Map Layers Research Articles

Assessing tropical cyclone risk for improving mitigation strategies in Coastal Odisha, India.

Tropical cyclone causes large-scale devastation and destruction in the coastal plains of India, particularly in Odisha, which is the most cyclone-affected state in the country. Tropical cyclones are projected to be more powerful and widespread due to changing climate. Hence, the risk assessment of tropical cyclone is necessary to identify cyclone-risk areas in coastal Odisha which may help in the mitigation of the damages caused by cyclones. Therefore, this study utilizes geospatial techniques to produce a comprehensive risk map posed by tropical cyclones and to estimate the degree of risk for coastal districts of Odisha. For this, we evaluated the district-level cyclone risk for coastal Odisha using multi-criteria decision-making (MCDM) technique by considering 21 parameters for each of the four components of risk, i.e., exposure, hazard, vulnerability, as well as mitigation capacity. For each criterion, thematic raster map layers were created and weighted using a fuzzy analytical hierarchy process (FAHP). We prepared individual risk component maps using weighted overlay techniques and finally integrated all indices to create the risk map. The study shows that 13% area of the study area comes under a very high-risk zone whereas, 25% area comes under a high-risk zone. The central (Cuttack, northern parts of Khordha, and south-western parts of Jajpur district) and the eastern part (most of the parts of Jagatsinghpur, Kendrapara, and northern parts of Puri district) of the study area come under high to very high tropical cyclone impact zone. Almost 67% of the total area is highly vulnerable to tropical cyclones and mainly concentrated near the shoreline. The applied approach and results can assist the local authorities in identifying vulnerable and hazardous locations and developing workable solutions for the mitigation of revised cyclone threats in the coastal districts of Odisha.

Applying an Analytic Hierarchy Process and a Geographic Information System for Assessment of Land Subsidence Risk Due to Drought: A Case Study in Ca Mau Peninsula, Vietnam

The increase in extreme weather events causes secondary hazards that can influence people and the environment enormously. The Ca Mau Peninsula is known as one of the areas most severely affected by drought, and excessive groundwater exploitation is one of the reasons leading to a higher risk of land subsidence. This study uses the Delphi method and the KAMET rule table to analyze and select indicators that affect subsidence. The study uses the analytic hierarchy process (AHP) analytical hierarchy method to evaluate the weights of influencing factors, combined with geographic information system (GIS) technology to overlay the map layers of the main influencing factors and build a subsidence risk warning zoning map of the study area. The influencing factors selected to evaluate the impact on land subsidence in the study area during the drought period included geological structure, soil characteristics, groundwater flow exploitation, water flow in the dry season, current land use status, and evaporation in the dry season. The weights of these factors were evaluated based on the synthesis of relevant documents as well as consultation with experts. The results indicate that nearly two-thirds of the Ca Mau Peninsula area is currently at very low or low risk of subsidence. Meanwhile, 23% of the area is at medium risk, nearly 9% is at high risk, and 0.1% of the study area is at very high risk. Subsidence risk warning zoning maps can provide a visual and general overview of areas with high subsidence risk, supporting managers in making reference plans for socio-economic development in the Ca Mau Peninsula.

Development of Smart Agriculture to detect the Arabica Coffee Leaf Disease using IAFSA based MSAB with Channel and Spatial Attention Network

Plant diseases provide challenges for the agriculture sector, notably to produce Arabica coffee. Recognising issues on Arabica coffee leaves is a first step in avoiding and curing illnesses to prevent crop loss. With the extraordinary advancements achieved in convolutional neural networks (CNN) in recent years, Arabica coffee leaf damage can now be identified without the aid of a specialist. However, the local characteristics that convolutional layers in CNNs record are typically redundant and unable to make efficient use of global data to support the prediction process. The proposed Hybrid Attention UNet, also known as CMSAMB-UNet due to its feature extraction and global modelling capabilities, integrates both the Channel and Spatial Attention Module (CSAM) as well as the Multi-head Self-Attention Block (MSAB). In this study, CMSAMB-UNet is built on Resnet50 to extract multi-level features from plant picture data. Two shallow layers of feature maps are used with CSAM according to local attention. used throughout the feature extraction process to enrich the features and adaptively disregard unwanted features. In order to recreate the spatial feature connection of the input pictures using high-resolution feature maps, two global attention maps produced by MSAB are combined.

