Image Noise Level Research Articles

Enhanced visualization in endoleak detection through iterative and AI-noise optimized spectral reconstructions

To assess the image quality parameters of dual-energy computed tomography angiography (DECTA) 40-, and 60 keV virtual monoenergetic images (VMIs) combined with deep learning-based image reconstruction model (DLM) and iterative reconstructions (IR). CT scans of 28 post EVAR patients were enrolled. The 60 s delayed phase of DECTA was evaluated. Objective [noise, contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR)] and subjective (overall image quality and endoleak conspicuity – 3 blinded readers assessment) image quality analyses were performed. The following reconstructions were evaluated: VMI 40, 60 keV VMI; IR VMI 40, 60 keV; DLM VMI 40, 60 keV. The noise level of the DLM VMI images was approximately 50% lower than that of VMI reconstruction. The highest CNR and SNR values were measured in VMI DLM images. The mean CNR in endoleak in 40 keV was accounted for as 1.83 ± 1.2; 2.07 ± 2.02; 3.6 ± 3.26 in VMI, VMI IR, and VMI DLM, respectively. The DLM algorithm significantly reduced noise and increased lesion conspicuity, resulting in higher objective and subjective image quality compared to other reconstruction techniques. The application of DLM algorithms to low-energy VMIs significantly enhances the diagnostic value of DECTA in evaluating endoleaks. DLM reconstructions surpass traditional VMIs and IR in terms of image quality.

Deep learning-accelerated image reconstruction in MRI of the orbit to shorten acquisition time and enhance image quality.

This study explores the use of deep learning (DL) techniques in MRI of the orbit to enhance imaging. Standard protocols, although detailed, have lengthy acquisition times. We investigate DL-based methods for T2-weighted and T1-weighted, fat-saturated, contrast-enhanced turbo spin echo (TSE) sequences, aiming to improve image quality, reduce acquisition time, minimize artifacts, and enhance diagnostic confidence in orbital imaging. In a 3-Tesla MRI study of 50 patients evaluating orbital diseases from March to July 2023, conventional (TSES ) and DL TSE sequences (TSEDL ) were used. Two neuroradiologists independently assessed the image datasets for image quality, diagnostic confidence, noise levels, artifacts, and image sharpness using a randomized and blinded 4-point Likert scale. TSEDL significantly reduced image noise and artifacts, enhanced image sharpness, and decreased scan time, outperforming TSES (p<.05). TSEDL showed superior overall image quality and diagnostic confidence, with relevant findings effectively detected in both DL-based and conventional images. In 94% of cases, readers preferred accelerated imaging. The study proved that using DL for MRI image reconstruction in orbital scans significantly cut acquisition time by 69%. This approach also enhanced image quality, reduced image noise, sharpened images, and boosted diagnostic confidence.

Assessing the robustness of a machine-learning model for early detection of pancreatic adenocarcinoma (PDA): evaluating resilience to variations in image acquisition and radiomics workflow using image perturbation methods.

To evaluate robustness of a radiomics-based support vector machine (SVM) model for detection of visually occult PDA on pre-diagnostic CTs by simulating common variations in image acquisition and radiomics workflow using image perturbation methods. Eighteen algorithmically generated-perturbations, which simulated variations in image noise levels (σ, 2σ, 3σ, 5σ), image rotation [both CT image and the corresponding pancreas segmentation mask by 45° and 90° in axial plane], voxel resampling (isotropic and anisotropic),gray-level discretization [bin width (BW) 32 and 64)], and pancreas segmentation (sequential erosions by 3, 4, 6, and 8 pixels and dilations by 3, 4, and 6 pixels from the boundary), were introduced to the original (unperturbed) test subset (n = 128; 45 pre-diagnostic CTs, 83 control CTs with normal pancreas). Radiomic features were extracted from pancreas masks of these additional test subsets, and the model's performance was compared vis-a-vis the unperturbed test subset. The model correctly classified 43 out of 45 pre-diagnostic CTs and 75 out of 83 control CTs in the unperturbed test subset, achieving 92.2% accuracy and 0.98 AUC. Model's performance was unaffected by a three-fold increase in noise level except for sensitivity declining to 80% at 3σ (p = 0.02). Performance remained comparable vis-a-vis the unperturbed test subset despite variations in image rotation (p = 0.99), voxel resampling (p = 0.25-0.31), change in gray-level BW to 32 (p = 0.31-0.99), and erosions/dilations up to 4 pixels from the pancreas boundary (p = 0.12-0.34). The model's high performance for detection of visually occult PDA was robust within a broad range of clinically relevant variations in image acquisition and radiomics workflow.

