Positron Emission Tomography (PET) is a critical imaging modality in nuclear medicine but requires radioactive tracer administration, which increases radiation exposure risks. While recent studies have investigated MR-guided low-dose PET denoising, they neglect two critical factors: the synergistic roles of multicontrast MR images and disease-specific denoising requirements. In this work, we propose a diffusion model that integrates T1-weighted, T2 fluid attenuated inversion recovery (T2 FLAIR), and hippocampal-optimized (T2 HIPPO) MR sequences to achieve ultra-low-dose PET denoising tailored for temporal lobe epilepsy (TLE). Our parallel cross-modal fusion (PCMF) module employs dedicated encoders to extract cross-modal features-which are dynamically integrated via attention mechanisms. Extensive experiments demonstrate that our method outperforms other approaches in preserving image quality. The PSNR and SSIM obtained were 37.0251 $\pm$ 1.5215 dB and 0.9760 $\pm$ 0.0057 (p < 0.01). Compared to the PET-only baseline model (IDDPM), our method achieved improvements of 8.4% in PSNR and 1.7% in SSIM, particularly excelling in diagnostically relevant temporal and hippocampal regions. This method provides a novel pathway for disease-specific PET denoising and has the potential to be generalized to other neurodegenerative diseases.

Hardware Trojans may cause security issues in cyber-physical systems (CPSs), and recently proposed mutual auditing frameworks have helped build trustworthy CPSs with untrustworthy devices by requiring neighboring devices from different vendors. However, this may cause severe multi-vendor integration challenges, such as expensive, hard-to-maintain, and insufficient vendors to purchase devices. In this work, we improve the mutual auditing framework by maintaining the security of the CPSs with fewer vendors. First, the vendor-constrained security auditing framework is introduced to enhance the security of the CPS network with limited vendors, where side auditing detects the hardware Trojan collusion between neighboring nodes and infected node isolation stops the spread of active HTs. Second, a multi-agent cooperative reinforcement learning-based method is proposed to assign devices with proper vendors in the context of security auditing, and it provides solutions with a minimized number of offline nodes due to the HT infection. The experimental results show that our proposed method reduces the number of vendors needed by 40.95%, and only causes an increment of 0.39% infected nodes.

ArrayInSAR has three-dimensional resolution capability, which can obtain three-dimensional information of the observation scene with a single flight, solving the problem of layover in InSAR mapping. In the process of ArrayInSAR data imaging, due to factors such as inconsistent antenna patterns, baseline errors, multipath effects, etc., there are noise points in the obtained 3D point cloud data of the observation scene, which reduces the quality of the 3D imaging results. This article proposes an ArrayInSAR point cloud denoising method based on an improved hierarchical moving surface fitting and adaptive threshold(AD-IHMSFA). Firstly, preprocess the point cloud data to remove discrete outliers in the air. Then, the improved hierarchical moving surface fitting algorithm is used to extract ground information from ArrayInSAR point cloud data. Finally, the DEM obtained from ground information interpolation is used to remove noise points from the SAR point cloud. The experimental results show that compared with traditional point cloud denoising methods, our method can effectively remove noise points in ArrayInSAR point cloud data and improve the accuracy of ArrayInSAR 3D reconstruction results.

The intent of this paper is to make an attempt at a large-scale forest height estimation technique framework construction via multi-data fusion of Quad-Pol TanDEM-X and GEDI Data. Faced with complex forest scenes, such as diverse tree species, sparse forest distribution and complex terrain, the authors design a strategy by fusing the traditional Random Volume over Ground (RVoG) model with ancillary forest height information from GEDI observations using BP neural network trained by sparse GEDI samples. To guarantee the accuracy for canopy height retrievals, the vertical wavenumber, as the critical parameter for RVoG construction, are refined by GEDI samples. This a large-scale forest height estimation technique framework is implemented over sparse forest in Danling county of China using quad-Pol TanDEM-X and GEDI Data. The results show that the proposed technique framework can meet the requirement of a large-scale forest height mapping in a resolution of 10m, with an average accuracy of 3.80 m, which is better than that of the traditional three-stage inversion method based on RVoG model.