A novel broad learning system integrated with restricted Boltzmann machine and echo state network for time series forecasting

Time series data prediction is crucial in system control, social management, and economic production. For the complex features of time series data and the massive amount of arithmetic in deep learning, a novel network model is proposed based on the broad learning architecture to handle time series prediction tasks. The model utilizes the restricted Boltzmann machine (RBM) in the mapping layer to learn feature information from the input data. Simultaneously, it employs the echo state network (ESN) as the fundamental unit in the enhancement layer to fit the learned feature information from the mapping layer. The proposed model's predictive performance has been validated on air quality index, PM2.5, and electric power datasets. Compared with the benchmark model, it has the advantage of training speed and can reduce prediction error by up to 36%, proving its effectiveness in time series prediction tasks.

Development of a Hydrogen Fuel Cell Prototype Vehicle Supported by Artificial Intelligence for Green Urban Transport

In the automotive sector, the zero emissions area has been dominated by battery electric vehicles. However, prospective users cite charging times, large batteries, and the deployment of charging stations as a counter-argument. Hydrogen will offer a solution to these areas, in the future. This research focuses on the development of a prototype three-wheeled vehicle that is named Neumann H2. It integrates state-of-the-art energy storage systems, demonstrating the benefits of solar-, battery-, and hydrogen-powered drives. Of crucial importance for the R&D platform is the system’s ability to record its internal states in a time-synchronous format, providing valuable data for researchers and developers. Given that the platform is equipped with the ROS2 Open-Source interface, the data are recorded in a standardized format. Energy management is supported by artificial intelligence of the “Reinforcement Learning” type, which selects the optimal energy source for operation based on different layers of high-fidelity maps. In addition to powertrain control, the vehicle also uses artificial intelligence to detect the environment. The vehicle’s environment-sensing system is essentially designed to detect, distinguish, and select environmental elements through image segmentation using camera images and then to provide feedback to the user via displays.

BSI-MVS: multi-view stereo network with bidirectional semantic information

The basic principle of multi-view stereo (MVS) is to perform 3D reconstruction by extracting depth information from multiple views. Most current SOTA MVS networks are based on Vision Transformer, which usually means expensive computational complexity. To reduce computational complexity and improve depth map accuracy, we propose a MVS network with Bidirectional Semantic Information (BSI-MVS). Firstly, we design a Multi-Level Spatial Pyramid module to generate multiple layers of feature map for extracting multi-scale information. Then we propose a 2D Bidirectional-LSTM module to capture bidirectional semantic information at different time steps in the horizontal and vertical directions, which contains abundant depth information. Finally, cost volumes are built based on various levels of feature maps to optimize the final depth map. We experiment on the DTU and BlendedMVS datasets. The result shows that our network, in terms of overall metrics, surpasses TransMVSNet, CasMVSNet, CVP-MVSNet, and AACVP-MVSNet respectively by 17.84%, 36.42%, 14.96%, and 4.86%, which also shows a noticeable performance enhancement in objective metrics and visualizations.

