Feature Preservation Research Articles

MATTopo: Topology-preserving Medial Axis Transform with Restricted Power Diagram

We present a novel topology-preserving 3D medial axis computation framework based on volumetric restricted power diagram (RPD), while preserving the medial features and geometric convergence simultaneously, for both 3D CAD and organic shapes. The volumetric RPD discretizes the input 3D volume into sub-regions given a set of medial spheres. With this intermediate structure, we convert the homotopy equivalency between the generated medial mesh and the input 3D shape into a localized contractibility checking for each restricted element (power cell, power face, power edge), by checking their connected components and Euler characteristics. We further propose a fractional Euler characteristic algorithm for efficient GPU-based computation of Euler characteristic for each restricted element on the fly while computing the volumetric RPD. Compared with existing voxel-based or point-cloud-based methods, our approach is the first to adaptively and directly revise the medial mesh without globally modifying the dependent structure, such as voxel size or sampling density, while preserving its topology and medial features. In comparison with the feature preservation method MATFP [Wang et al. 2022], our method provides geometrically comparable results with fewer spheres and more robustly captures the topology of the input 3D shape.

A Cubical Persistent Homology-Based Technique for Image Denoising with Topological Feature Preservation

Image denoising is a fundamental challenge in image processing, where the objective is to remove noise while preserving critical image features. Traditional denoising methods, such as Wavelet, Total Variation (TV) minimization, and Non-Local Means (NLM), often struggle to maintain the topological integrity of image features, leading to the loss of essential structures. This study proposes a Cubical Persistent Homology-Based Technique (CPHBT) that leverages persistence barcodes to identify significant topological features and reduce noise. The method selects filtration levels that preserve important features like loops and connected components. Applied to digit images, our method demonstrates superior performance, achieving a Peak Signal-to-Noise Ratio (PSNR) of 46.88 and a Structural Similarity Index Measure (SSIM) of 0.99, outperforming TV (PSNR: 21.52, SSIM: 0.9812) and NLM (PSNR: 22.09, SSIM: 0.9822). These results confirm that cubical persistent homology offers an effective solution for image denoising by balancing noise reduction and preserving critical topological features, thus enhancing overall image quality.

Security of End-to-End medical images encryption system using trained deep learning encryption and decryption network

The Internet of Medical Things (IoMT) links medical devices and wearable, enhancing healthcare. To secure sensitive patient data over the IoMT, encryption is vital to retain confidentiality, prevent tampering, ensure authenticity, and secure data transfer. The intricate neural network architecture of deep learning models adds a layer of complexity and non-linearity to the encryption process, rendering it highly resistant to plaintext attacks. Specifically, the Cycle_GAN network is used as the leading learning network. This work suggests deep learning-based encryption for medical images using Cycle_GAN, a Generative Adversarial Network. Cycle_GAN changes images without paired training data that improves quality and feature preservation. Unlike conventional image-to-image translation techniques, Cycle_GAN doesn’t require a dataset with corresponding input–output pairs. Traditional methods typically needs paired data to learn the mapping between input and output images. Paired data can be challenging to obtain, specifically in medical imaging where gathering and annotating data can be time-consuming, laborious and expensive. The use of Cycle_GAN overwhelms this constraint by using unpaired data, where the input and output images are not explicitly paired. This method ensures confidentiality, authenticity, and secure transfer. Cycle_GAN consists of two major components: a generator used to modify the images, and a discriminator used to distinguish between real and fake images. Further, the Binary-Cross Entropy loss function is employed to train the network for precise predictions. The experiments are carried out on skin cancer datasets. The results demonstrate high-level efficient, systematic and coherent encryption as compared with other modernized medical image encryption methods. The proposed technique offers several benefits, including efficient encryption and decryption and robustness against unauthorized access.

