This paper proposes an inverse neural network approach for stochastic model calibration, focusing on the conversion of high-dimensional system sequential responses into RGB (Red, Green, and Blue) images, which signifi...
详细信息
This paper proposes an inverse neural network approach for stochastic model calibration, focusing on the conversion of high-dimensional system sequential responses into RGB (Red, Green, and Blue) images, which significantly enhances the efficiency of calibration processes. By encoding multi-nodal, multi-directional data sequence into RGB images and employing advanced neural network architectures, including the Visual Geometry Group (VGG) network for frequency response data and Long Short-Term Memory (LSTM) integrated with Residual Networks (ResNet) for sequential time-domain data, the proposed method effectively decodes complex structural responses into stochastic model parameters. This process eliminates the need for conventional iterative optimization or Bayesian sampling methods, reducing computational costs while maintaining high accuracy in parameter identification. Two case studies, the NASA Langley Uncertainty Quantification Challenge and a satellite finite element model calibration task, demonstrate the effectiveness of the approach. The novel encoding-decoding framework enables real-time model calibration for high-dimensional data, making it a promising solution for complex engineering systems with large scale, high-dimensional data and inevitable uncertainties.
To achieve a visually captivating nocturnal image that closely resembles its natural daytime counterpart, people employ a range of techniques to process the nighttime image. The primary focus lies in achieving rapid a...
详细信息
To achieve a visually captivating nocturnal image that closely resembles its natural daytime counterpart, people employ a range of techniques to process the nighttime image. The primary focus lies in achieving rapid and stable unsupervised image enhancement effects specifically tailored for nocturnal scenes, without relying on daytime contrast image. However, existing neural network-based methods for enhancing nighttime image often rely on supervised paired training data, which presents challenges in practical production scenarios. The acquisition of image pairs depicting the same scene and the creation of a large-scale, feature-rich training dataset pose significant difficulties. In this study, we propose a fast pure nighttime image enhancement technique based on preprocessing inspired by the varying light sensitivity exhibited by fish during night fishing. The sensitivity of fish to light varies at different depths, analogous to the concealed richness of effective information within seemingly dark nighttime image, which can be effectively and comprehensively unveiled through preprocessing techniques. Subsequently, we employ an improved dual logarithmic imageprocessing method based on type-ii fuzzy sets to fuse the layer information obtained from preprocessing, resulting in enhanced contrast, noise reduction, color enhancement, and improved illumination with superior quality. The extensive experimental and comparative results demonstrate that our method's robust enhancement and restoration capabilities surpass even those of state-of-the-art supervised methods.
Deep unrolling, or unfolding, is an emerging learning-to-optimize method that unrolls a truncated iterative algorithm in the layers of a trainable neural network. However, the convergence guarantees and generalizabili...
详细信息
Deep unrolling, or unfolding, is an emerging learning-to-optimize method that unrolls a truncated iterative algorithm in the layers of a trainable neural network. However, the convergence guarantees and generalizability of the unrolled networks are still open theoretical problems. To tackle these problems, we provide deep unrolled architectures with a stochastic descent nature by imposing descending constraints during training. The descending constraints are forced layer by layer to ensure that each unrolled layer takes, on average, a descent step toward the optimum during training. We theoretically prove that the sequence constructed by the outputs of the unrolled layers is then guaranteed to converge for in-distribution problems. We then analyze the generalizability to certain out-of-distribution (OOD) shifts in the optimization problems being solved. Our analysis shows that the descending nature imposed by the proposed constraints is transferable under these distribution shifts, subject to a generalization error, thereby providing the unrolled networks with OOD robustness. We numerically assess unrolled architectures trained with the proposed constraints in two different applications, including the sparse coding using learnable iterative shrinkage and thresholding algorithm (LISTA) and image inpainting using proximal generative flow (GLOW-Prox), and demonstrate the performance and robustness advantages of the proposed method.
Recently, implicit neural representation (INR) has been applied to image compression. However, the rate-distortion performance of most existing INR-based image compression methods is still obviously inferior to the st...