SC-Net: Symmetrical conical network for colorectal pathology image segmentation

Background and ObjectiveImage segmentation of histopathology of colorectal cancer is a core task of computer aided medical image diagnosis system. Existing convolutional neural networks generally extract multi-scale information in linear flow structures by inserting multi-branch modules, which is difficult to extract heterogeneous semantic information under multi-level and different receptive field and tough to establish context dependency among different receptive field features. MethodsTo address these issues, we propose a symmetric spiral progressive feature fusion encoder-decoder network called the Symmetric Conical Network (SC-Net). First, we design a Multi-scale Feature Extraction Block (MFEB) matching with the Symmetric Conical Network to obtain multi-branch heterogeneous semantic information under different receptive fields, so as to enrich the diversity of extracted feature information. The encoder is composed of MFEB through spiral and multi-branch arrangement to enhance context dependence between different information flow. Secondly, the information loss of contour, color and others in high-level semantic information through causally stacking MFEB, the Feature Mapping Layer (FML) is designed to map low-level features to high-level semantic features along the down-sampling branch and solve the problem of insufficient global feature extraction in deep levels. ResultsThe SC-Net was evaluated on our self-constructed colorectal cancer dataset, a publicly available breast cancer dataset and a polyp dataset. The results revealed that the mDice of segmentation reached 0.8611, 0.7259 and 0.7144. We compare our model with the state-of-art semantic segmentation UNet++, PSPNet, Attention U-Net, R2U-Net and other advanced segmentation networks. The experimental results demonstrate that we achieve the most advanced performance. ConclusionsThe results indicate that the proposed SC-Net excels in segmenting H&E stained pathology images, effectively preserving morphological features and spatial information even in scenarios with weak texture, poor contrast, and variations in appearance.

Spatiotemporal monitoring of climate change impacts on water resources using an integrated approach of remote sensing and Google Earth Engine

In this study, a data-driven approach employed by utilizing the product called JRC-Global surface water mapping layers V1.4 on the Google Earth Engine (GEE) to map and monitor the effects of climate change on surface water resources. Key climatic variables affecting water bodies, including air temperature (AT), actual evapotranspiration (ETa), and total precipitation, were analyzed from 2000 to 2021 using the temperature-vegetation index (TVX) and Moderate Resolution Imaging Spectroradiometer (MODIS) products. The findings demonstrate a clear association between global warming and the shrinking of surface water resources in the LUB. According to the results, an increase in AT corresponded to a decrease in water surface area, highlighting the significant influence of AT and ETa on controlling the water surface in the LUB (partial rho of − 0.65 and − 0.68, respectively). Conversely, no significant relationship was found with precipitation and water surface area (partial rho of + 0.25). Notably, the results of the study indicate that over the past four decades, approximately 40% of the water bodies in the LUB remained permanent. This suggests a loss of around 30% of the permanent water resources, which have transitioned into seasonal water bodies, accounting for nearly 13% of the total. This research provides a comprehensive framework for monitoring surface water resource variations and assessing the impact of climate change on water resources. It aids in the development of sustainable water management strategies and plans, supporting the preservation and effective use of water resources.

Link Prediction Based on Feature Mapping and Bi-Directional Convolution

A considerable amount of research on link prediction has recently been driven by missing relationships between knowledge graph entities and the problem of the incompleteness of knowledge graphs. Some recent studies have shown that convolutional neural networks based on knowledge embeddings are highly expressive and have good performance in link prediction. However, we found that the convolutional neural network (CNN)-based models do not handle the link between relations and entities well. For this reason, this paper proposes a link prediction model (LPM) based on feature mapping and bi-directional convolution. For the modeling of the task, an encoding layer–mapping layer–decoding layer structure is used. Among these layers, the encoding layer adopts a graph attention network to encode multi-hop triad information and obtains richer encoding of entities and relationships. The mapping layer can realize the mapping transformation between entities and relations and project the entity encoding in the space of relation encoding to capture the subtle connection between entities and relations. The decoding layer adopts bidirectional convolution to merge and decode the triples in a sequential inverse order, which makes the decoding layer model more advantageous in prediction. In addition, the decoding layer also adopts the r-drop training method to effectively reduce the distribution error generated by training between models and enhance the robustness of the model. Our experiments demonstrated the effectiveness of mapping relations, bidirectional convolution, and r-drop, and the accuracy of the proposed model showed significant improvements for each evaluation metric on two datasets, WN18RR and FB15k-237.

