Edge Artifacts Research Articles

Role of Lesion-to-Background Activity Ratio and Background Activity in Optimization of Reconstruction Protocols for FDG PET/CT Images of Overweight Patients: A Phantom Study

Background: The quality of positron emission tomography/computed tomography (PET/CT) images plays an important role in tumor detection. This imaging method often yields poor-quality images of overweight patients due to the high level of noise, originating from scattering and photon attenuation. Objectives: The point spread function (PSF) is mostly used to enhance the spatial resolution and signal-to-noise ratio (SNR); however, it is known to increase the edge artifacts. The time-of-flight (TOF) principle can reduce edge artifacts in PSF modeling and improve lesion detection, especially in the thorax. The present study aimed to assess these two new techniques by applying different reconstruction parameters. Materials and Methods: An in-house phantom with an inner diameter of 35 cm was used for the simulation of overweight patients. Lesion-to-background ratios (LBRs) of 2: 1 and 8: 1, as well as background activity concentrations of 3 and 5 kBq/cc, were considered in this study. The list-mode data were reconstructed with various reconstruction protocols, numbers of subsets, and filter sizes. Quantitative analyses, including the coefficient of variation (COV), SNR, and recovery coefficient (RC), were also carried out. Moreover, box-and-whisker plots were performed. Results: At LBR of 2: 1, by changing the protocol from ordered subset expectation maximization (OSEM) to OSEM + PSF + TOF, the median value of SNR for 13-mm lesions (37 mm) increased by 39.25% and 53.45% (42.22% and 56.21%), at background activity concentrations of 3 and 5 kBq/cc respectively. However, at LBR of 8: 1, the corresponding values were 33.22% and 48.94% (40.22% and 52.15%) at background activity concentrations of 3 and 5 kBq/cc respectively. Conclusion: The TOF protocols were strongly recommended for both background activity concentrations at LBR of 2: 1 and for the low background activity concentration at LBR of 8: 1, especially when using smaller filter sizes. Moreover, subset numbers of 18 and 24 were appropriate for all protocols. However, a smaller subset number was suitable when a low background activity concentration and a smaller filter size were applied, especially at a lower LBR.

Reconstructing Floorplans from Point Clouds Using GAN

This paper proposed a method for reconstructing floorplans from indoor point clouds. Unlike existing corner and line primitive detection algorithms, this method uses a generative adversarial network to learn the complex distribution of indoor layout graphics, and repairs incomplete room masks into more regular segmentation areas. Automatic learning of the structure information of layout graphics can reduce the dependence on geometric priors, and replacing complex optimization algorithms with Deep Neural Networks (DNN) can improve the efficiency of data processing. The proposed method can retain more shape information from the original data and improve the accuracy of the overall structure details. On this basis, the method further used an edge optimization algorithm to eliminate pixel-level edge artifacts that neural networks cannot perceive. Finally, combined with the constraint information of the overall layout, the method can generate compact floorplans with rich semantic information. Experimental results indicated that the algorithm has robustness and accuracy in complex 3D indoor datasets; its performance is competitive with those of existing methods.

Synthetic aperture radar image despeckling neural network based on maximum a posteriori probability estimation

ABSTRACT Automatic despeckling from synthetic aperture radar (SAR) imagery plays a significant role in many urban applications. Recently, owing to the impressive performance of deep learning, various SAR despeckling methods based on the convolutional neural network (CNN) have been proposed for optical SAR images. However, existing CNN-based despeckling methods and spatial-domain-based methods exhibit certain limitations. The introduction of prior knowledge is not considered in the designing of classical denoising and certain CNN-based methods, thus leading to the loss of edge information and artefacts in the results. To address this issue, an a priori-based deep neural network SAR image despeckling method (SAR-PBDNN) was proposed. In this approach, the Bayesian formula is modelled to calculate the objective function suitable for the despeckling problem and is optimized using the alternating direction method of multipliers algorithm and deep learning method. The introduction of prior information effectively improved the denoising effect of SAR images. The SAR-PBDNN was evaluated using both simulated and real SAR images. The comparisons demonstrate the effectiveness of the proposed method. We conclude that the SAR-PBDNN has the potential to automatically despeckle SAR images with an accuracy that renders it a useful tool for practical application scenarios.

Smoothing of inter-layer edge artifacts in depth-map computer-generated holograms.