详细信息
Recently, implicit neural representation (INR) has been applied to image compression. However, the rate-distortion performance of most existing INR-based image compression methods is still obviously inferior to the state-of-the-art image compression methods. In this letter, we propose an Enhanced Quantified Local Implicit neural Representation (EQLINR) for image compression by enhancing the utilization of local relationships of INR and narrow the quantization gap between training and encoding to further improve the performance of INR-based image compression. Our framework consists of latent representation and the corresponding implicit neural network consisting of MLP and CNN, which can transform the latent representation into the image space. To enhance local relationships utilization, we design a local enhancement module (LEM) consisted of CNN to capture the neighborhood relationships of the reconstructed image from MLP. Furthermore, to mitigate the performance loss caused by quantization of latent representation, we employ an enhanced quantization scheme (EQS) in our training process. We use uniform noise for network initialization and then use stochastic Gumbel Annealing (SGA) with dynamic temperature regulation as a proxy function for quantization during training. Extensive experimental results demonstrate that our approach significantly the compression performance of INR-based image compression, and even better than BPG.
Introduction: Tremendous developments in multimedia technology have promoted a massive amount of research in image and video processing. As imaging technologies are rapidly increasing, it is becoming essential to use ...
详细信息
Introduction: Tremendous developments in multimedia technology have promoted a massive amount of research in image and video processing. As imaging technologies are rapidly increasing, it is becoming essential to use images in almost every application in our day-to-day life. Materials and methods: This paper presents a comparative analysis of various image restoration approaches, ranging from fundamental methods to advanced techniques. These approaches aim to improve the quality of images that have been degraded during acquisition or transmission. A brief overview of the image restoration approaches is mentioned in the paper, which are as follows: (i) Wiener Filter: The Wiener filter is a classical approach used for image restoration. It is a linear filter that minimizes the mean square error between the original image and the restored image. (ii) Inverse Filter: The inverse filter is another traditional restoration technique. It attempts to invert the degradation process to recover the original image. However, inverse filtering is highly sensitive to noise and tends to amplify noise artifacts. (iii) Linear and Nonlinear Filtering: These methods involve applying linear or nonlinear filters to the degraded image to enhance its quality. Linear filters, such as Gaussian filters, can effectively reduce noise but may blur the image. Nonlinear filters, such as median filters, can preserve edges while reducing noise. (iv) Compressive Sensing and Restoration Approaches: Compressive sensing is a signalprocessing technique that exploits the sparsity of signals or images to reconstruct them from fewer measurements. CS-based restoration methods aim to recover high-quality images from compressed or incomplete measurements. (v) neural Networks Approaches: With the advancements in deep learning, neural networks have been widely used for image restoration tasks. Convolutional neural networks (CNNs) and generative adversarial networks (GANs) have shown promising results in restorin
As a noninvasive, nonradiative and high-speed imaging modality, fluorescence imaging in the second near-infrared window (NIR-ii, 1,000-1,700 nm) has demonstrated great potential for biomedical research and clinical st...
详细信息
As a noninvasive, nonradiative and high-speed imaging modality, fluorescence imaging in the second near-infrared window (NIR-ii, 1,000-1,700 nm) has demonstrated great potential for biomedical research and clinical study. The NIR-ii window can be further divided into two spectral regions: NIR-iia (1,000-1,300 nm) and NIR-iib (1,500-1,700 nm). Compared to NIR-iia, imaging in NIR-iib region affords high-resolution imaging at subcentimeter tissue depths due to suppressed photon scattering and diminished tissue autofluorescence at long wavelengths, but relies on probes with high toxicity. To address the problem, researchers employ deep learning networks to attain NIR-iib images from NIR-iia images. However, current methods require numerous paired or unpaired images (more than 2800 images) as training sets, which can hardly acquire. In this work, an innovative convolutional neural network (BRCycle-GAN) is trained based on a small training set (merely 63 images) to transform NIR-iia images into images with NIR-iib imaging qualities. The NIR-iib images generated by BRCycle-GAN outperform previous network models in terms of peak signal-to-noise ratio, cosine similarity and other image evaluation indices.
To address the problem that traditional convolutional neural networks cannot classify facial expression image features precisely, an interpretable face expression recognition method combining ResNet18 residual network...
详细信息
To address the problem that traditional convolutional neural networks cannot classify facial expression image features precisely, an interpretable face expression recognition method combining ResNet18 residual network and support vector machines (SVM) is proposed in the paper. The SVM classifier is used to enhance the matching ability of feature vectors and labels under the expression image feature space, to improve the expression recognition effect of the whole model. The class activation mapping and t-distributed stochastic neighbor embedding methods are used to visualize and interpret facial expression recognition's feature analysis and decision making under the residual neural network. The experimental results and the interpretable visualization analysis show that the algorithm structure can effectively improve the recognition ability of the network. Under the FER2013, JAFFE, and CK+ datasets, it achieved 67.65%, 84.44%, and 96.94% emotional recognition accuracy, respectively, showing a certain generalization ability and superior performance.