Validation of Optimal Imaging Conditions for Coronary Computed Tomography Angiography Using High-definition Mode and Deep Learning Image Reconstruction Algorithm

To verify the optimal imaging conditions for coronary computed tomography angiography (CCTA) examinations when using high-definition (HD) mode and deep learning image reconstruction (DLIR) in combination. A chest phantom and an in-house phantom using 3D printer were scanned with a 256-row detector CT scanner. The scan parameters were as follows - acquisition mode: ON (HD mode) and OFF (normal resolution [NR] mode), rotation time: 0.28 s/rotation, beam coverage width: 160 mm, and the radiation dose was adjusted based on CT-AEC. Image reconstruction was performed using ASiR-V (Hybrid-IR), TrueFidelity Image (DLIR), and HD-Standard (HD mode) and Standard (NR mode) reconstruction kernels. The task-based transfer function (TTF) and noise power spectrum (NPS) were measured for image evaluation, and the detectability index (d') was calculated. Visual evaluation was also performed on an in-house coronary phantom. The in-plane TTF was better for the HD mode than for the NR mode, while the z-axis TTF was lower for DLIR than for Hybrid-IR. The NPS values in the high-frequency region were higher for the HD mode compared to those for the NR mode, and the NPS was lower for DLIR than for Hybrid-IR. The combination of HD mode and DLIR showed the best value for in-plane d', whereas the combination of NR mode and DLIR showed the best value for z-axis d'. In the visual evaluation, the combination of NR mode and DLIR showed the best values from a noise index of 45 HU. The optimal combination of HD mode and DLIR depends on the image noise level, and the combination of NR mode and DLIR was the best imaging condition under noisy conditions.

A multi-tasking novel variational model for image decolorization and denoising

Image decolorization has a wide range of applications in digital printing, photo rendering and single-channel image processing. During the process of image decolorization, it is significant to retain more contrast information and avoid increasing the noise level of the original image. As we know, noise was hardly considered in previous decolorization methods. In this paper, we propose a new multi-tasking variational model with two tasks: decolorizing and denoising. The first task is image decolorization, that is, to transform color images to grayscale images. The other task is to control the noise level, and even eliminate the noise. Inspired by TV-Stokes model, we introduce the unit normal vector containing the basic geometry information of the color image in the model. The existence and the uniqueness of the solution to the variational problems are proved. For the variational model, two numerical methods are adopted. At first, we use gradient descent method to obtain evolution equations and construct finite difference schemes to solve the corresponding equations of proposed variational model. Furthermore, an efficient augmented Lagrangian approach is applied to solve the multi-tasking variational model directly. For noisy images and clean images, we employ two output patterns of grayscale images respectively. Especially, our model is stable for the noisy images. Numerical results and comparisons show that the proposed model can produce highly competitive results in terms of decolorization quality of noisy images, edge preservation and adjustment of parameters.

Adapting to the Dark: A Novel Adaptive Low Light Illumination Correction Algorithm for Video Sequences in Wireless Communications

The lower image quality and higher noise levels of low light images can have a substantial effect on wireless communication. Images typically have less brightness, less clarity, and more picture noise in low light situations. The precision and dependability of picture-based communication may be impacted by this decline in image quality. Low light levels can be detrimental to video chats since it becomes difficult for the camera to get clear pictures of the user's face. This may lead to decreased video quality, trouble with facial recognition, and a generally worse user experience when communicating over video. Low light illumination correction algorithms are frequently made to work quickly and effectively in order to enable real-time video-based wireless communication. Correcting uneven illumination or low light in images is crucial for various fields; as such images pose challenges for human recognition, computer vision algorithms, and multimedia algorithms. An adaptive illumination correction algorithm based on a simple log transformation approach enhanced by a hyperbolic beta transformation and a logarithmic image processing is presented in this research. It has an adaptive optimization function that adjusts itself based on the image’s illumination and reflectance. Low computation and minimal parameter optimization are the hallmarks of this method. A variety of illuminated images and video sequences from various datasets are used to evaluate the suggested adaptive illumination approach. This adaptive illumination approach's experimental findings are impressive compared to those produced using prior techniques. The results reveal that the proposed system can adapt to various image-based communication and surveillance video sequences taken under various illumination conditions.