In the depth-map computer-generated hologram (CGH), inter-layer edge artifacts are observed in the discontinuous edges of section-wise depth-map objects. CGH synthesis, utilizing the hybrid smoothing method of silhouette masking and edge-apodization, alleviates unwanted inter-layer edge artifacts. The proposed method achieves improved de-artifact filtering that generates holographic images closer to the ground truth image of the depth-map object unattainable by the conventional CGH synthesis method.

Single image super‐resolution based on sparse representation using edge‐preserving regularization and a low‐rank constraint

Sparse representation-based non-local self-similarity approaches have demonstrated promising performance in single image super-resolution reconstruction. This type of method, however, cannot always effectively preserve the key details of an image, resulting in edge artifacts and local structure blurring. To better preserve the image edge information, in this paper, a sparse representation super-resolution method based on non-local self-similarity is proposed. First, we impose slide window gradient domain guided filtering on both the low-resolution input image and the degraded restored image. Then, we utilize their difference as an edge-preserving regularization term and incorporate this regularization term into the non-local self-similarity-based sparse representation model to build a sparse coding model, which can enhance the restored high-resolution image patches’ detail information. Finally, the iterative threshold algorithm is used to calculate the sparse representation coefficients so that a high-resolution image can be estimated. Furthermore, to explore the potential structures of the subspaces spanned by similar patches, we enforce a low-rank matrix recovery technique on the generated super-resolution image, which can further refine the reconstruction quality. Experimental results prove that the new approach preserves the critical edge structures while suppressing noise and exceeding some popular methods both quantitatively and qualitatively.

Towards Semantic Management of On-Device Applications in Industrial IoT

The Internet of Things (IoT) is revolutionizing the industry. Powered by pervasive embedded devices, the Industrial IoT (IIoT) provides a unique solution for retrieving and analyzing data near the source in real-time. Many emerging techniques, such as Tiny Machine Learning (TinyML) and Complex Event Processing (CEP) , are actively being developed to support decision making at the edge, shifting the paradigm from centralized processing to distributed computing. However, distributed computing presents management challenges, as IoT devices are diverse and constrained, and their number is growing exponentially. The situation is even more challenging when various on-device applications (so-called artifacts) are deployed across decentralized IoT networks. Questions to be addressed include how to discover an appropriate function, whether that function can be executed on a certain device, and how to orchestrate a cross-platform service. To tackle these challenges, we propose an approach for the scalable management of on-device applications among distributed IoT devices. By leveraging the W3C Web of Things (WoT) , the capabilities of each IoT device, or more precisely, its interaction patterns, can be semantically expressed in a Thing Description (TD) . In addition, we introduce semantic modeling of on-device applications to supplement an TD with additional information regarding applications on the device. Specifically, we demonstrate two examples of semantic modeling: neural networks (NN) and CEP rules. The ontologies are evaluated by answering a set of competency questions. By hosting the enriched semantic knowledge of the entire IoT system in a Knowledge Graph (KG) , we can discover and interoperate edge devices and artifacts across the decentralized network. This can reduce fragmentation and increase the reusability of IoT components. We demonstrate the feasibility of our concept on an industrial workstation consisting of a conveyor belt and several IoT devices. Finally, the requirements for constructing an IoT semantic management system are discussed.

Structure-Aware Color Halftoning with Adaptive Sharpness Control

Two-and-a-half and 3D printing are becoming increasingly popular, and consequently the demand for high quality surface reproduction is also increasing. Halftoning plays an important role in the quality of the surface reproduction. Three dimensional halftoning methods, that adapt the halftone structures to the geometrical structure of 3D surfaces or to the viewing direction, could further improve surface reproduction quality. In this paper, a 3D adaptive halftoning method is proposed, that incorporates different halftone structures on the same 3D surface. The halftone structures are firstly adapted to the 3D geometrical structure of the surface. Secondly, the halftone structures are adapted based on the normal vector to the surface at a specific voxel. Two simple approaches to approximate the normal vector are also proposed. The problem of edge artefacts that might occur in the previously proposed 3D Iterative Method Controlling the Dot Placement (IMCDP) halftoning method is discussed and a solution to reduce these artefacts is given. The results show that the proposed adaptive halftoning can combine different halftone structures on the same 3D surface with no transition artefacts between different halftone structures. It is also shown that using second-order frequency modulation (FM) halftone, in comparison to first-order FM, can result in more homogeneous appearance of 3D surfaces with undesirable structures on them.