Failure recovery is one of the most essential problems in Internet of Things (IoT) systems, and the conventional snapshot method is an effective way to solve this problem. However, snapshot methods lack specialized designs for heterogeneous IoT devices, and when implemented in edge devices, serious system interruptions occur and performance is impacted. To address these problems, a dynamic checkpointing strategy is proposed for IoT systems that consist of heterogeneous devices. Firstly, an anomaly detection network for snapshots (i.e., ADSnet) that combines long short-term memory networks with multilayer convolutional networks is used to learn the multidimensional features of system resource usage. Secondly, ADSnet is tuned during deployment to learn the behaviors of target devices, so that ADSnet can report the anomalies of target devices in the near future. Finally, a dynamic checkpointing strategy is proposed to dynamically create snapshots on the basis of the anomaly detection results. The experimental results show that the proposed ADSnet achieves 97.73% accuracy in detecting anomalies in the target device; furthermore, our proposed dynamic checkpointing strategy reduces 25.4% snapshots than that created by the recently proposed ResCheck.

Anti-islanding detection (AILD) for distributed power sources plays an important role on the stable operation of power grid systems and the safety of electrical systems. In order to improve the detection accuracy, we propose neural network architecture search (NAS) based approach for anti-islanding image detection and optimization of distributed power sources. It combines the fast region-based convolutional neural network (Faster-R-CNN) with the differentiable architecture search (Darts), utilizing electrical signal image data acquired from cameras or photovoltaic (PV) inverters, thereby enhancing the accuracy and efficiency of detection. We conduct a comparative analysis of two methods for obtaining distributed power supply (DPS) anti-islanding image data: camera-based acquisition and PV inverter-based acquisition. Our observation reveals that the image data acquired through cameras is more conducive for the learning process of DCNN. The proposed algorithm was compared with other convolutional neural network (CNN) models, validating its performance superiority. This provides a novel perspective and methodology for the advancement of AILD for distributed power sources, and enhances the security and stability of DPS.

Many algorithms have been proposed and studied for the registration of data acquired by light detection and ranging (LiDAR) and photogrammetry techniques. However, these algorithms cannot achieve accurate performance for data obtained from the array-interferometric synthetic aperture radar system (array-InSAR) because of the sampling effect of array-InSAR data. To address the registration problem of array-InSAR 3-D data, we propose a local intensity gradient (LIG)-based keypoint detector, along with an intensity-based feature descriptor in this paper. Then, the transformation model can be calculated based on the matched keypoints. We applied our proposed algorithm to real-life array-InSAR data and evaluated its accuracy by manually selecting corresponding points. The experimental results demonstrate that our algorithm improves the relative accuracy (root mean square error, RMSE) from 2.24 meters to 0.20 meters, indicating its effectiveness in improving registration accuracy for array-InSAR data.

The uncertainty of crop phenological cycle is an important issue in crop classification with time series PolSAR data. The time series alignment algorithm represented by dynamic time warping (DTW) can supply a potential solution, which realigns curves based on shape matching, dealing with the distortion of feature curves caused by uncertain crop phenological development. However, previous studies mainly focused on shape characteristics of time-varying feature curves, which is hard to comprehensively evaluate the similarity degree of crop phenological cycles. Furthermore, it ignored the differences in scattering signal and polarimetric statistical distribution of crops, which limited the accuracy of crop classification. In this letter, a novel crop classification method based on phenology alignment is proposed. Firstly, the dual-branch time series alignment method is proposed, including the time-weighted dynamic time warping (TWDTW) alignment and the Wishart distance-based TWDTW (WD-TWDTW) alignment, which combines the feature curve characteristics and the polarimetric statistical information to correctly describe the similarity degree of phenological cycles. Secondly, a multi-similarity measure (including shape similarity, feature similarity and polarimetric similarity) is defined to improve capacity of crop discrimination. The multi-similarity measure can describe the differences of crop types from three aspects, including crop growth trend, growth status, and statistical distribution. The proposed method is evaluated with time series full-polarization Radarsat-2 data in Flevoland area. The results show that our method is superior to traditional method with single TWDTW alignment and shape similarity, and the corresponding overall accuracy is improved by 6%.

