes a novel geometric-algebra-based spectral–spatial hierarchical fusion network (GASSF-Net), which uses geometric algebra for the first time to process multi-source remote sensing images, enabling a more holistic approach to handling these images by simultaneously leveraging the real and imaginary components of geometric algebra to express structural information. This method captures the internal and external relationships between remote sensing image features and spatial information, effectively fusing the features of different remote sensing data to improve classification accuracy. GASSF-Net uses geometric algebra (GA) to represent pixels from different bands as multivectors, thus capturing the intrinsic relationships between spectral bands while preserving spatial information. The network begins by deeply mining the spectral–spatial features of a hyperspectral image (HSI) using pairwise covariance operators. These features are then extracted through two branches: a geometric-algebra-based branch and a real-valued network branch. Additionally, the geometric-algebra-based network extracts spatial information from light detection and ranging (LiDAR) to complement the elevation data lacking in the HSI. Finally, a genetic-algorithm-based cross-fusion module is introduced to fuse the HSI and LiDAR data for improved classification. Experiments conducted on three well-known datasets, Trento, MUUFL, and Houston, demonstrate that GASSF-Net significantly outperforms traditional methods in terms of classification accuracy and model efficiency. Full article

Rolling bearing is the core component of industrial machines, but it is difficult for common single signal source-based fault diagnosis methods to ensure reliable results since sensor signals are vulnerable to the pollution of background noises and the attenuation of transmitted information. Recently, multi-source information-based fault diagnosis methods have become popular, but the information redundancy between multiple signals is a tough problem that will negatively impact the representational capacity of deep learning algorithms and the precision of fault diagnosis methods. Besides that, the characteristics of various signals are actually different, but this problem was usually omitted by researchers, and it has potential to further improve the diagnosing performance by adaptively adjusting the feature extraction process for every input signal source. Aimed at solving the above problems, a novel model for bearing fault diagnosis called multi-branch selective fusion deep residual network is proposed in this paper. The model adopts a multi-branch structure design to enable every input signal source to have a unique feature processing channel, avoiding the information of multiple signal sources blindly coupled by convolution kernels. And in each branch, different convolution kernel sizes are assigned according to the characteristics of every input signal, fully digging the precious fault components on respective information sources. Lastly, the dropout technique is used to randomly throw out some activated neurons, alleviating the redundancy and enhancing the quality of the multiscale features extracted from different signals. The proposed method was experimentally compared with other intelligent methods on two authoritative public bearing datasets, and the experimental results prove the feasibility and superiority of the proposed model. Full article

In recent years, parcel volumes have reached record highs, prompting the logistics industry to explore innovative solutions to meet growing demand. In densely populated areas, delivery robots offer a promising alternative to traditional truck-based delivery systems. These autonomous electric robots operate on sidewalks and deliver time-sensitive goods, such as express parcels, medicine and meals. However, their limited cargo capacity and battery life require a return to a depot after each delivery. This challenge can be modeled as an electric vehicle-routing problem with soft time windows and single-unit capacity constraints. The objective is to serve all customers while minimizing the quadratic sum of delivery delays and ensuring each vehicle operates within its battery limitations. To address this problem, we propose a mixed-integer quadratic programming model and introduce an enhanced formulation using a layered graph structure. For this layered graph, we present two solution approaches based on relaxations that reduce the number of nodes and arcs compared to the expanded formulation. The first approach, Iterative Refinement, solves the current relaxation to optimality and refines the graph when the solution is infeasible for the expanded formulation. This process continues until a proven optimal solution is obtained. The second approach, Branch and Refine, integrates graph refinement into a branch-and-bound framework, eliminating the need for restarts. Computational experiments on modified Solomon instances demonstrate the effectiveness of our solution approaches, with Branch and Refine consistently outperforming Iterative Refinement across all tested parameter configurations. Full article

Glass is one of the most common materials in society, and the float glass process is the main production method of glass used at present, which involves adopting a melting furnace with a single cooler. However, this structure has been difficult to fit to the requirements of modern glass production, such as producing multiple types of glass and large-scale production. Therefore, a large-tonnage float glass melting furnace with a double cooler is studied, which is rising in popularity in the glass sector. The aim of this paper is to clarify the characteristics of the new glass furnace. A numerical simulation technique is applied to analyze the thermal and flow characteristics of molten glass in the new structure so as to clarify the feasibility of production by checking the temperature distribution and flow field of the molten glass. The results show that the new structure also exhibits flow behavior similar to the original structure in the branch line. Due to the addition of the branch line, the stability of the temperature is improved, with a 60 K and 43 K difference between the surface and bottom in the main and branch lines, respectively. Similar stability is shown in the flow field, specifically low acceleration in the cooler (0.006 m/s²). The bubble clarification time is about 2700 s, less than the 3000 s required for flow. The parameters of the branch line meet the requirements of glass production. In theory, a glass-melting furnace with a double cooler has the capacity to produce two types of glass. Full article