A Pixel-Level Segmentation-Synthesis Framework for Dynamic Texture Video Compression

Dynamic textures are time-varying motion patterns that exhibit certain temporal stationarity. The motion of such patterns is usually anisotropic, which is a great challenge for video coding. In this paper, we firstly analyze the predictive and temporal characteristics of dynamic textures and, based on the analysis, a pixel-level segmentation-synthesis framework for dynamic texture compression is proposed to improve predictive-coding efficiency. This framework consists of three sub-modules: dynamic texture segmentation, dynamic texture synthesis and fusion process. A deep-learning-based segmentation method is introduced to obtain a pixel-level mask. Subsequently, optical flow fields are used to generate flowlines and help analyze the motion periodicity of a given sequence. The flowlines and period values are utilized as inputs for flow-based dynamic texture synthesis. The segmented dynamic texture area is then reconstructed using the synthesized results at the decoder side at pixel-level, such that the overhead for transmitting these contents could be saved. The segmented mask is transmitted through an All-Zero-Run-Length coding algorithm and an effective fusion module is further proposed to reduce edge artifacts that occur at the boundary of dynamic texture areas. Coding blocks have been classified into three types and a weighting reconstruction scheme is performed. The proposed framework has been integrated into VTM-10.0. Experimental results on a collected dynamic texture test set demonstrate that 15.1% and 16.1% MOS BD-rate savings can be achieved on average for LDB and LDP configurations, respectively.

Image manipulation detection by multiple tampering traces and edge artifact enhancement

Image manipulation detection has attracted considerable attention owing to the increasing security risks posed by fake images. Previous studies have proven that tampering traces hidden in images are essential for detecting manipulated regions. However, existing methods have limitations in generalization and the ability to tackle post-processing methods. This paper presents a novel Network to learn and Enhance Multiple tampering Traces (EMT-Net), including noise distribution and visual artifacts. For better generalization, EMT-Net extracts global and local noise features from noise maps using transformers and captures local visual artifacts from original RGB images using convolutional neural networks. Moreover, we enhance fused tampering traces using the proposed edge artifacts enhancement modules and edge supervision strategy to discover subtle edge artifacts hidden in images. Thus, EMT-Net can prevent the risks of losing slight visual clues against well-designed post-processing methods. Experimental results indicate that the proposed method can detect manipulated regions and outperform state-of-the-art approaches under comprehensive quantitative metrics and visual qualities. In addition, EMT-Net shows robustness when various post-processing methods further manipulate images.

Identifying Calcaneal Anatomical Regions of Interest (ROI) for Quantitative Ultrasound Application in Subadults

Whilst quantitative ultrasound can be reliably used to assess bone health in adults, the fixed location of the transducers in current devices may result in inaccurate and unreliable measurements in bone assessment in children due to the variation in foot size during growth. To improve positioning for paediatric assessment, Jaworski et al. (1995) created an anatomical method to identify the region of interest (ROI), however, there have been no medical imaging studies to confirm that the Jaworski method results in consistent placement of the transducer on the centre of the calcaneal body to avoid edge artefacts. In this study, computed tomography scans of the tarsus were collected from 498 individuals (258 females; 240 males) aged 2 to 20 years and used to create three novel anatomical methods to identify ROI on the calcaneus using palpable landmarks. In addition, the established Jaworski method was applied to the same scans and compared to our novel methods. The maximum ROI significantly increased with age with males having significantly greater diameters, supporting the recommendation that ½ inch diameter transducers should be used on individuals younger than 7 years of age. We identified that 79% of the ‘Jaworski points’ lied anterosuperior to the ROI centre point identified in this study, with 10% of the points lying outside the ROI. Of the three novel methods, only the calcaneal insertion method demonstrated small measurement variance between individuals of the same age in each sex and is therefore the preferred method for ultrasound clinical application.

CCTNet: Coupled CNN and Transformer Network for Crop Segmentation of Remote Sensing Images

