Pooling Operations Research Articles

MFHARFNet: multi-branch feature hybrid and adaptive receptive field network for image segmentation

Abstract Accurate segmentation of medical images is crucial for disease diagnosis and understanding disease changes. Deep learning methods, utilizing encoder-decoder structures, have demonstrated cutting-edge performance in various medical image segmentation tasks. However, the pooling operation in the encoding stage results in feature loss, which makes the network lack the ability to fuse multi-scale information at different levels, hinders its effective perception of multi-scale information, and leads to poor segmentation performance. Drawing inspiration from the U-shaped network, this study introduces a multi-branch feature hybrid attention and adaptive receptive field network (MFHARFNet) for medical image segmentation. Building upon the encoder-decoder framework, we initially devise a multi-branch feature hybrid attention module (MFHAM) to seamlessly integrate feature maps of varying scales, capturing both fine-grained features and coarse-grained semantics across the entire scale. Furthermore, we redesign the skip connection to amalgamate feature information from different branches in the encoder stage and efficiently transmit it to the decoder, providing the decoder with global context feature maps at different levels. Finally, the adaptive receptive field (ARF) module is introduced in the decoder feature reconstruction stage to adapt and focus on related fields, ensuring the model’s adaptation to different segmentation target features, and achieving different weights for the output of different convolution kernels to improve segmentation performance. We comprehensively evaluate our method on medical image segmentation tasks, by using four public datasets across CT and MRI. Remarkably, MFHARFNet method consistently outperforms other state-of-the-art methods, exceeding UNet by 2.1%, 0.9%, 6.6% and 1.0% on Dice on ATLAS, LiTs, BraTs2019 and Spine and intervertebral disc datasets, respectively. In addition, MFHARFNet minimizes network parameters and computational complexity as much as possible. The source codes are in https://github.com/OneHundred99/MFHARFNet.

Fall detection method based on spatio-temporal coordinate attention for high-resolution networks

Fall behavior is closely related to the high mortality rate of the elderly, so fall detection has become an important and urgent research area in human behavior recognition. However, the existing fall detection methods, suffer from the loss of detailed action information during feature extraction due to the downsampling operation, resulting in subpar performance when detecting falls with similar behaviors such as lying and sitting. To solve the challenges, this study proposes a high-resolution spatio-temporal feature extraction method based on a spatio-temporal coordinate attention mechanism. The method employs 3D convolutions to extract spatio-temporal features and utilizes gradual down-sampling to generate a multi-resolution sub-network, thus realizing multi-scale fusion and perception enhancement of details. In particular, this study designs a pseudo-3D basic block, which simulates the ability of 3D convolution, to ensure the running speed of the network while controlling the number of parameters. Further, a spatio-temporal coordinate attention mechanism is designed to accurately extract the spatio-temporal positional changes of key skeletal points and the interrelationships among them. Long-term dependencies in horizontal, vertical, temporal directions are captured through three one-dimensional global pooling operations. Then the long-range relationships and channel correlations among features are captured by cascading and slicing operations. Finally, the key information is effectively highlighted by performing dot-multiplication operations between the feature maps from the horizontal, vertical and temporal directions and the input feature maps. Experimental results on three typical public datasets show that the proposed method can better extract motion features and improve the accuracy of fall detection.

Asphalt pavement raveling recognition based on deep learning algorithms

Abstract Pavement raveling can lead to the development of issues such as potholes. Monitoring and providing real-time early warnings for road surface developments is crucial for timely maintenance planning and effectively controlling the occurrence and progression of surface issues. To achieve intelligent identification of raveling distress on asphalt pavement surfaces, we utilized a portable laser-based pavement surface scanning device to collect raveling images, followed by histogram equalization to eliminate noise caused by environmental factors. Additionally, data enhancement strategies such as rotation and mirroring were used to enrich training data and expand the sample size. A Convolutional Neural Network (CNN) was employed to perform multi-layer convolution and pooling operations on the images to quickly extract features of distress. Finally, five pre-trained transfer learning models—ResNet-18, DenseNet, VGG16, Inception, and EfficientNet—were introduced to identify the asphalt surface disease data and the normal surface data. The results showed that the accuracy and loss of the training sets of several models exhibited a positive trend, with the EfficientNet model demonstrating stable and superior performance under the same training conditions. The precision matrix, confusion matrix, precision, recall, and F1 score were used as evaluation indexes to assess the effectiveness of the models for identifying pavement raveling. The EfficientNet and Inception models had the best accuracy and comprehensive performance, with accuracy rates of 0.96 and 0.95, respectively. The precision, recall, and F1 score are 0.95, 0.96, and 0.96 for the EfficientNet model respectively. A total of 146 images correctly predicted raveling disease and only 4 normal images were incorrectly predicted as diseased. For example, the EfficientNet model’s identification accuracy for pockmark disease on the PC platform was 0.963, verifying the model’s feasibility in practical applications. The principal contribution of this paper lies in the efficient and precise identification of asphalt pavement raveling, which adversely affects vehicular traveling comfortability. This provides theoretical and technical support for the intelligent identification of asphalt pavement defects and enhancing road asset maintenance management in the pavement engineering industry.