Mapping past and future shipping noise in European seas

Against the backdrop of a steadily increasing demand for sea transport of goods and people, the development of a reliable marine shipping soundscape model is an essential planning requirement to assess the effect on ocean noise of operational and technological changes aimed at mitigating the environmental impact of the shipping sector. The NAVISON (Navis Sonus) project, conducted with the support of the European Maritime Safety Agency, employs a specially developed parametric vessel source model with the objective of producing shipping sound maps in European seas for past, present, and potential future conditions over a time span from 2016 to 2050. The source model is combined with historical ship tracking data from the automated identification system (AIS), or projected shipping densities and mitigation scenarios, to calculate spatial ship noise emissions data for input to a sound mapping tool. The mapping tool computes underwater sound propagation using the parabolic-equation method, drawing upon ocean-scale databases of bathymetric, oceanographic, and sediment properties. Project outputs are provided as map layers of sound pressure level and sound energy according to vessel type, season, region, year, and operational conditions; from these layers, maps can be generated for user-specified combinations of mitigation measures. Maps are presented in two frequency bands (centred at 63 Hz and 125 Hz) selected for assessing Good Environmental Status in the context of the European Union’s Marine Strategy Framework Directive.

Seasonal dynamics of fallow and cropping lands in the broadacre cropping region of Australia

Fallowing is an important strategy for enhancing soil health, water harvesting and crop yields, thus improving sustainability and reducing production risks in dryland farming systems in Australia. However, accurate data regarding the location, frequency, extent, and duration of fallow land is not readily available at high spatio-temporal resolutions before and during a cropping season. As a result, continental maps depicting seasonal dynamics of fallow and cropping lands and their responses to climate change and human activities remain missing. This study set out to develop an automated approach to discriminate fallow areas from cropping areas within Australian broadacre cropping region. This was done by applying a mathematical algorithm of turning point detection technique to a curve fitted vegetation index time series from 2016 to 2022, derived from high spatial (10 m) and temporal (5-days) resolution Sentinel-2 data. Specifically, we aimed to derive: (i) 5-day revisit fallow/cropping map layers; (ii) key attributes of the fallow and cropping periods (e.g., start, end, duration); (iii) cropping intensity and area. The proposed approach yielded a high and significant overall accuracy, surpassing 92% with substantial balanced accuracy (>0.9) against >3000 fields for individual winter and summer seasons across Australia. In addition, this study's temporal and geographical outputs, encompassing seasonal attributes, cropping intensity and cropping areas, robustly enhanced the understanding of seasonal cropping dynamics and historical shifts in dryland farming practices. Findings derived from this analysis aligned well with extreme climatic events and changes in crop activities adopted in response to such events. Future research will aim to extract key seasonal dynamics attributes influenced by cropping practices, enhancing the conversion of high-resolution earth observation data into actionable industry knowledge.

GIS-AHP APPROACH FOR A COMPREHENSIVE FRAMEWORK TO DETERMINE THE SUITABLE REGIONS FOR GEOTHERMAL POWER PLANTS IN IZMIR, TÜRKİYE

Geothermal energy is gaining more reputation and importance around the world. Correspondingly, suitable location selection is a critical step and has become necessary for the successful installation and operation of geothermal power plants. This study investigated suitability of İzmir region, located in the Aegean part of Türkiye, in terms of geothermal power plants applications by using the combination of Geographical Information System and Analytic Hierarchy Process. Based on the request of power plants, thirteen important criteria were evaluated under three main categories named as physical (C1), environmental (C2) and technical (C3). Moreover, expert’s opinions were taken into consideration to calculate the importance of these criteria. Key results showed that İzmir was suitable for geothermal power plants. The final suitability map layer pointed out that %8.73 (1.037 km2) of total area were determined as highly suitable regions in terms of installation. In addition, the obtained suitability map layer was compared with actual geothermal power plants. Based on the comparison study, power plants in Seferihisar were moderately suitable for geothermal power plants while the location of Balçova power plant was highly suitable. Regarding the suitability assessment in the present study, the location of Dikili power plants had the least suitability score.