A hybrid‐ensemble model for software defect prediction for balanced and imbalanced datasets using AI‐based techniques with feature preservation: SMERKP‐XGB

AbstractMaintaining software quality is a significant challenge as the complexity of software is increasing with the rise of the software industry. Software defects are a primary concern in complex modules, and predicting them in the early stages of the software development life cycle (SDLC) is difficult. Previous techniques to address this issue have not been very promising. We have proposed “A hybrid ensemble model for software defect prediction using AI‐based techniques with feature preservation” to overcome this problem. We have used the National Aeronautics and Space Administration (NASA) dataset from the PROMISE repository for testing and validation. By applying exploratory data analysis (EDA), feature engineering, scaling, and standardization, we found that the dataset is imbalanced, which can negatively affect the model's performance. To address this, we have used the Synthetic Minority Oversampling (SMOTE) technique and the edited nearest neighbor (ENN) (SMOTE‐ENN). We have also used recursive feature elimination cross‐validation (RFE‐CV) with a pipeline to prevent data leaking in CV and kernel‐based principal component analysis (K‐PCA) to minimize dimensionality and selectively relevant features. The reduced dimensional data is then given to the eXtreme Gradient Boosting (XGBoost) for classification, resulting in the hybrid‐ensemble (SMERKP‐XGB) model. The proposed SMERKP‐XGB model is better than previously developed models in terms of accuracy (CM1: 97.53%, PC1: 92.05%, and PC2: 97.45%, KC1:95.65%), and area under the receiver operating characteristic curve values (CM1:96.30%, PC1:98.30%, and PC2:99.30%: KC1: 93.54) and other evaluation criteria mentioned in the literature.

An adaptive trajectory compression and feature preservation method for maritime traffic analysis

Ship trajectory data extracted from Automatic Identification System (AIS) has been extensively used for maritime traffic analysis. Yet the enormous volume of AIS data has come with substantial challenges related to storing, processing, analyzing, transmitting, and transferring. Trajectory compression techniques have been widely investigated to remedy the challenge. However, conventional compression techniques such as Douglas-Peucker (DP) algorithm mainly depend on line simplification algorithms, falling short in accurately identifying and preserving crucial information within trajectories. Moreover, using kinematic information from AIS data has posed difficulties associated with compression threshold determination. Hence, an adaptive method capable of considering multiple information from AIS is required. In this paper, a Top-Down Kinematic Compression (TDKC) algorithm aimed at adaptive trajectory compression and feature preservation is proposed. By incorporating time, position, speed, and course attributes from AIS data, TDKC exploits a Compression Binary Tree (CBT) method to address the recursion termination problem and determine the threshold automatically. A case study was conducted to evaluate the performance of TDKC using AIS data from Gulf of Finland, where a comparison with conventional algorithms and their improved versions based on specific performance evaluation metrics was involved. The results demonstrate TDKC's superiority in facilitating maritime traffic analysis.

Beyond boundaries: Unifying classification and segmentation in wildfire detection systems

AbstractForest fires, a dangerous natural phenomenon, cause large-scale destruction in forests and nearby communities. In this paper, we leverage the capabilities of classification and fast prediction of machine learning and deep learning to detect forest fires by proposing a dual-stage approach of identification and segmentation of burnt/burning regions. The initial identification step uses simple classification by harnessing custom convolutional neural network and AlexNet to detect satellite images displaying signs of burning, followed by the next step of segmentation, which employs UNet and SegNet architectures to precisely localize the burnt/burning areas. We chose UNet considering its ability of efficient feature preservation, and SegNet prominent use for precise boundary delineation. For training our models, we have curated a dataset using Google Earth Engine, querying latitude, longitude, and dates. Image processing techniques facilitated the extraction of segmentation masks, bounding box images, and segmented images. The post-training results portray that CNN outperforms AlexNet with an 88.19% accuracy for classification, while in segmentation, UNet excelled over SegNet with a dice score of 0.6869. If deployed, this study stands to benefit forest officials, local communities, and biodiversity. Moreover, it aids afforestation teams in identifying burnt zones for targeted reforestation efforts, ensuring a more effective response to forest fires.

Non-Rigid Registration Via Intelligent Adaptive Feedback Control.