A novel image classification framework based on variational quantum algorithms

Image classification is a crucial task in machine learning with widespread practical applications. The existing classical framework for image classification typically utilizes a global pooling operation at the end of the network to reduce computational complexity and mitigate overfitting. However, this operation often results in a significant loss of information, which can affect the performance of classification models. To overcome this limitation, we introduce a novel image classification framework that leverages variational quantum algorithms (VQAs) hybrid approaches combining quantum and classical computing paradigms within quantum machine learning. The major advantage of our framework is the elimination of the need for the global pooling operation at the end of the network. In this way, our approach preserves more discriminative features and fine-grained details in the images, which enhances classification performance. Additionally, employing VQAs enables our framework to have fewer parameters than the classical framework, even in the absence of global pooling, which makes it more advantageous in preventing overfitting. We apply our method to different state-of-the-art image classification models and demonstrate the superiority of the proposed quantum architecture over its classical counterpart through a series of state vector simulation experiments on public datasets. Our experiments show that the proposed quantum framework achieves up to a 9.21% increase in accuracy and up to a 15.79% improvement in F1 score, compared to the classical framework. Additionally, we explore the impact of shot noise on our method through shot-based simulation and find that increasing the number of measurements does not always lead to better results. Selecting an appropriate number of measurements can yield optimal results, even surpassing those obtained from state vector simulation.

Topology-Agnostic Graph U-Nets for Scalar Field Prediction on Unstructured Meshes

Abstract Machine-learned surrogate models to accelerate lengthy computer simulations are becoming increasingly important as engineers look to streamline the product design cycle. In many cases, these approaches offer the ability to predict relevant quantities throughout a geometry, but place constraints on the form of the input data. In a world of diverse data types, a preferred approach would not restrict the input to a particular structure. In this paper, we propose Topology-Agnostic Graph U-Net (TAG U-Net), a graph convolutional network that can be trained to input any mesh or graph structure and output a prediction of a target scalar field at each node. The model constructs coarsened versions of each input graph and performs a set of convolution and pooling operations to predict the node-wise outputs on the original graph. By training on a diverse set of shapes, the model can make strong predictions, even for shapes unlike those seen during training. A 3-D additive manufacturing dataset is presented, containing Laser Powder Bed Fusion simulation results for thousands of parts. The model is demonstrated on this dataset, and it performs well, predicting both 2-D and 3-D scalar fields with a median R-squared > 0.85 on test geometries.

EIDU-Net: edge-preserved inception DenseGCN U-Net for LiDAR point cloud segmentation

With the development of laser scanners and machine learning, point cloud semantic segmentation plays a significant role in autonomous driving, scene reconstruction, human-computer interaction, and other fields. In recent years, point cloud semantic segmentation based on deep learning has become one of the key research directions in point cloud processing. Due to the limited ability to exploit geometric details and contextual information in point clouds, most methods that adopt encoder-decoder architecture lose local structural information easily, especially detailed features, and extract features insufficiently. To address this issue, the edge-preserving inception DenseGCN U-Net (entitled as EIDU-Net) is proposed. EIDU-Net makes full use of the complementation between geometric details in the original point cloud and high-level features. The edge-preserved graph pooling (EGP) layer, the key module of the EIDU-Net, is designed to retain additional edge feature information from the original point cloud during pooling operations. Accordingly, the edge-preserved graph unpooling (EGU) layer can restore the feature graph more efficiently based on the additionally retained edge features. Extensive experiments demonstrate that our proposed EIDU-Net has remarkable improvements on semantic segmentation tasks under whatever S3DIS or Terracotta Warrior fragments. Our code is publicly available at https://github.com/caoxin918/EIDU-Net.