A Spatial–Spectral Transformer for Hyperspectral Image Classification Based on Global Dependencies of Multi-Scale Features

Vision transformers (ViTs) are increasingly utilized for HSI classification due to their outstanding performance. However, ViTs encounter challenges in capturing global dependencies among objects of varying sizes, and fail to effectively exploit the spatial–spectral information inherent in HSI. In response to this limitation, we propose a novel solution: the multi-scale spatial–spectral transformer (MSST). Within the MSST framework, we introduce a spatial–spectral token generator (SSTG) and a token fusion self-attention (TFSA) module. Serving as the feature extractor for the MSST, the SSTG incorporates a dual-branch multi-dimensional convolutional structure, enabling the extraction of semantic characteristics that encompass spatial–spectral information from HSI and subsequently tokenizing them. TFSA is a multi-head attention module with the ability to encode attention to features across various scales. We integrated TFSA with cross-covariance attention (CCA) to construct the transformer encoder (TE) for the MSST. Utilizing this TE to perform attention modeling on tokens derived from the SSTG, the network effectively simulates global dependencies among multi-scale features in the data, concurrently making optimal use of spatial–spectral information in HSI. Finally, the output of the TE is fed into a linear mapping layer to obtain the classification results. Experiments conducted on three popular public datasets demonstrate that the MSST method achieved higher classification accuracy compared to state-of-the-art (SOTA) methods.

Dense Connected Edge Feature Enhancement Network for Building Edge Detection from High Resolution Remote Sensing Imagery

Deep-learning-based methods for building-edge-detection have been widely researched and applied in the field of image processing. However, these methods often emphasis the analysis of deep features, which may result in neglecting crucial shallow information representation. Furthermore, abstract features in the deep layers can potentially interfere with the accuracy of edge extraction. To address these challenges, we propose a densely connected edge-detection enhancement network (DCEFE-Net) for building-edge-detection in high-resolution remote sensing images. Firstly, by introducing spatial land channel attention modules, we effectively captured low-level spatial information and high-level semantic information from the input image. Secondly, the proposed edge-aware feature enhancement (EAFE) module emphasis the representation of informative edge features. By alliteratively generating multiple layers of edge-detection maps, it addresses the issue of edge detail loss and enhances edge-detection accuracy. Finally, the dense connectivity blocks strengthen the connections between the convolutional layers, thereby preventing the loss of edge features. Experimental results on the WHU and the Inria Aerial Image Labeling datasets validate the effectiveness of DCEFE-Net, as it consistently produces clear and reliable building-edge results.

Comprehensive visual information acquisition for tomato picking robot based on multitask convolutional neural network

The tomato picking robot's vision system faces two difficult tasks: precise tomato pose acquisition and stem location. Tomato pose and stem location can help determine the end effector pose and achieve collision-free picking. To realise efficient crop picking, the tasks of target location, pose detection, and obstacle semantic segmentation should be completed in one model to obtain comprehensive visual information. Therefore, the multitask convolutional neural network YOLO-MCNN is proposed, a new method to complete the above tasks in one model. By fusing multi-scale features and determining the optimal locations for the semantic segmentation branch, four strategies are proposed for enhancing the segmentation ability. The experiment results show that fusing the semantic segmentation branch with the second layer of shallow feature maps and placing the branch after the 17th layer can result in the best segmentation performance. Fusing shallow feature maps improves small target detection while merging multi-scale feature maps enhances semantic segmentation performance. Moreover, ablation experiments are conducted to understand the influence between multitask convolutional and single task networks. It proves that running multiple tasks on the same backbone network does not affect their performance. The YOLO-MCNN's target detection performance F1 is 87.8%, the semantic segmentation performance mIoU is 74.8%, the keypoint detection performance dlmk is 6.95 pixels, the network size is 15.2 MB, and the inference speed is 19.9ms. Compared with other target detection and semantic segmentation networks, it shows that the comprehensive performance of the YOLO-MCNN is the best. The method provides theoretical foundation for constructing multitask convolutional neural networks.