Semantic segmentation by using remote sensing images is an efficient method for agricultural crop classification. Recent solutions in crop segmentation are mainly deep-learning-based methods, including two mainstream architectures: Convolutional Neural Networks (CNNs) and Transformer. However, these two architectures are not sufficiently good for the crop segmentation task due to the following three reasons. First, the ultra-high-resolution images need to be cut into small patches before processing, which leads to the incomplete structure of different categories’ edges. Second, because of the deficiency of global information, categories inside the crop field may be wrongly classified. Third, to restore complete images, the patches need to be spliced together, causing the edge artifacts and small misclassified objects and holes. Therefore, we proposed a novel architecture named the Coupled CNN and Transformer Network (CCTNet), which combines the local details (e.g., edge and texture) by the CNN and global context by Transformer to cope with the aforementioned problems. In particular, two modules, namely the Light Adaptive Fusion Module (LAFM) and the Coupled Attention Fusion Module (CAFM), are also designed to efficiently fuse these advantages. Meanwhile, three effective methods named Overlapping Sliding Window (OSW), Testing Time Augmentation (TTA), and Post-Processing (PP) are proposed to remove small objects and holes embedded in the inference stage and restore complete images. The experimental results evaluated on the Barley Remote Sensing Dataset present that the CCTNet outperformed the single CNN or Transformer methods, achieving 72.97% mean Intersection over Union (mIoU) scores. As a consequence, it is believed that the proposed CCTNet can be a competitive method for crop segmentation by remote sensing images.

Pitfalls" in the path of quantitative assessment of the severity of oncological lesions in diagnostic nuclear medicine

When examining patients by single-photon emission computed tomography, the distribution of a radiopharmaceutical drug in biological tissues is evaluated. An active fast-growing tumor has an increased uptake of the drug compared to healthy tissues. In recent years, there has been considerable interest in the level of activity of a radiopharmaceutical in tumor lesions, since it can be useful for monitoring the effectiveness of prescribed therapy. To obtain quantitative results, it is necessary that the applied reconstruction algorithms include the main effects that contribute to the formation of images. One of these factors is the geometric resolution of the collimator. It was expected that taking into account this effect would improve the quantitative assessment of tumor lesions in reconstructed images. However, there were "pitfalls" in the way of obtaining quantitative images. In the diagnostic images obtained taking into account the "geometric resolution of the collimator", so-called edge artifacts were observed, which manifested themselves in the form of a sharp increase in intensity of radiopharmaceutical accumulation in small tumors and as oscillations in the intensity of radiopharmaceutical accumulation at the edges in large tumors. The problem of edge artifacts casts doubt on the correctness of quantitative assessment the level of intensity of drug accumulation in tumor lesions. In this paper, the problem of edge artifacts is investigated using the method of mathematical modeling. In numerical experiments, the dependence of the quantitative estimate of the intensity of RPD accumulation on the parameters of the reconstruction algorithm was investigated for the first time. Keywords: Nuclear medicine, mathematical modeling, single-photon emission computed tomography, edge artifacts.

Подводные камни" на пути количественной оценки тяжести онкологических поражений в диагностической ядерной медицине

When examining patients by single-photon emission computed tomography, the distribution of a radiopharmaceutical drug in biological tissues is evaluated. An active fast-growing tumor has an increased uptake of the drug compared to healthy tissues. In recent years, there has been considerable interest in quantifying the level of activity of a radiopharmaceutical in tumor lesions, since it can be useful for monitoring the effectiveness of prescribed therapy. To obtain quantitative results, it is necessary that the applied reconstruction algorithms include the main effects that contribute to the formation of images. One of these effects is the "blurring" of the image of a point source in the form of a spot. It was expected that taking into account this effect would improve estimates. However, there were "pitfalls" in the way of obtaining quantitative images. In the clinical images obtained taking into account the "blurring effect", so-called edge artifacts were observed, which manifested themselves in the form of a sharp increase in intensity in small tumors and a surge in intensity at the edges in large tumors. The problem of edge artifacts casts doubt on the correctness of quantifying the level of drug activity in tumor lesions. In this paper, the problem of edge artifacts is investigated using the method of mathematical modeling. In numerical experiments, the dependence of the quantitative assessment of activity on the parameters of the reconstruction algorithm was investigated for the first time.

An iterative reconstruction method for sparse-projection data for low-dose CT.

Reducing X-ray radiation is beneficial for reducing the risk of cancer in patients. There are two main approaches for achieving this goal namely, one is to reduce the X-ray current, and another is to apply sparse-view protocols to do image scanning and projections. However, these techniques usually lead to degradation of the reconstructed image quality, resulting in excessive noise and severe edge artifacts, which seriously affect the diagnosis result. In order to overcome such limitation, this study proposes and tests an algorithm based on guided kernel filtering. The algorithm combines the characteristics of anisotropic edges between adjacent image voxels, expresses the relevant weights with an exponential function, and adjusts the weights adaptively through local gray gradients to better preserve the image structure while suppressing noise information. Experiments show that the proposed method can effectively suppress noise and preserve the image structure. Comparing with similar algorithms, the proposed algorithm greatly improves the peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and root mean square error (RMSE) of the reconstructed image. The proposed algorithm has the best effect in quantitative analysis, which verifies the effectiveness of the proposed method and good image reconstruction performance. Overall, this study demonstrates that the proposed method can reduce the number of projections required for repeated CT scans and has potential for medical applications in reducing radiation doses.