With the launch of various multipolarimetric satellites, many scholars have introduced the persistent scatterer (PS)-oriented polarimetric optimization methods and extended the persistent scatterer interferometry (PSI) method to multipolarimetric data configuration, called polarimetric PSI (PolPSI) technology. Most PolPSI methods mainly take the amplitude dispersion index (ADI) as the optimization criterion and evaluate the temporal amplitude stationarity of each polarimetric channel for finding an optimal one. However, due to the unstable statistical characteristics of the quality indicator, many non-PS pixels are easily mistaken for the PS candidates (PSCs), and the performance of interferometric phase optimization is also limited. To overcome these restrictions, in this article, a novel PolPSI method is proposed based on the following two improved innovations. First, in terms of PSC selection, the trace moment (TM)-based statistical properties of time-series polarimetric coherency matrices are utilized for selecting the scatterers with the temporal polarimetric stationarity. Second, in terms of interferometric phase optimization, all interferometric coherency matrices of multipolarization channels are added up together to construct the total power (TP) interferogram for suppressing the effect of speckle noise and decorrelation. In the experiment, 13 scenes of quad-polarization ALOS PALSAR-1 image are selected to verify the algorithm’s effectiveness. The experimental results demonstrate that the proposed PolPSI method can better improve the deformation monitoring performance in three aspects than both the single-polarimetric HH and traditional exhaustive search polarimetric optimization (ESPO) methods, including phase quality improvement, density of PSs, and computational efficiency.

Synthetic aperture radar (SAR) coherent change detection (CCD) predominantly uses the degree of coherence or similar change metrics as a measurement of the changes that have occurred between two data collections. Many existing coherence estimators have shown some performances for change detection, but are still relatively limited because the change areas do not stand out well from all decorrelation areas due to low clutter-to-noise ratio (CNR) and volume scattering. Besides, many estimators require the equal-variance assumption between two SAR images of the same scene. However, the assumption is less likely to be met in areas with significant intensity differences, especially in change regions. In this paper, a novel weighted coherence estimator is proposed to address these problems. The estimator is derived based on the statistical characteristics of SAR images by using the maximum-likelihood (ML) principle. The introduction of weight parameters not only makes the estimator no longer need to satisfy the equal-variance assumption but also combines the advantages of coherent and noncoherent algorithms to a certain extent because the weights are closely related to the ratio change statistic. Experiments on simulated and real SAR image pairs demonstrate the effectiveness of the proposed estimator in highlighting change areas and the boundaries between change and other areas in theory and practice.

As an emerging molecular imaging modality, Photoacoustic Tomography (PAT) is capable of mapping tissue physiological metabolism and exogenous contrast agent information with high specificity. Due to its ultrasonic detection mechanism, the precise localization of targeted lesions has long been a challenge for PAT imaging. The poor soft-tissue contrast of the PAT image makes this process difficult and inaccurate. To meet this challenge, in this study, we first make use of the rich and clear structural information brought about by another advanced imaging modality, Magnetic Resonance Imaging (MRI), to assist organ segmentation and correct for the light fluence attenuation of PAT. We demonstrate improved feature visibility and enhanced localization of endogenous and exogenous agents in the fluence corrected PAT images. Compared with PAT-based methods, the contrast-to-noise ratio (CNR) of our MRI-assisted method increases by 29.1% in live animal experiments. Furthermore, we show that the co-registered MRI image can also be incorporated into PAT image restoration, and achieves improved anatomical landscape and soft-tissue contrast (CNR increased by 25.36%) while preserving similar spatial resolution. This PAT-MRI combination provides excellent structural, functional and molecular images of the subject, and may enable more comprehensive analysis of various preclinical research applications.