Lightweight and fast visual detection method for 3C assembly

In the context of 3C assembly scenarios, characterized by numerous semi-flexible, heterogeneous, and small slender targets, traditional target detection algorithms face significant challenges such as low accuracy, weak generalization, large model sizes, and slow inference speeds. To address these issues, this study introduces an enhanced method based on the YOLOv5 model, named YOLOv5-GTB. This method integrates Bidirectional Feature Pyramid Network (BiFPN), Ghost lightweight convolution, Vision Transformer (ViT), and adaptive activation function technologies, aiming to improve the accuracy and speed of target detection. Our approach utilizes Ghost convolution to construct a Ghost bottleneck layer, optimizing the feature extraction network while significantly reducing computational costs and enhancing the convolutional neural network’s feature extraction capabilities. Additionally, the Cross-Stage Partial Network (CSPNet) architecture is employed to effectively segment the data flow of the input feature map layer, thereby improving the efficiency of gradient processing. Furthermore, we introduce a fusion structure of CNN and Transformer, leveraging the strengths of convolutional neural networks in local feature extraction and the ViT’s capability in long-range information capture, thereby further enhancing the overall network’s feature extraction performance. Regarding feature fusion, considering the limitations of traditional top-down unidirectional information flow in effectively merging features with both location and semantic information, BiFPN is incorporated into the YOLOv5-GTB. This enhances the fusion of feature layers extracted from both the end of the backbone network and the first module’s fused feature layer, thus improving detection accuracy. Ablation and comparative experiments conducted on the 3C assembly scenario target detection dataset demonstrate the significant advantages of the YOLOv5-GTB model in terms of accuracy and speed. Ultimately, the application of this model to the 3C assembly platform successfully achieves rapid and accurate target recognition in this scenario.

Broad Multitask Learning System With Group Sparse Regularization.

The broad learning system (BLS) featuring lightweight, incremental extension, and strong generalization capabilities has been successful in its applications. Despite these advantages, BLS struggles in multitask learning (MTL) scenarios with its limited ability to simultaneously unravel multiple complex tasks where existing BLS models cannot adequately capture and leverage essential information across tasks, decreasing their effectiveness and efficacy in MTL scenarios. To address these limitations, we proposed an innovative MTL framework explicitly designed for BLS, named group sparse regularization for broad multitask learning system using related task-wise (BMtLS-RG). This framework combines a task-related BLS learning mechanism with a group sparse optimization strategy, significantly boosting BLS's ability to generalize in MTL environments. The task-related learning component harnesses task correlations to enable shared learning and optimize parameters efficiently. Meanwhile, the group sparse optimization approach helps minimize the effects of irrelevant or noisy data, thus enhancing the robustness and stability of BLS in navigating complex learning scenarios. To address the varied requirements of MTL challenges, we presented two additional variants of BMtLS-RG: BMtLS-RG with sharing parameters of feature mapped nodes (BMtLS-RGf), which integrates a shared feature mapping layer, and BMtLS-RGf and enhanced nodes (BMtLS-RGfe), which further includes an enhanced node layer atop the shared feature mapping structure. These adaptations provide customized solutions tailored to the diverse landscape of MTL problems. We compared BMtLS-RG with state-of-the-art (SOTA) MTL and BLS algorithms through comprehensive experimental evaluation across multiple practical MTL and UCI datasets. BMtLS-RG outperformed SOTA methods in 97.81% of classification tasks and achieved optimal performance in 96.00% of regression tasks, demonstrating its superior accuracy and robustness. Furthermore, BMtLS-RG exhibited satisfactory training efficiency, outperforming existing MTL algorithms by 8.04-42.85 times.

ViT‐AMD: A New Deep Learning Model for Age‐Related Macular Degeneration Diagnosis From Fundus Images

Age‐related macular degeneration (AMD) diagnosis using fundus images is one of the critical missions of the eye‐care screening program in many countries. Various proposed deep learning models have been studied for this research interest, which aim to achieve the mission and outperform human‐based approaches. However, research efforts are still required for the improvement of model classification accuracy, sensitivity, and specificity values. In this study, we proposed the model named as ViT‐AMD, which is based on the latest Vision Transformer (ViT) structure, to diagnosis a fundus image as normal, dry AMD, or wet AMD types. Unlike convolution neural network models, ViT consists of the attention map layers, which show more effective performance for image classification task. Our training process is based on the 5‐fold cross‐validation and transfer learning techniques using Chula‐AMD dataset at the Department of Ophthalmology, the King Chulalongkorn Memorial Hospital, Bangkok. Furthermore, we also test the performance of trained model using an independent image datasets. The results showed that for the 3‐classes AMD classification (normal vs. dry AMD vs. wet AMD) on the Chula‐AMD dataset, the averaged accuracy, precision, sensitivity, and specificity of our trained model are about 93.40%, 92.15%, 91.27%, and 96.57%, respectively. For result testing on independent datasets, the averaged accuracy, precision, sensitivity, and specificity of trained model are about 74, 20%, 75.35%, 74.13%, and 87.07%, respectively. Compared with the results from the baseline CNN‐based model (DenseNet201), the trained ViT‐AMD model has outperformed significantly. In conclusion, the ViT‐AMD model have proved their usefulness to assist the ophthalmologist to diagnosis the AMD disease.