Sparse and continuous sampling approaches to fMRI of overt vocalization tasks

Sparse temporal sampling has become the dominant paradigm for functional magnetic resonance imaging studies of auditory stimuli or verbal responses, as it allows the presentation or production of stimuli during the relatively quiet periods when there is no gradient switching and ensures that task-related movements are not occurring during scan acquisitions. To date, however, there has been no direct comparison between sparse and continuous acquisition protocols for overt auditory-verbal studies (i.e., speaking or singing). The aim of this study was to determine whether sparse temporal sampling would reduce movement artefacts and show better network activation for overt singing compared to continuous imaging. Fifteen healthy adults performed the same overt singing task under both sparse and continuous scanning conditions. We noted significant variations in signal intensity between adjacent slices in our sparse acquisition, with (odd-numbered) slices acquired in the second half of each volume acquisition being of lower intensity and showing less reliable task-related activation, and thus requiring the removal of these slices prior to preprocessing. Edge artefacts, presumably due to movement, were observed in both acquisition types at a subthreshold level, although ventricular space artefacts were more apparent in the continuous data. However, statistical comparison revealed no significant differences in functional activation nor in motion correction parameters. Our results show that sparse imaging has the potential to introduce significant image artefacts affecting downstream analyses. While sparse sampling provides benefits that may be essential for certain studies (e.g., periods free from scanner noise), the technical requirements of such sequences should not be overlooked and inspection of raw data is essential. Our data also show that continuous imaging can be used for overt response auditory-verbal studies and may be of wider utility than previously appreciated.

Correcting Artifacts in Ratiometric Biosensor Imaging; an Improved Approach for Dividing Noisy Signals.

The accuracy of biosensor ratio imaging is limited by signal/noise. Signals can be weak when biosensor concentrations must be limited to avoid cell perturbation. This can be especially problematic in imaging of low volume regions, e.g., along the cell edge. The cell edge is an important imaging target in studies of cell motility. We show how the division of fluorescence intensities with low signal-to-noise at the cell edge creates specific artifacts due to background subtraction and division by small numbers, and that simply improving the accuracy of background subtraction cannot address these issues. We propose a new approach where, rather than simply subtracting background from the numerator and denominator, we subtract a noise correction factor (NCF) from the numerator only. This NCF can be derived from the analysis of noise distribution in the background near the cell edge or from ratio measurements in the cell regions where signal-to-noise is high. We test the performance of the method first by examining two noninteracting fluorophores distributed evenly in cells. This generated a uniform ratio that could provide a ground truth. We then analyzed actual protein activities reported by a single chain biosensor for the guanine exchange factor (GEF) Asef, and a dual chain biosensor for the GTPase Cdc42. The reduction of edge artifacts revealed persistent Asef activity in a narrow band (∼640 nm wide) immediately adjacent to the cell edge. For Cdc42, the NCF method revealed an artifact that would have been obscured by traditional background subtraction approaches.

Ultrasound segmentation-guided edge artifact reduction in diffuse optical tomography using connected component analysis.

Ultrasound (US)-guided diffuse optical tomography (DOT) has demonstrated potential value for breast cancer diagnosis and treatment response assessment. However, in clinical use, the chest wall, poor probe-tissue contact, and tissue heterogeneity can all cause image artifacts. These image artifacts, appearing commonly as hot spots in the non-lesion regions (edge artifacts), can decrease the reconstruction accuracy and cause misinterpretation of lesion images. Here we introduce an iterative, connected component analysis-based image artifact reduction algorithm. A convolutional neural network (CNN) is used to segment co-registered US images to extract the lesion location and size to guide the artifact reduction. We demonstrate its performance using Monte Carlo simulations on VICTRE digital breast phantoms and breast patient images. In simulated tissue mismatch models, this algorithm successfully reduces edge artifacts without significantly changing the reconstructed target absorption coefficients. With clinical data it improves the optical contrast between malignant and benign groups, from 1.55 without artifact reduction to 1.91 with artifact reduction. The proposed algorithm has a broad range of applications in other modality-guided DOT imaging.