Preserving features or local shape characteristics of a mesh using conventional non-rigid registration methods is always difficult, as the preservation and deformation are competing with each other. The challenge is to find a balance between these two terms in the process of the registration, especially in presence of artefacts in the mesh. We present a non-rigid Iterative Closest Points (ICP) algorithm which addresses the challenge as a control problem. An adaptive feedback control scheme with global asymptotic stability is derived to control the stiffness ratio for maximum feature preservation and minimum mesh quality loss during the registration process. A cost function is formulated with the distance term and the stiffness term where the initial stiffness ratio value is defined by an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based predictor regarding the source mesh and the target mesh topology, and the distance between the correspondences. During the registration process, the stiffness ratio of each vertex is continuously adjusted by the intrinsic information, represented by shape descriptors, of the surrounding surface as well as the steps in the registration process. Besides, the estimated process-dependent stiffness ratios are used as dynamic weights for establishing the correspondences in each step of the registration. Experiments on simple geometric shapes as well as 3D scanning datasets indicated that the proposed approach outperforms current methodologies, especially for the regions where features are not eminent and/or there exist interferences between/among features, due to its ability to embed the inherent properties of the surface in the process of the mesh registration.

Filtering of 3D point clouds using maximum likelihood algorithm

Recently, the 3D point cloud (PC) has become more popular as an innovative object representation. However, there is usually noise and outliers in the raw point cloud. It is essential to eliminate the noise from the point cloud and outlier data while maintaining the features and finer details intact. This paper presents a comprehensive method for filtering and classification point clouds using a maximum likelihood algorithm (ML). TOPCON GLS-2000 3D terrestrial laser scanners (TLS) have been used to collect the 3D PC data set; the scan range is up to 350 m. About 30 m apart from the study area. ScanMaster software has been used to import, view, and filter point cloud information. The position information of the points is linked with the training point cloud and the filtered point cloud to derive the nonlinear model using MATLAB software. To evaluate the quality of the denoising results, two error metrics have been used: the average angle (δ) and distance (Dmean) between the ground truth point and the resulting point. The experimental findings demonstrate that the suggested approach can effectively filter out background noise while improving feature preservation. The filtering and classifying technique is more effective and efficient compared to the selected filtering methods when applied to 3D point clouds containing a large number of points and a variety of natural characteristics.

ITFuse: An interactive transformer for infrared and visible image fusion

Infrared and visible image fusion (IVIF) has attracted increasing attention from the community because of its pleasing results in downstream applications. However, most existing deep fusion models are either feature–level fusion or image–level fusion, leading to information loss. In this paper, we propose an interactive transformer for IVIF, termed ITFuse. In contrast to previous algorithms, ITFuse consists of feature interactive modules (FIMs) and a feature reconstruction module (FRM) to alternatively extract and integrate important features. Specifically, to adequately exploit the common properties of different source images, we design a residual attention block (RAB) for mutual feature representation. To aggregate the distinct characteristics that existed in the corresponding input images, we leverage interactive attention (ITA) to incorporate the complementary information for comprehensive feature preservation and interaction. In addition, cross-modal attention (CMA) and transformer block (TRB) are presented to fully merge the capitalized features and construct long-range relationships. Furthermore, we devise a pixel loss and a structural loss to train the proposed deep fusion model in an unsupervised manner for excess performance amelioration. Massive experiments on popular databases illustrate that our ITFuse performs better than other representative state-of-the-art methods in terms of both qualitative and quantitative assessments. The source code of the proposed method is available at https://github.com/tthinking/ITFuse.

SimNFND: A Forward-Looking Sonar Denoising Model Trained on Simulated Noise-Free and Noisy Data

Given the propagation characteristics of sound waves and the complexity of the underwater environment, denoising forward-looking sonar image data presents a formidable challenge. Existing studies often add noise to sonar images and then explore methods for its removal. This approach neglects the inherent complex noise in sonar images, resulting in inaccurate evaluations of traditional denoising methods and poor learning of noise characteristics by deep learning models. To address the lack of high-quality data for FLS denoising model training, we propose a simulation algorithm for forward-looking sonar data based on RGBD data. By utilizing rendering techniques and noise simulation algorithms, high-quality noise-free and noisy sonar data can be rapidly generated from existing RGBD data. Based on these data, we optimize the loss function and training process of the FLS denoising model, achieving significant improvements in noise removal and feature preservation compared to other methods. Finally, this paper performs both qualitative and quantitative analyses of the algorithm’s performance using real and simulated sonar data. Compared to the latest FLS denoising models based on traditional methods and deep learning techniques, our method demonstrates significant advantages in denoising capability. All inference results for the Marine Debris Dataset (MDD) have been made open source, facilitating subsequent research and comparison.