Piecewise convolutional neural network relation extraction with self-attention mechanism

The task of relation extraction in natural language processing is to identify the relation between two specified entities in a sentence. However, the existing model methods do not fully utilize the word feature information and pay little attention to the influence degree of the relative relation extraction results of each word. In order to address the aforementioned issues, we propose a relation extraction method based on self-attention mechanism (SPCNN-VAE) to solve the above problems. First, we use a multi-head self-attention mechanism to process word vectors and generate sentence feature vector representations, which can be used to extract semantic dependencies between words in sentences. Then, we introduce the word position to combine the sentence feature representation with the position feature representation of words to form the input representation of piecewise convolutional neural network (PCNN). Furthermore, to identify the word feature information that is most useful for relation extraction, an attention-based pooling operation is employed to capture key convolutional features and classify the feature vectors. Finally, regularization is performed by a variational autoencoder (VAE) to enhance the encoding ability of model word information features. The performance analysis is performed on SemEval 2010 task 8, and the experimental results show that the proposed relation extraction model is effective and outperforms some competitive baselines.

Point cloud upsampling network based on pyramid pooling and self-attention mechanism

For the problems of irregular point cloud data and non-unique input size, a point-cloud upsampling network (PPSA-PU) based on Pyramid Pooling and Self-Attention mechanism is proposed. Multi-scale features are first extracted from the input point cloud using dense map convolution with different expansion rates, and the pyramid pooling module and self-attention mechanism are added. By applying the same pooling operation to regions of different sizes and numbers, the network is made to accept inputs of arbitrary size. Self-attention is used to weight the features to establish global correlation and improve the feature extraction capability. The features are then extended by the Self-Attention NodeShuffle (SA-NS) module, which uses a graph convolution layer to fuse neighbour information and learn new point features from the latent space. Finally, the coordinate reconstruction module is used to generate a dense up-sampled point cloud. Using the existing dataset to train and test the network and compare it with the existing up-sampling network, the experimental results show that PPSA-PU is better in terms of evaluation metrics and performance, and has better generalisation and robustness.

Wildlife target detection based on improved YOLOX-s network

To addresse the problem of poor detection accuracy or even false detection of wildlife caused by rainy environment at night. In this paper, a wildlife target detection algorithm based on improved YOLOX-s network is proposed. Our algorithm comprises the MobileViT-Pooling module, the Dynamic Head module, and the Focal-IoU module.First, the MobileViT-Pooling module is introduced.It is based on the MobileViT attention mechanism, which uses a spatial pooling operator with no parameters as a token mixer module to reduce the number of network parameters. This module performs feature extraction on three feature layers of the backbone network output respectively, senses the global information and strengthens the weight of the effective information. Second, the Dynamic Head module is used on the downstream task of network detection, which fuses the information of scale sensing, spatial sensing, and task sensing and improves the representation ability of the target detection head. Lastly, the Focal idea is utilized to improve the IoU loss function, which balances the learning of high and low quality IoU for the network. Experimental results reveal that our algorithm achieves a notable performance boost with mAP@0.5 reaching 87.8% (an improvement of 7.9%) and mAP@0.5:0.95 reaching 62.0% (an improvement of 5.3%). This advancement significantly augments the night-time wildlife detection accuracy under rainy conditions, concurrently diminishing false detections in such challenging environments.

PointCloud-At: Point Cloud Convolutional Neural Networks with Attention for 3D Data Processing

The rapid growth in technologies for 3D sensors has made point cloud data increasingly available in different applications such as autonomous driving, robotics, and virtual and augmented reality. This raises a growing need for deep learning methods to process the data. Point clouds are difficult to be used directly as inputs in several deep learning techniques. The difficulty is raised by the unstructured and unordered nature of the point cloud data. So, machine learning models built for images or videos cannot be used directly on point cloud data. Although the research in the field of point clouds has gained high attention and different methods have been developed over the decade, very few research works directly with point cloud data, and most of them convert the point cloud data into 2D images or voxels by performing some pre-processing that causes information loss. Methods that directly work on point clouds are in the early stage and this affects the performance and accuracy of the models. Advanced techniques in classical convolutional neural networks, such as the attention mechanism, need to be transferred to the methods directly working with point clouds. In this research, an attention mechanism is proposed to be added to deep convolutional neural networks that process point clouds directly. The attention module was proposed based on specific pooling operations which are designed to be applied directly to point clouds to extract vital information from the point clouds. Segmentation of the ShapeNet dataset was performed to evaluate the method. The mean intersection over union (mIoU) score of the proposed framework was increased after applying the attention method compared to a base state-of-the-art framework that does not have the attention mechanism.