A Novel Iterative MLEM Image Reconstruction Algorithm Based on Beltrami Filter: Application to ECT Images.

The implementation of emission-computed tomography (ECT), including positron emission tomography and single-photon emission-computed tomography, has been an important research topic in recent years and is of significant and practical importance. However, the slow rate of convergence and the computational complexity have severely impeded the efficient implementation of iterative reconstruction. By combining the maximum-likelihood expectation maximization (MLEM) iteratively along with the Beltrami filter, this paper proposes a new approach to reformulate the MLEM algorithm. Beltrami filtering is applied to an image obtained using the MLEM algorithm for each iteration. The role of Beltrami filtering is to remove mainly out-of-focus slice blurs, which are artifacts present in most existing images. To improve the quality of an image reconstructed using MLEM, the Beltrami filter employs similar structures, which in turn reduce the number of errors in the reconstructed image. Numerical image reconstruction tomography experiments have demonstrated the performance capability of the proposed algorithm in terms of an increase in signal-to-noise ratio (SNR) and the recovery of fine details that can be hidden in the data. The SNR and visual inspections of the reconstructed images are significantly improved compared to those of a standard MLEM. We conclude that the proposed algorithm provides an edge-preserving image reconstruction and substantially suppress noise and edge artifacts.

Standardisation of conventional and advanced iterative reconstruction methods for Gallium-68 multi-centre PET-CT trials

PurposeTo assess the applicability of the Fluorine-18 performance specifications defined by EANM Research Ltd (EARL), in Gallium-68 multi-centre PET-CT trials using conventional (ordered subset expectation maximisation, OSEM) and advanced iterative reconstructions which include the systems’ point spread function (PSF) and a Bayesian penalised likelihood algorithm (BPL) commercially known as Q.CLEAR. The possibility of standardising the two advanced reconstruction methods was examined.MethodsThe NEMA image quality phantom was filled with Gallium-68 and scanned on a GE PET-CT system. PSF and BPL with varying post-reconstruction Gaussian filter width (2–6.4 mm) and penalisation factor (200–1200), respectively, were applied. The average peak-to-valley ratio from six profiles across each sphere was estimated to inspect any edge artefacts. Image noise was assessed using background variability and image roughness. Six GE and Siemens PET-CT scanners provided Gallium-68 images of the NEMA phantom using both conventional and advanced reconstructions from which the maximum, mean and peak recoveries were drawn. Fourteen patients underwent 68Ga-PSMA PET-CT imaging. BPL (200-1200) reconstructions of the data were compared against PSF smoothed with a 6.4-mm Gaussian filter.ResultsA Gaussian filter width of approximately 6 mm for PSF and a penalisation factor of 800 for BPL were needed to suppress the edge artefacts. In addition, those reconstructions provided the closest agreement between the two advanced iterative reconstructions and low noise levels with the background variability and the image roughness being lower than 7.5% and 11.5%, respectively. The recoveries for all methods generally performed at the lower limits of the EARL specifications, especially for the 13- and 10-mm spheres for which up to 27% (conventional) and 41% (advanced reconstructions) lower limits are suggested. The lesion standardised uptake values from the clinical data were significantly different between BPL and PSF smoothed with a Gaussian filter of 6.4 mm wide for all penalisation factors except for 800 and 1000.ConclusionIt is possible to standardise the advanced reconstruction methods with the reconstruction parameters being also sufficient for minimising the edge artefacts and noise in the images. For both conventional and advanced reconstructions, Gallium-68 specific recovery coefficient limits were required, especially for the smallest phantom spheres.

A multi-focus image fusion method based on attention mechanism and supervised learning

Multi-focus image fusion is always a difficult problem in digital image processing. To achieve efficient integration, we propose a new end-to-end network. This network uses the residual atrous spatial pyramid pooling module to extract multi-level features from the space of different scales and share parameters to ensure the consistency and correspondence of features. We also introduced a disparities attention module for the network which allows for information retention. These two parts can make our method overcome the difficulties of target edge artifacts, small range blur, poor detail capture, and so on. In addition, in order to improve the semantic ambiguity easily caused by unsupervised learning, we also proposed a new multi-focus image fusion dataset with groundtruth for supervised learning. We performed sufficient experiments, and the results show that the network can quickly capture the corresponding features of multi-focus images, and improve the fusion performance with less computation and lower storage cost. Compared with the existing nine fusion methods, our network is superior to other methods in subjective visual evaluation and objective evaluation, reaching a higher level.