Generalizing feature preservation in iso-surface extraction from triple dexel models

We present a method to resolve visual artifacts of a state-of-the-art iso-surface extraction algorithm by generating feature-preserving surface patches for isolated arbitrarily complex, single voxels without the need for further adaptive subdivision. In the literature, iso-surface extraction from a 3D voxel grid is limited to a single sharp feature per minimal unit, even for algorithms such as Cubical Marching Squares that produce feature-preserving surface reconstructions. In practice though, multiple sharp features can meet in a single voxel. This is reflected in the triple dexel model, which is used in simulation of CNC manufacturing processes. Our approach generalizes the use of normal information to perfectly preserve multiple sharp features for a single voxel, thus avoiding visual artifacts caused by state-of-the-art procedures.

Adaptively Isotropic Remeshing Based on Curvature Smoothed Field.

With the development of 3D digital geometry technology, 3D triangular meshes are becoming more useful and valuable in industrial manufacturing and digital entertainment. A high quality triangular mesh can be used to represent a real world object with geometric and physical characteristics. While anisotropic meshes have advantages of representing shapes with sharp features (such as trimmed surfaces) more efficiently and accurately, isotropic meshes allow more numerically stable computations. When there is no anisotropic mesh requirement, isotropic triangles are always a good choice. In this paper, we propose a remeshing method to convert an input mesh into an adaptively isotropic one based on a curvature smoothed field (CSF). With the help of the CSF, adaptively isotropic remeshing can retain the curvature sensitivity, which enables more geometric features to be kept, and avoid the occurrence of obtuse triangles in the remeshed model as much as possible. The remeshed triangles with locally isotropic property benefit various geometric processes such as neighbor-based feature extraction and analysis. The experimental results show that our method achieves better balance between geometric feature preservation and mesh quality improvement compared to peers. We provide the implementation codes of our resampling method at github.com/vvvwo/Adaptively-Isotropic-Remeshing.

AGD-GAN: Adaptive Gradient-Guided and Depth-supervised generative adversarial networks for ancient mural sketch extraction

To address the overlooked issues of multi-scale detail feature extraction and disease noise suppression in mural sketch extraction, we proposed a novel generative adversarial network with Adaptive Gradient-guided and Depth-supervised (AGD-GAN), which can generate high-quality mural sketches in an unsupervised manner. AGD-GAN first enriches mural feature details at various scales by introducing a cross-channel residual attention module, significantly improving the detail extraction effects. The Adaptive Gradient-guided strategy is based on the gradient attention maps, which can adaptively adjust the weights between gradient information and detail feature preservation according to the degree of damage in the mural, further balancing the preservation of mural detail features and the suppression of disease noise. Finally, the Depth-supervised reinforces constraints on the position and shape of the sketches by introducing a depth-predicted loss function, thereby reducing background noise interference and controlling the shape of the generated sketches. We compared with eight state-of-the-art algorithms quantitatively and qualitatively, experimental results demonstrate the promising capability of the proposed AGD-GAN to extract clear, coherent, and comprehensive mural sketches. We release the source code at https://github.com/Alice77bai/AGD-GAN.