A Dual-Dimension Convolutional-Attention Module for Remaining Useful Life Prediction of Aeroengines

Remaining useful life (RUL) prediction of aeroengines not only enhances aviation safety and operational efficiency but also significantly lowers operational costs, offering substantial economic and social benefits to the aviation industry. Aiming at RUL prediction, this paper proposes a novel dual-dimension convolutional-attention (DDCA) mechanism. DDCA consists of two branches: one includes channel attention and spatial attention mechanisms, while the other applies these mechanisms to the inverted dimensions. Pooling and feature-wise pooling operations are employed to extract features from different dimensions of the input data. These branches operate in parallel to capture more complex temporal and spatial feature correlations in multivariate time series data. Subsequently, an end-to-end DDCA-TCN network is constructed by integrating DDCA with a temporal convolutional network (TCN) for RUL prediction. The proposed prediction model is evaluated using the C-MAPSS dataset and compared to several state-of-the-art RUL prediction models. The results show that the RMSE and SCORE metrics of DDCA-TCN decreased by at least 12.8% and 4.6%, respectively, compared to other models on the FD002 subset, and by at least 10.6% and 18.4%, respectively, on the FD004 subset, which demonstrates that the DDCA-TCN model exhibits excellent performance in RUL prediction, particularly under multiple operating conditions.

Feature diffusion reconstruction mechanism network for crop spike head detection.

Monitoring crop spike growth using low-altitude remote sensing images is essential for precision agriculture, as it enables accurate crop health assessment and yield estimation. Despite the advancements in deep learning-based visual recognition, existing crop spike detection methods struggle to balance computational efficiency with accuracy in complex multi-scale environments, particularly on resource-constrained low-altitude remote sensing platforms. To address this gap, we propose FDRMNet, a novel feature diffusion reconstruction mechanism network designed to accurately detect crop spikes in challenging scenarios. The core innovation of FDRMNet lies in its multi-scale feature focus reconstruction and lightweight parameter-sharing detection head, which can effectively improve the computational efficiency of the model while enhancing the model's ability to perceive spike shape and texture.FDRMNet introduces a Multi-Scale Feature Focus Reconstruction module that integrates feature information across different scales and employs various convolutional kernels to capture global context effectively. Additionally, an Attention-Enhanced Feature Fusion Module is developed to improve the interaction between different feature map positions, leveraging adaptive average pooling and convolution operations to enhance the model's focus on critical features. To ensure suitability for low-altitude platforms with limited computational resources, we incorporate a Lightweight Parameter Sharing Detection Head, which reduces the model's parameter count by sharing weights across convolutional layers. According to the evaluation experiments on the global wheat head detection dataset and diverse rice panicle detection dataset, FDRMNet outperforms other state-of-the-art methods with mAP@.5 of 94.23%, 75.13% and R 2 value of 0.969, 0.963 between predicted values and ground truth values. In addition, the model's frames per second and parameters in the two datasets are 227.27,288 and 6.8M, respectively, which maintains the top three position among all the compared algorithms. Extensive qualitative and quantitative experiments demonstrate that FDRMNet significantly outperforms existing methods in spike detection and counting tasks, achieving higher detection accuracy with lower computational complexity.The results underscore the model's superior practicality and generalization capability in real-world applications. This research contributes a highly efficient and computationally effective solution for crop spike detection, offering substantial benefits to precision agriculture practices.

Mesh Convolution With Continuous Filters for 3-D Surface Parsing.

Geometric feature learning for 3-D surfaces is critical for many applications in computer graphics and 3-D vision. However, deep learning currently lags in hierarchical modeling of 3-D surfaces due to the lack of required operations and/or their efficient implementations. In this article, we propose a series of modular operations for effective geometric feature learning from 3-D triangle meshes. These operations include novel mesh convolutions, efficient mesh decimation, and associated mesh (un)poolings. Our mesh convolutions exploit spherical harmonics as orthonormal bases to create continuous convolutional filters. The mesh decimation module is graphics processing unit (GPU)-accelerated and able to process batched meshes on-the-fly, while the (un)pooling operations compute features for upsampled/downsampled meshes. We provide an open-source implementation of these operations, collectively termed Picasso. Picasso supports heterogeneous mesh batching and processing. Leveraging its modular operations, we further contribute a novel hierarchical neural network for perceptual parsing of 3-D surfaces, named PicassoNet++. It achieves highly competitive performance for shape analysis and scene segmentation on prominent 3-D benchmarks. The code, data, and trained models are available at https://github.com/EnyaHermite/Picasso.