To characterize intra-tumor heterogeneity comprehensively, we propose a multi-level fusion strategy to combine PET and CT information at the image-, matrix-and feature-levels towards improved prognosis. Specifically, we developed fusion radiomics in the context of 3 prognostic outcomes in a multi-center setting (4 centers) involving 296 head & neck cancer patients. Eight clinical parameters were first utilized to build a (1) clinical model. We also built models by extracting 127 radiomics features from (2) PET images alone; (3-8) PET and CT images fused via wavelet-based fusion (WF) using CT-weights of 0.2, 0.4, 0.6 and 0.8, gradient transfer fusion (GTF), and guided filtering-based fusion (GFF); (9) fused matrices (sumMat); (10-11) fused features constructed via feature averaging (avgFea) and feature concatenation (conFea); and finally, (12) CT images alone; above models were also expanded to include both clinical and radiomics features. Seven variations of training and testing partitions were investigated. Highest performance in 5, 6 and 5 partitions was achieved by image-level fusion strategies for RFS, MFS and OS prediction, respectively. Among all partitions, WF0.6 and WF0.8 showed significantly higher performance than CT model for RFS (C-index: 0.60 ± 0.04 vs. 0.56 ± 0.03, p-value: 0.015) and MFS (C-index: 0.71 ± 0.13 vs. 0.62 ± 0.08, p-value: 0.020) predictions, respectively. In partition CER 23 vs. 14, WF0.6 significantly outperformed Clinical model for RFS prediction (C-index: 0.67 vs. 0.53, p-value: 0.003); both avgFea and WF0.6 showed C-index of 0.64 and significantly higher than that of PET only (C-index: 0.51, p-value: 0.018 and 0.031, respectively) for OS prediction. Fusion radiomics modeling showed varying improvements compared to single modality models for different outcome predictions in different partitions, highlighting the importance of generalizing radiomics models. Image-level fusion holds potential to capture more useful characteristics.

Conducting an accurate motion correction of liver dynamic contrast-enhanced magnetic resonance (DCE-MR) imaging remains challenging because of intensity variations caused by contrast agents. Such variations lead to the failure of the traditional intensity-based registration method. To address this problem, we propose a correlation-weighted sparse representation framework to separate the contrast agent from original liver DCE-MR images. This framework allows the robust registration of motion components over time without intensity variances. Existing sparse coding techniques recover a 3D image containing only contrast agents (named contrast enhancement component) from a manually labeled dictionary, whose column has the same size with the original 3D volume (3D-t mode). The high dimension of the recovery target (3D volume) and the indistinguishability between the unenhanced and enhanced images make accurate coding difficult. In this paper, we predefine an ideal time-intensity curve containing only contrast agents (named contrast agent curve) and recover it from the transpose dictionary (t-3D mode), whose column has been updated into the original time-intensity curves. The low dimension of the target (1D curve) and the significant intergroup difference between contrast agent curves and non-contrast agent curves can estimate a series of pure contrast agent curves. A "correlation-weighted" constraint is introduced for the selection of a coding subset with more contrast agent curves, leading to an efficient and accurate sparse recovery process. Then, the contrast enhancement component can be estimated by the solved sparse coefficients' map and the ideal curve and subtracted from the original DCE-MRI. Finally, we register the de-enhanced images and apply the obtained deformation fields for the original DCE-MRI to achieve the goal of motion correction. We conduct the experiments on both simulated and real liver DCE-MRI data. Compared with other state-of-the-art DCE-MRI registration methods, the experimental results show that our method achieves a better registration performance with less computational efficiency.

This paper proposes a fast, high-precision of multi-scale C-V model (M-C-V) method for coastline extraction with SAR imagery. (1) The multi-scale technology to the traditional C-V model is introduced that reduces the image size and obtains a series of images under different spatial resolution; (2) Low-pass filter polishes small scale image sequence is used, which is easy to form the image sequence with the relatively smooth boundary; (3) The image sequence in different scales and polished degree, splits the coastline based on the C-V model of level set one by one. In the division of coastline, high spatial resolution image inherits the boundary which is extracted by low spatial resolution images in the higher level and refines coastline further through the C-V model. Experiments show that the method accelerates the acquisition of initial level set formation, shortens the time of the extraction of coastline, at the same time, removes the non-coastline body part and improves the identification precision of the main body coastline, which makes the extracted process of coastline robust.