Improving image quality via Deep learning: Aspects of speckle noise reduction applying the optimal weighted filter

This article explores the dynamic field of image denoising, with a specific emphasis on the novel Adaptive Euclidean Weighted Filters (AEWF) and their utilization in mitigating speckle noise. The inadequacy of conventional filters to handle the intricacies of contemporary applications has prompted researchers to investigate the potential benefits of integrating deep learning and optimization algorithms. By combining Convolutional Neural Networks (CNNs) and the Bat Optimization Algorithm (BOA), we develop an adaptive dynamic AEWF that prioritizes the reduction of speckle noise while accommodating various image attributes. By emulating the echolocation behavior of bats, BOA optimizes denoising parameters to guarantee outputs of superior quality. Convolutional and ReLU layers of CNNs are crucial during the denoising phase, while pooling layers are utilized to reduce the size of the image. Our comprehensive investigation contrasts the suggested methodology with traditional filters, evaluating its efficacy in mitigating speckle noise. The benchmarking metrics, namely the Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR), have respective mean values of 32.05 and 0.98. Through an exhaustive evaluation of feature preservation, noise reduction, and overall image quality, this study underscores the integrated AEWF's effectiveness in tackling current obstacles in the field of image processing.

Surface remeshing with preservation of sharp features through iterative identification and optimization of sample points

High-quality triangular meshes with sharp features are necessary for geometric processing applications, e.g., CAD models. Existing approaches either directly utilize the original sharp features as outputs, or predetermine feature sample points and reconstruct feature edges after an optimization process. However, these may result in poor-quality triangular facets close to feature edges or even feature preservation failure. With this regard, we present a feature-preservation surface remeshing method to generate high-quality triangular meshes with sharp feature preservation from raw manifold meshes. Unlike existing predetermination strategy, we introduce a heuristic strategy to dynamically determine feature sample points in each iteration. Based on the projection distance, each feature sample point is identified and projected onto the original feature edges, achieving the preservation of feature sample points. Besides, to eliminate the large gaps between two adjacent feature sample points and edge-close sample points, a feature edge continuity preservation strategy is proposed. Some additional sample points are further projected onto the large gaps and the edge-close sample points are deviated from the feature edges, achieving the preservation of reconstructed feature edges. Experimental results on several models demonstrate the effectiveness and efficiency of our method in generating high-quality triangular meshes with sharp feature preservation.

Feature-preserving shrink wrapping with adaptive alpha

Recent advancements in shrink-wrapping-based mesh approximation have shown tremendous advantages for non-manifold defective meshes. However, these methods perform unsatisfactorily when maintaining the regions with sharp features and rich details of the input mesh. We propose an adaptive shrink-wrapping method based on the recent Alpha Wrapping technique, offering improved feature preservation while handling defective inputs. The proposed approach comprises three main steps. First, we compute a new sizing field with the capability to assess the discretization density of non-manifold defective meshes. Then, we generate a mesh feature skeleton by projecting input feature lines onto the offset surface, ensuring the preservation of sharp features. Finally, an adaptive wrapping approach based on normal projection is applied to preserve the regions with sharp features and rich details simultaneously. By conducting experimental tests on various datasets including Thingi10k, ABC, and GrabCAD, we demonstrate that our method exhibits significant improvements in mesh fidelity compared to the Alpha Wrapping method, while maintaining the advantage of manifold property inherited from shrink-wrapping methods.

A Serial Multi-Scale Feature Fusion and Enhancement Network for Amur Tiger Re-Identification.

The Amur tiger is an important endangered species in the world, and its re-identification (re-ID) plays an important role in regional biodiversity assessment and wildlife resource statistics. This paper focuses on the task of Amur tiger re-ID based on visible light images from screenshots of surveillance videos or camera traps, aiming to solve the problem of low accuracy caused by camera perspective, noisy background noise, changes in motion posture, and deformation of Amur tiger body patterns during the re-ID process. To overcome this challenge, we propose a serial multi-scale feature fusion and enhancement re-ID network of Amur tiger for this task, in which global and local branches are constructed. Specifically, we design a global inverted pyramid multi-scale feature fusion method in the global branch to effectively fuse multi-scale global features and achieve high-level, fine-grained, and deep semantic feature preservation. We also design a local dual-domain attention feature enhancement method in the local branch, further enhancing local feature extraction and fusion by dividing local feature blocks. Based on the above model structure, we evaluated the effectiveness and feasibility of the model on the public dataset of the Amur Tiger Re-identification in the Wild (ATRW), and achieved good results on mAP, Rank-1, and Rank-5, demonstrating a certain competitiveness. In addition, since our proposed model does not require the introduction of additional expensive annotation information and does not incorporate other pre-training modules, it has important advantages such as strong transferability and simple training.