Optimization of steel plate quality inspection driven by PscSE and SPPFELAN

AbstractBased on the improved YOLOv8n, a steel plate defect detection and recognition method is proposed to address the high labor costs and workload of traditional tasks. SPPFELAN processes inputs in parallel to enhance computational efficiency by executing multiple pooling operations simultaneously. The parallel feature fusion module PscSE, using a mixed‐dimension SE attention mechanism (scSE), captures global and channel‐related information better, improving characterization capability. The EIOU loss function addresses the ambiguous aspect ratio definition of CIOU loss, enhancing detection accuracy and accelerating convergence. Results show the YOLOv8n‐PscSE‐SPPFELAN model achieves 76.9% mAP@0.5 on the Northeastern University steel plate dataset, a 4.6% improvement over the original YOLOv8n, with a computation amount of 7.7 GFLOPs, reducing resource usage and greatly improving detection speed.

LUT-SLS: A Lightweight Transformer Network Based on U-Net for Skin Lesion Segmentation

In recent years, deep learning has achieved significant advancements in medical image segmentation, primarily focusing on CNN structures and Transformer-based architectures. However, these network architectures often suffer from high computational complexity and a large number of parameters. To address these challenges, this paper proposes a lightweight, novel structured network called LUT-SLS, based on U-Net and Transformer. Firstly, the overall structure integrates U-Net and Transformer, which effectively captures long-range dependencies in image relationships and contextual information, thereby improving segmentation accuracy. Secondly, a novel PLTS module is designed, which replaces the traditional self-attention mechanism with average pooling operations to extract global features and local details. Additionally, a novel MMLP structure is introduced, incorporating residual depth-separable operations into the traditional fully-connected framework. This enhances the processing of pooled features and further improves feature expression capability. Finally, the encoder and decoder parts are connected by the MSBN module, which facilitates the extraction of deep features while fusing encoder features. Experimental results demonstrate that the proposed model achieves competitive advantages in balancing the number of parameters, computational complexity, and performance compared to current leading models on multiple public datasets. This solution enables model deployment on IoT terminals, assisting doctors in making more accurate clinical decisions.

Skin Lesion Segmentation through Generative Adversarial Networks with Global and Local Semantic Feature Awareness

The accurate segmentation of skin lesions plays an important role in the diagnosis and treatment of skin cancers. However, skin lesion areas are rich in details and local features, including the appearance, size, shape, texture, etc., which pose challenges for the accurate localization and segmentation of the target area. Unfortunately, the consecutive pooling and stride convolutional operations in existing convolutional neural network (CNN)-based solutions lead to the loss of some spatial information and thus constrain the accuracy of lesion region segmentation. In addition, using only the traditional loss function in CNN cannot ensure that the model is adequately trained. In this study, a generative adversarial network is proposed, with global and local semantic feature awareness (GLSFA-GAN) for skin lesion segmentation based on adversarial training. Specifically, in the generator, a multi-scale localized feature fusion module and an effective channel-attention module are designed to acquire the multi-scale local detailed information of the skin lesion area. In addition, a global context extraction module in the bottleneck between the encoder and decoder of the generator is used to capture more global semantic features and spatial information about the lesion. After that, we use an adversarial training strategy to make the discriminator discern the generated labels and the segmentation prediction maps, which assists the generator in yielding more accurate segmentation maps. Our proposed model was trained and validated on three public skin lesion challenge datasets involving the ISIC2017, ISIC2018, and HAM10000, and the experimental results confirm that our proposed method provides a superior segmentation performance and outperforms several comparative methods.

FBSM: Foveabox-based boundary-aware segmentation method for green apples in natural orchards

Image segmentation of target fruits is an essential part of machine vision systems, aiming to facilitate more accurate early fruit measurement and machine harvesting in natural orchard environments. Given that instance-level pooling and down-sampling operations in conventional segmentation models often lead to the loss of detailed information, resulting in coarse partitioning masks, we endeavor to restore the boundary information of the masks. To achieve high-quality fruit segmentation with clear boundaries and refined masks, a Foveabox-based boundary-aware segmentation model (FBSM) is constructed by adding a multi-stage mask prediction head incorporating fine-grained features to the anchor-free detection model FoveaBox to realize accurate segmentation of green apple fruits. At each stage, a bi-layer fusion structure (BFS) is employed to fully fuse the fine-grained features in a double-fusion manner to guide the subsequent instance mask prediction. Finally, the boundary recovery module (BRM) is leveraged to recover the lost boundary detail information and obtain more accurate boundaries for the continuous optimization process of the fruit instance mask. Experimental results demonstrate that the proposed FBSM model achieves a mean average precision (mAP) of 60.7 % for green apple instance segmentation in unstructured natural orchard environments, surpassing traditional instance segmentation models. Moreover, it enhances base model detection accuracy by 1.3 % while striking a better balance between detection and segmentation performance.