MCAFNet: multiscale cross-layer attention fusion network for honeycomb lung lesion segmentation.

Accurate segmentation of honeycomb lung lesions from lung CT images plays a crucial role in the diagnosis and treatment of various lung diseases. However, the availability of algorithms for automatic segmentation of honeycomb lung lesions remains limited. In this study, we propose a novel multi-scale cross-layer attention fusion network (MCAFNet) specifically designed for the segmentation of honeycomb lung lesions, taking into account their shape specificity and similarity to surrounding vascular shadows. The MCAFNet incorporates several key modules to enhance the segmentation performance. Firstly, a multiscale aggregation (MIA) module is introduced in the input part to preserve spatial information during downsampling. Secondly, a cross-layer attention fusion (CAF) module is proposed to capture multiscale features by integrating channel information and spatial information from different layers of the feature maps. Lastly, a bidirectional attention gate (BAG) module is constructed within the skip connection to enhance the model's ability to filter out background information and focus on the segmentation target. Experimental results demonstrate the effectiveness of the proposed MCAFNet. On the honeycomb lung segmentation dataset, the network achieves an Intersection over Union (IoU) of 0.895, mean IoU (mIoU) of 0.921, and mean Dice coefficient (mDice) of 0.949, outperforming existing medical image segmentation algorithms. Furthermore, experiments conducted on additional datasets confirm the generalizability and robustness of the proposed model. The contribution of this study lies in the development of the MCAFNet, which addresses the lack of automated segmentation algorithms for honeycomb lung lesions. The proposed network demonstrates superior performance in accurately segmenting honeycomb lung lesions, thereby facilitating the diagnosis and treatment of lung diseases. This work contributes to the existing literature by presenting a novel approach that effectively combines multi-scale features and attention mechanisms for lung lesion segmentation. The code is available at https://github.com/Oran9er/MCAFNet .

INSTANCE SEGMENTATION OF LIDAR DATA WITH VISION TRANSFORMER MODEL IN SUPPORT INUNDATION MAPPING UNDER FOREST CANOPY ENVIRONMENT

Abstract. Inundation mapping in forest and dense vegetated areas requires the ability to generate well defined Digital Terrain Models (DTM) to derive floodwater extent, depth, and duration. Due to the occlusion caused by overlapping leaves and branch structures of forest canopies, the ability to extract elevation point clouds through UAV and airborne optical imagery and photogrammetry is challenging. LiDAR is an active sensor that acquires direct 3D measurements by transmitting hundreds of thousands of laser measurements per second producing incredibly detailed mapping layers of not only the terrain but also forest attributes such as crown diameter, tree density and height that can support inundation mapping as well as hydrological models and monitoring of floods.In this research, we propose a methodology to map the inundated areas under canopies by using photon base Geiger Mode LiDAR point cloud dataset and a deep learning model to conduct instance segmentation of tree canopy. The method is to segment the vegetation from water and determine the gap fraction between trees to quantify the penetration through canopy for the detection of water pixels in vegetated areas. To conduct the segmentation Masked-attention Mask Transformer (Mask2Former) for universal segmentation model was implemented and trained to automate the extraction of tree crown segments from the LiDAR data. Furthermore, a semi-automatic experimental approach using a Canopy Height Model and watershed segmentation was applied to develop a rapid tree crown annotation strategy.