Regression of the Rician Noise Level in 3D Magnetic Resonance Images from the Distribution of the First Significant Digit

This paper investigates the distribution characteristics of Fourier, discrete cosine, and discrete sine transform coefficients in T1 MRI images. This paper reveals their adherence to Benford’s law, characterized by a logarithmic distribution of first digits. The impact of Rician noise on the first digit distribution is examined, which causes deviations from the ideal distribution. A novel methodology is proposed for noise level estimation, employing metrics such as the Bhattacharyya distance, Kullback–Leibler divergence, total variation distance, Hellinger distance, and Jensen–Shannon divergence. Supervised learning techniques utilize these metrics as regressors. Evaluations on MRI scans from several datasets coming from a wide range of different acquisition devices of 1.5 T and 3 T, comprising hundreds of patients, validate the adherence of noiseless T1 MRI frequency domain coefficients to Benford’s law. Through rigorous experimentation, our methodology has demonstrated competitiveness with established noise estimation techniques, even surpassing them in numerous conducted experiments. This research empirically supports the application of Benford’s law in transforms, offering a reliable approach for noise estimation in denoising algorithms and advancing image quality assessment.

Defect detection enhancement using Gaussian Sparse Models

Detecting defects is critical in industrial fabrication, such as pipe welding, where radiography testing (RT) is the gold standard as a non-destructive method for monitoring weld quality and weld corrosion. The extraction of seamless information from radiographic images is critical for this approach. Image processing techniques can improve the quality of radiographic images by enhancing image contrast, especially in flawed regions. In this study, a method based on Gaussian mixture models was implemented and applied to radiographs of welded objects to improve visualization and detectability. In the Sparse Coding and Gaussian Scale Mixture method (SSC-GSM), the local image patches are described as a mixture of Gaussian distributions. Given the different levels of noise in the individual images, the background was determined and subtracted from each original image. The results show that the proposed techniques can maintain and improve the edge information in radiographs and defective regions. The density of pixels along the analyzed profile lines has yielded enhancement by a factor of approximately two for the reconstructed images as compared to the original images. We tested the SSC-GSM method on GD-Xray database with 68 radiographs of the weld. In total, the SSC-GSM technique can improve the contrast of radiography image by back ground removal method and show the defects better that exist in weld radiographs, compared to traditional method for the detection of weld defects.

Exit wave function reconstruction from two defocus images using neural network

Wave function reconstruction from one or two defocus images is promising for live atomic resolution imaging in transmission electron microscopy. However, a robust and accurate reconstruction method we still need more attention. Here, we present a neural-network-based wave function reconstruction method, EWR-NN, that enables accurate wave function reconstruction from only two defocus images. Results from both simulated and two different experimental defocus series show that the EWR-NN method has better performance than the widely-used iterative wave function reconstruction (IWFR) method. Influence of image number, defocus deviation, residual image shifts and noise level were considered to validate the performance of EWR-NN under practical conditions. It is seen that these factors will not influence the arrangement of atom columns in the reconstructed phase images, while they can alter the absolute values of all-atom columns and degrade the contrast of the phase images.

Segmentation of Images Used in Unmanned Aerial Vehicles Navigation Systems

The paper presents the results of the study of a two-stage procedure for selecting a reference object in the current image formed by a correlation-extreme system used for autonomous navigation of unmanned aerial vehicles. The aim of this paper is to theoretically evaluate the probability of selecting low-dimensional low-contrast objects in the segmented current image according to the proposed two-stage procedure. To achieve this goal, the problem of segmentation of images of the sighting surface and subsequent selection of the reference object in the presence of heterogeneous objects differing in brightness and area characteristics is solved. The most significant result is the justification of application of two-stage procedure of selection of the reference object in the current image by brightness and area parameters using the set thresholds. The significance of the obtained results consists in establishing the dependence of the probability of correct selection of the reference object on the noise level of the current images. It is shown that the probability of correct selection of the object in the image is a function of the threshold value and can be maximised by choosing its value. This approach allows to consider the influence of various factors leading to image noise on the quality of images formed by the navigation system. It is shown that when noise distorts more than 31% of the image pixels, the proposed two-stage procedure allows to ensure the selection of the reference object in the image with a probability not lower than 0.9.