The upcoming Square Kilometre Array Observatory will produce images of neutral hydrogen distribution during the epoch of reionization by observing the corresponding 21-cm signal. However, the 21-cm signal will be subj...
详细信息
The upcoming Square Kilometre Array Observatory will produce images of neutral hydrogen distribution during the epoch of reionization by observing the corresponding 21-cm signal. However, the 21-cm signal will be subject to instrumental limitations such as noise and galactic foreground contamination that pose a challenge for accurate detection. In this study, we present the SegU-Net v2 framework, an enhanced version of our convolutional neural network, built to identify neutral and ionized regions in the 21-cm signal contaminated with foreground emission. We trained our neural network on 21-cm image data processed by a foreground removal method based on Principal Component Analysis achieving an average classification accuracy of 71 per cent between redshift z = 7 and 11. We tested SegU-Net v2 against various foreground removal methods, including Gaussian Process Regression, Polynomial Fitting, and Foreground-Wedge Removal. Results show comparable performance, highlighting SegU-Net v2's independence on these pre-processingmethods. Statistical analysis shows that a perfect classification score with AUC = 95 is possible for 8 < z < 10. While the network prediction lacks the ability to correctly identify ionized regions at higher redshift and differentiate well the few remaining neutral regions at lower redshift due to low contrast between 21-cm signal, noise, and foreground residual in images. Moreover, as the photon sources driving reionization are expected to be located inside ionized regions, we show that SegU-Net v2 can be used to correctly identify and measure the volume of isolated bubbles with V-ion > (10cmpc)(3 )at z > 9, for follow-up studies with infrared/optical telescopes to detect these sources.
Prostate cancer (PCa) represents the general type of cancer and is considered the third leading reason of death worldwide. As a combined part of computer-aided detection (CAD) applications, magnetic resonance imaging ...
详细信息
Prostate cancer (PCa) represents the general type of cancer and is considered the third leading reason of death worldwide. As a combined part of computer-aided detection (CAD) applications, magnetic resonance imaging (MRI) is extensively studied for the precise detection of PCa. However, various issues rely on MRI, which includes the complexity of interpretation and increased time. Thus, deep learning-based tumor detection and segmentation methods attempt to be imperative techniques for radiologists to execute their tasks more precisely. The objective is to present Gradient Bald vulture optimization (GBVO)-based Deep Convolution neural Networks (DCNN) with U-Net++ for segmenting and detecting prostate cancer. Initially, image pre-processing is done using Non-Local Means (NLM) filter. After pre-processing, image segmentation is carried out using the proposed optimized multi-objective Unet++, where the objective function in Unet++ is modified using pixel-wise cross entropy and the Jaccard coefficient. In addition, the training of Unet++ is done using the newly designed Gradient Bald Eagle Optimization (GBEO), which is a combination of stochastic Gradient Descent (SGD) and Bald Eagle optimization (BEO). Finally, cancer detection is done using DCNN. DCNN is trained using GBVO, which is the integration of the African Vultures Optimization Algorithm (AVOA) and GBEO. The proposed method out-performed state-of-the-art techniques. The developed method achieved the highest accuracy of 0.916, a false negative ratio (FNR) of 0.104, a false positive ratio (FPR) of 0.100, and a negative predictive value (NPV) of 0.903.
Pest's infection affects the crop production and annual income. From the past decade, many traditional methods anticipated the optimum accuracy while categorizing the infected tomato-plants. Every technique has th...
详细信息
Pest's infection affects the crop production and annual income. From the past decade, many traditional methods anticipated the optimum accuracy while categorizing the infected tomato-plants. Every technique has their pros and specifically the cons. As an upgradation, this paper introduces appropriate unsupervised detection & categorization of the diseased/healthy tomato plant using neural-net techniques. image dataset is congregation of both online and naturally accessible samples for healthy & diseased tomato crops. The current algorithm executes three steps to attain utmost performance: (i) Data pre-processing using Non-Subsampled Contourlet to acquire energy-detail components, (ii) Modified K-means processing to extract colored clusters, that are in-turn utilized for tomato-leaf detection, and (iii) finally Modified Convolution-neural Network features are fused to SVM for automated categorization. The work was tested for Kaggle PlantVillage and Mendeley datatset constituting 20,283 images, forming one healthy and 10 disease classes. The model undergoes the subjective performance metric evaluation and achieved the model accuracy as 99.15% and average precision of 95.6%. Technique produces highly intense, automatic and accurate classifier results over state-of-the-art approaches.
暂无评论