A dense and U-shaped transformer with dual-domain multi-loss function for sparse-view CT reconstruction.

CT image reconstruction from sparse-view projections is an important imaging configuration for low-dose CT, as it can reduce radiation dose. However, the CT images reconstructed from sparse-view projections by traditional analytic algorithms suffer from severe sparse artifacts. Therefore, it is of great value to develop advanced methods to suppress these artifacts. In this work, we aim to use a deep learning (DL)-based method to suppress sparse artifacts. Inspired by the good performance of DenseNet and Transformer architecture in computer vision tasks, we propose a Dense U-shaped Transformer (D-U-Transformer) to suppress sparse artifacts. This architecture exploits the advantages of densely connected convolutions in capturing local context and Transformer in modelling long-range dependencies, and applies channel attention to fusion features. Moreover, we design a dual-domain multi-loss function with learned weights for the optimization of the model to further improve image quality. Experimental results of our proposed D-U-Transformer yield performance improvements on the well-known Mayo Clinic LDCT dataset over several representative DL-based models in terms of artifact suppression and image feature preservation. Extensive internal ablation experiments demonstrate the effectiveness of the components in the proposed model for sparse-view computed tomography (SVCT) reconstruction. The proposed method can effectively suppress sparse artifacts and achieve high-precision SVCT reconstruction, thus promoting clinical CT scanning towards low-dose radiation and high-quality imaging. The findings of this work can be applied to denoising and artifact removal tasks in CT and other medical images.

ADMM-1DNet: Online Monitoring Method for Outdoor Mechanical Equipment Part Signals Based on Deep Learning and Compressed Sensing

To solve the problem that noise seriously affects the online monitoring of parts signals of outdoor machinery, this paper proposes a signal reconstruction method integrating deep neural network and compression sensing, called ADMM-1DNet, and gives a detailed online vibration signal monitoring scheme. The basic approach of the ADMM-1DNet network is to map the update steps of the classical Alternating Direction Method of Multipliers (ADMM) into the deep network architecture with a fixed number of layers, and each phase corresponds to an iteration in the traditional ADMM. At the same time, what differs from other unfolded networks is that ADMM-1DNet learns a redundant analysis operator, which can reduce the impact of outdoor high noise on reconstruction error by improving the signal sparse level. The implementation scheme includes the field operation of mechanical equipment and the operation of the data center. The empirical network trained by the local data center conducts an online reconstruction of the received outdoor vibration signal data. Experiments are conducted on two open-source bearing datasets, which verify that the proposed method outperforms the baseline method in terms of reconstruction accuracy and feature preservation, and the proposed implementation scheme can be adapted to the needs of different types of vibration signal reconstruction tasks.

CenterADNet: Infrared Video Target Detection Based on Central Point Regression.

Infrared video target detection is a fundamental technology within infrared warning and tracking systems. In long-distance infrared remote sensing images, targets often manifest as circular spots or even single points. Due to the weak and similar characteristics of the target to the background noise, the intelligent detection of these targets is extremely complex. Existing deep learning-based methods are affected by the downsampling of image features by convolutional neural networks, causing the features of small targets to almost disappear. So, we propose a new infrared video weak-target detection network based on central point regression. We focus on suppressing the image background by fusing the different features between consecutive frames with the original image features to eliminate the background's influence. We also employ high-resolution feature preservation and incorporate a spatial-temporal attention module into the network to capture as many target features as possible and improve detection accuracy. Our method achieves superior results on the infrared image weak aircraft target detection dataset proposed by the National University of Defense Technology, as well as on the simulated dataset generated based on real-world observation. This demonstrates the efficiency of our approach for detecting weak point targets in infrared continuous images.