Calibrating VLBI Polarization Data Using GPCAL. II. Time-dependent Calibration

We present a new method of time-dependent instrumental polarization calibration for very long baseline interferometry (VLBI). This method has been implemented in the recently developed polarization calibration pipeline GPCAL. Instrumental polarization, also known as polarimetric leakage, is a direction-dependent effect, and it is not constant across the beam of a telescope. Antenna pointing model accuracy is usually dependent on time, resulting in off-axis polarimetric leakages that can vary with time. The method is designed to correct for the off-axis leakages with large amplitudes that can severely degrade linear polarization images. Using synthetic data generated based on real Very Long Baseline Array (VLBA) data observed at 43 GHz, we evaluate the performance of the method. It was able to reproduce the off-axis leakages assumed in the synthetic data, particularly those with large amplitudes. The method has been applied to two sets of real VLBA data, and the derived off-axis leakages show very similar trends over time for pairs of nearby sources. Furthermore, the amplitudes of the off-axis leakages are strongly correlated with the antenna gain correction factors. The results demonstrate that the method is capable of correcting for the off-axis leakages present in VLBI data. By calibrating time-dependent instrumental polarization, the rms noise levels of the updated linear polarization images have been significantly reduced. The method is expected to substantially enhance the quality of linear polarization images obtained from existing and future VLBI observations.

Blind deep-learning based preprocessing method for Fourier ptychographic microscopy

Fourier ptychographic microscopy (FPM) is a technique for tackling the trade-off between the resolution and the imaging field of view by combining the techniques from aperture synthesis and phase retrieval to estimate the complex object from a series of low-resolution intensity images captured under angle-varied illumination. The captured images are commonly corrupted by multiple noise, leading to the degradation of the reconstructed image quality. Typically speaking, the noise model and noise level of the experimental images are unknown, and the traditional image denoising methods have limited effect. In this paper we model the FPM forward imaging process corrupted by noise and divide the noise in the captured images into two parts: the signal-dependent part and the signal-independent part. Based on the noise model we propose a novel blind deep-learning based Fourier ptychographic microscopy preprocessing method, termed BDFP, for removing these two components of noise. First, from a portion of the captured low-resolution images, a set of blocks corresponding to the smooth area of the object are extracted to model signal-independent noise. Second, under the assumption that the signal-dependent noise follows a Poisson distribution, we add Poisson noise and signal-independent noise blocks to clean images to form a paired training dataset, which is then used for training a deep convolutional neural network (CNN) model to reduce both signal-dependent noise and signal-independent noise. The proposed blind preprocessing method, combining with typical FPM reconstruction algorithms, is tested on simulated data and experimental images. Experimental results show that our preprocessing method can significantly reduce the noise in the captured images and bring about effective improvements in reconstructed image quality.

Single image noise level estimation by artificial noise

In this paper, we propose a new method for noise level estimation from a single noisy image contaminated by additive white Gaussian noise and gamma noise, which usually appear due to different physical imaging systems. Our method is motivated by the key observation that the curves of small eigenvalues of the patch covariance matrices corresponding to the original noisy image and the artificial noisy image are nearly parallel. Based on some assumptions, we derive a proportional relation equation between eigenvalues and variances of the noisy images in the redundant dimensions. The solution of this equation gives the basic noise estimation formula. However, the assumptions are not strictly satisfied for natural images with textures and fine details. To overcome this drawback, we modified the basic noise estimation formula by introducing a correction parameter. The proposed method is also extended to multiplicative Gamma noise estimation. Numerical results demonstrate that the proposed noise level estimation method outperforms the existing state-of-the-art approaches in accuracy for various scenes. In particular, the decreasing of Root Mean Square Error (RMSE) is an average of over 10% for the testing databases. We further demonstrate that the proposed noise estimation method can be used for blind denoising.

Adaptive image noise level estimation with Chi-square distribution on the flat patches selected by improved PCANet and ResNet101

Additive Gaussian white noise present in images adversely affects their quality, impeding effective human visualization. Current denoising approaches commonly rely on manual setting of the noise level, leading to insufficient noise removal or excessive smoothing. Traditional noise level estimation methods are not adaptive and their generalization ability is limited. To address this issue, this paper developed an adaptive image noise level estimation with Chi-square distribution on the flat patches selected by improved PCANet and ResNet101, including two key procedures: flat image patches selection and noise level estimation. The former was achieved using the improved PCANet and ResNet101 methods while the latter was completed by applying Chi-square distribution to the selected flat patches. Experimental results demonstrated that the reported method achieved high accuracy and powerful generalization ability, making it more suitable for extensive applications compared to existing methods.