The intent of this paper is to propose the method of dual-aspect radiometric and geometric terrain correction to solve the two problems, radiometric variation and radiometric information loss. Firstly, the radiometric terrain correction method which considered gamma naught based image plane as measure of backscattering coefficient, which can correct radiometric variation of PolSAR images at the facing slope and back slope areas. Secondly, two color PolSAR images with dual opposite viewing angle are mutually fused by replacing invalid radiometric value with valid radiometric value from the opposite viewing image, which can generate integral radiometric information for orthophoto creation.

In this letter, a comprehensive study of the mapping of building collapse levels by the use of postearthquake synthetic aperture radar (SAR) images is addressed. Although previous studies have successfully quantified the collapse level by the use of postevent SAR, the types of features that are of benefit to the final accuracy still remain unknown. This letter takes the Yushu earthquake as a case study to contribute in two key areas. First, the Chinese dual-band airborne SAR mapping system (CASMSAR), which collected very-high-resolution X-band and P-band images (0.50 and 1.1 m, respectively) by interferometric and polarimetric modes, is comprehensively evaluated for the first time. Second, to discriminate intact and fallen structures, multiconfiguration SAR data features are analyzed, including 40 polarimetric, 3 interferometric, and 138 texture features. Furthermore, a random-forest decision framework is introduced to quantify the importance score of each feature and to improve the discrimination accuracy. The CASMSAR experiment results indicate that the texture is recommended as a powerful input to quantify the extent of building collapse in Yushu.

In this letter, a uniform scale-invariant feature transform (SIFT)-like algorithm is proposed for synthetic aperture radar (SAR) image registration, which can extract enough robust, reliable, and uniformly distributed features by the strategies of optimal feature selection based on a Voronoi diagram and feature scale-space proportional extraction. SAR images, taken from different viewpoints by an airborne sensor and at different times by spaceborne sensors, were used as test data to validate the effectiveness of the proposed algorithm. The indexes of local density and global coverage were used to assess the spatial distribution of matches. Compared with the traditional SIFT-like algorithm for SAR images (SAR-SIFT), the results show that the proposed algorithm can increase the number of matches and optimize their spatial distribution.

A new approach is presented in this paper to produce Digital Elevation Model (DEM) in mountainous areas with steep slope using ERS-1/2 tandem data. In order to reduce the impact of phase errors on the Interferometric Synthetic Aperture Radar (InSAR)-generated DEM, an external DEM such as that from Shuttle Radar Topography Mission (SRTM) is utilized in this approach. The proposed algorithm includes two steps: The first step is to model and remove phase trends with a linear regression analysis before converting phase to height; the second step is to filter unreliable height points before interpolating the DEM from the InSAR height map. The critical points are the following: 1) determining the one-to-one correspondence between the interferogram and the SRTM DEM before knowing the InSAR-derived elevation values and 2) estimating the elevation range of every pixel from SRTM DEM. To solve the first problem, an iteratively geocoding algorithm is performed. A DEM interpolation error model solves the second one. For InSAR data processing, the SRTM DEM is not only usable for modeling systematic phase errors but also for filtering gross height errors. The experiments in Zhangbei and the Three Gorges areas in China show that our approach has improved the accuracy of the resulting DEMs significantly without any ground control points.

This paper examines the synchronization of turbo coded modulation systems at low SNR where these systems are designed to operate. Maximum likelihood estimation theory is used to derive a causal, discrete-time synchronization structure for estimating the carrier phase and symbol timing of a general QAM signal, using a decision-directed structure which is known to be superior at low SNR. Simulation shows that efficient joint carrier phase and symbol timing estimates can be acquired using our synchronizer for turbo codes. Bit error performance for selected loop bandwidths is presented.