Graph pooling in graph neural networks: methods and their applications in omics studies

Graph neural networks (GNNs) process the graph-structured data using neural networks and have proven successful in various graph processing tasks. Currently, graph pooling operators have emerged as crucial components that bridge the gap between node representation learning and diverse graph-level tasks by transforming node representations into graph representations. Given the rapid growth and widespread adoption of graph pooling, this review aims to summarize the existing graph pooling operators for GNNs and their representative applications in omics. Specifically, we first present a comprehensive taxonomy of existing graph pooling algorithms, expanding the categorization for both global and hierarchical pooling operators, and for the first time reviewing the inverse operation of graph pooling, named unpooling. Next, we describe the general evaluation framework for graph pooling operators, encompassing three fundamental aspects: experimental setup, ablation analysis, and model interpretation. We also discuss open issues that significantly influence the design of graph pooling operators, including complexity, connectivity, adaptability, additional loss, and attention mechanisms. Finally, we summarize bioinformatics applications of graph pooling operators in omics, including graphs of gene interaction, medical images, and protein structures for drug discovery and disease diagnosis. Furthermore, we showcase the impact of graph pooling operators on research in specific real-world domains, with a focus on prediction performance and model interpretability. This review provides methodological insights in machine learning based graph modeling and related omics research, as well as an ongoing resource by gathering related papers and code in a dedicated GitHub repository (https://github.com/Hou-WJ/Graph-Pooling-Operators-and-Bioinformatics-Applications).

Automatic Water Body Extraction from SAR Images Based on MADF-Net

Water extraction from synthetic aperture radar (SAR) images has an important application value in wetland monitoring, flood monitoring, etc. However, it still faces the problems of low generalization, weak extraction ability of detailed information, and weak suppression of background noises. Therefore, a new framework, Multi-scale Attention Detailed Feature fusion Network (MADF-Net), is proposed in this paper. It comprises an encoder and a decoder. In the encoder, ResNet101 is used as a solid backbone network to capture four feature levels at different depths, and then the proposed Deep Pyramid Pool (DAPP) module is used to perform multi-scale pooling operations, which ensure that key water features can be captured even in complex backgrounds. In the decoder, a Channel Spatial Attention Module (CSAM) is proposed, which focuses on feature areas that are critical for the identification of water edges by fusing attention weights in channel and spatial dimensions. Finally, the high-level semantic information is effectively fused with the low-level edge features to achieve the final water detection results. In the experiment, Sentinel-1 SAR images of three scenes with different characteristics and scales of water body are used. The PA and IoU of water extraction by MADF-Net can reach 92.77% and 89.03%, respectively, which obviously outperform several other networks. MADF-Net carries out water extraction with high precision from SAR images with different backgrounds, which could also be used for the segmentation and classification of other tasks from SAR images.

A Novel Deep Learning Approach for Real-Time Critical Assessment in Smart Urban Infrastructure Systems

The swift advancement of communication and information technologies has transformed urban infrastructures into smart cities. Traditional assessment methods face challenges in capturing the complex interdependencies and temporal dynamics inherent in these systems, risking urban resilience. This study aims to enhance the criticality assessment of geographic zones within smart cities by introducing a novel deep learning architecture. Utilizing Convolutional Neural Networks (CNNs) for spatial feature extraction and Long Short-Term Memory (LSTM) networks for temporal dependency modeling, the proposed framework processes inputs such as total electricity use, flooding levels, population, poverty rates, and energy consumption. The CNN component constructs hierarchical feature maps through successive convolution and pooling operations, while the LSTM captures sequence-based patterns. Fully connected layers integrate these features to generate final predictions. Implemented in Python using TensorFlow and Keras on an Intel Core i7 system with 32 GB RAM and an NVIDIA GTX 1080 Ti GPU, the model demonstrated a superior performance. It achieved a mean absolute error of 0.042, root mean square error of 0.067, and an R-squared value of 0.935, outperforming existing methodologies in real-time adaptability and resource efficiency.