Advanced Denoising Model for QR Code Images Using Hough Transformation and Convolutional Neural Networks

Quick Response (QR) code, a trademark for a two-dimensional code, has gained significant popularity in various sectors due to its innovative automatic identification and data detection capabilities in images. This research aims to enhance QR code identification rates by employing an effective pre-processing and detection method to mitigate noise levels in images with complicated backgrounds or uneven illumination. High-speed transformations on image blocks are utilized to improve recognition in these challenging conditions. A Convolutional Neural Network (CNN), a specialized network architecture for deep learning algorithms, is employed for QR image recognition and other pixel-based processing tasks. CNNs simplify the visuals without sacrificing essential information required for accurate predictions. In this paper, we propose an efficient Noise Removal in Quick Response Code Images using Hough Transformation (NRQRCI-HT) combined with CNN for noise reduction and accurate data identification. This method is benchmarked against traditional techniques, demonstrating superior performance levels in both noise removal and data identification accuracy.

Noise-robustness test for ultrasound breast nodule neural network models as medical devices.

Deep learning technology has been widely applied to medical image analysis. But due to the limitations of its own imaging principle, ultrasound image has the disadvantages of low resolution and high Speckle Noise density, which not only hinder the diagnosis of patients' conditions but also affect the extraction of ultrasound image features by computer technology. In this study, we investigate the robustness of deep convolutional neural network (CNN) for classification, segmentation, and target detection of breast ultrasound image through random Salt & Pepper Noise and Gaussian Noise. We trained and validated 9 CNN architectures in 8617 breast ultrasound images, but tested the models with noisy test set. Then, we trained and validated 9 CNN architectures with different levels of noise in these breast ultrasound images, and tested the models with noisy test set. Diseases of each breast ultrasound image in our dataset were annotated and voted by three sonographers based on their malignancy suspiciousness. we use evaluation indexes to evaluate the robustness of the neural network algorithm respectively. There is a moderate to high impact (The accuracy of the model decreased by about 5%-40%) on model accuracy when Salt and Pepper Noise, Speckle Noise, or Gaussian Noise is introduced to the images respectively. Consequently, DenseNet, UNet++ and Yolov5 were selected as the most robust model based on the selected index. When any two of these three kinds of noise are introduced into the image at the same time, the accuracy of the model will be greatly affected. Our experimental results reveal new insights: The variation trend of accuracy with the noise level in Each network used for classification tasks and object detection tasks has some unique characteristics. This finding provides us with a method to reveal the black-box architecture of computer-aided diagnosis (CAD) systems. On the other hand, the purpose of this study is to explore the impact of adding noise directly to the image on the performance of neural networks, which is different from the existing articles on robustness in the field of medical image processing. Consequently, it provides a new way to evaluate the robustness of CAD systems in the future.

Determining Body Condition of Dairy Cows for Early Diagnosis of Physiological Diseases

The paper points out the problem with automated diagnostics of body condition in dairy cattle, including ketosis. The conducted research is aimed at determining the possibility for non-contact automated diagnostics of the cattle physiological state on a daily basis. (Research purpose) To develop an algorithm for complex operational diagnostics of the physiological state of dairy cows by their live weight and body condition. (Materials and methods) Field data were collected in 2021-2022 on the FSUE Grigoryevskoye (Yaroslavl Region), Istra Cheese Factory and Lenin Dairy State Farm (Moscow Region). A commercial 3D ToF (Time-of-Flight) camera O3D303 was used. The 3D camera is capable of calculating and displaying the Point Cloud space as a multidimensional array. The program received 144 images, 136 images passed the filtering stage, 6 images did not detect the areas of interest, because of the high level of image noise, and the sacrum was not detected. 62 cows were subject to research. (Results and discussion) The sample and the dependence are proved to be representative as the Pearson correlation coefficient equals R=0.849, which shows a strong linear relationship between the body condition score and live weight. It was determined that in 24 percent of cases the body condition score is less than the least normal one. An algorithm was developed to help veterinarians to detect the animals that need additional examination. (Conclusions) It was found that the developed algorithm helps to quickly detect ketosis in dairy cows and automatically diagnose physiological diseases at an early stage, without additional labor and monetary costs.

Evaluation of reconstruction methods and image noise levels concerning visual assessment of simulated liver lesions in 111In-octreotide SPECT imaging

BackgroundEarly cancer detection is crucial for patients’ survival. The image quality in 111In-octreotide SPECT imaging could be improved by using Monte Carlo (MC)-based reconstruction. The aim of this observational study was to determine the detection rate of simulated liver lesions for MC-based ordered subset expectation maximization (OSEM) reconstruction compared to conventional attenuation-corrected OSEM reconstruction.MethodsThirty-seven SPECT/CT examinations with 111In-octreotide were randomly selected. The inclusion criterion was no liver lesions at the time of examination and for the following 3 years. SPECT images of spheres representing lesions were simulated using MC. The raw data of the spheres were added to the raw data of the established healthy patients in 26 of the examinations, and the remaining 11 examinations were not modified. The images were reconstructed using conventional OSEM reconstruction with attenuation correction and post filtering (fAC OSEM) and MC-based OSEM reconstruction without and with post filtering (MC OSEM and fMC OSEM, respectively). The images were visually and blindly evaluated by a nuclear medicine specialist. The criteria evaluated were liver lesion yes or no, including coordinates if yes, with confidence level 1–3. The percentage of detected lesions and accuracy (percentage of correctly classified cases), as well as tumor-to-normal tissue concentration (TNC) ratios and signal-to-noise ratios (SNRs), were evaluated.ResultsThe detection rates were 30.8% for fAC OSEM, 42.3% for fMC OSEM, and 50.0% for MC OSEM. The accuracies were 45.9% for fAC OSEM, 45.9% for fMC OSEM, and 54.1% for MC OSEM. The number of false positives was higher for fMC and MC OSEM. The observer’s confidence level was higher in filtered images than in unfiltered images. TNC ratios were significantly higher, statistically, with MC OSEM and fMC OSEM than with AC OSEM, but SNRs were similar due to higher noise with MC OSEM.ConclusionOne in two lesions were found using MC OSEM versus one in three using conventional reconstruction. TNC ratios were significantly improved, statistically, using MC-based reconstruction, but the noise levels increased and consequently the confidence level of the observer decreased. For further improvements, image noise needs to be suppressed.

Comparative study of the quantitative accuracy of oncological PET imaging based on deep learning methods.

[18F] Fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is an important tool for tumor assessment. Shortening scanning time and reducing the amount of radioactive tracer remain the most difficult challenges. Deep learning methods have provided powerful solutions, thus making it important to choose an appropriate neural network architecture. A total of 311 tumor patients who underwent 18F-FDG PET/CT were retrospectively collected. The PET collection time was 3 min/bed. The first 15 and 30 s of each bed collection time were selected to simulate low-dose collection, and the pre-90s was used as the clinical standard protocol. Low-dose PET was used as input, convolutional neural network (CNN, 3D Unet as representative) and generative adversarial network (GAN, P2P as representative) were used to predict the full-dose images. The image visual scores, noise levels and quantitative parameters of tumor tissue were compared. There was high consistency in image quality scores among all groups [Kappa =0.719, 95% confidence interval (CI): 0.697-0.741, P<0.001]. There were 264 cases (3D Unet-15s), 311 cases (3D Unet-30s), 89 cases (P2P-15s) and 247 cases (P2P-30s) with image quality score ≥3, respectively. There was significant difference in the score composition among all groups (χ2=1,325.46, P<0.001). Both deep learning models reduced the standard deviation (SD) of background, and increased the signal-to-noise ratio (SNR). When 8%PET images were used as input, P2P and 3D Unet had similar enhancement effect on SNR of tumor lesions, but 3D Unet could significantly improve the contrast-noise ratio (CNR) (P<0.05). There was no significant difference in SUVmean of tumor lesions compared with s-PET group (P>0.05). When 17%PET image was used as input, SNR, CNR and SUVmax of tumor lesion of 3D Unet group had no statistical difference with those of s-PET group (P>0.05). Both GAN and CNN can suppress image noise to varying degrees and improve image quality. However, when 3D Unet reduces the noise of tumor lesions, it can improve the CNR of tumor lesions. Moreover, quantitative parameters of tumor tissue are similar to those under the standard acquisition protocol, which can meet the needs of clinical diagnosis.

Invariant Pattern Recognition with Log-Polar Transform and Dual-Tree Complex Wavelet-Fourier Features.

In this paper, we propose a novel method for 2D pattern recognition by extracting features with the log-polar transform, the dual-tree complex wavelet transform (DTCWT), and the 2D fast Fourier transform (FFT2). Our new method is invariant to translation, rotation, and scaling of the input 2D pattern images in a multiresolution way, which is very important for invariant pattern recognition. We know that very low-resolution sub-bands lose important features in the pattern images, and very high-resolution sub-bands contain significant amounts of noise. Therefore, intermediate-resolution sub-bands are good for invariant pattern recognition. Experiments on one printed Chinese character dataset and one 2D aircraft dataset show that our new method is better than two existing methods for a combination of rotation angles, scaling factors, and different noise levels in the input pattern images in most testing cases.