High Frequency Information Research Articles

High frequency domain enhancement and channel attention module for multi-view stereo

Multi-view stereo based on deep learning is increasingly popular as a method for 3D reconstruction. Existing methods have made significant advancements in pixel-level depth estimation. However, challenges such as occlusions and non-Lambertian surfaces in images hinder accurate confidence estimation. Moreover, cost volume regularization often results in excessive smoothing at object boundaries. To tackle these challenges, we propose integrating the High Frequency Information Compensator and 3D Channel Attention Module into the Multi-View Stereo Network, termed HFCA-MVS. Firstly, in the feature volume aggregation stage, we introduce a high-frequency information compensator module to enhance the correlation between 2D semantics and 3D space. Subsequently, in the cost volume regularization stage, a 3D channel attention module is introduced to enhance the representation of channel features by capturing relationships among different channels. Lastly, the 3DCNN network employs the GELU activation function to boost the activation response and mitigate excessive object boundary smoothing. HFCA-MVS demonstrates competitive performance in 3D reconstruction across three benchmark datasets: DTU, BlendMVS, and Tanks&Temples. Particularly, compared to CasMVSNet, MVSTER, and Geo-MVSNet on the DTU benchmark, HFCA-MVS achieves a relative improvement in completeness of 33%, 6.5%, and 0.4%, respectively, and an enhancement in overall performance of 15% and 4.2% compared to CasMVSNet and MVSTER. Furthermore, our model yields comparable reconstruction results to existing models on the Tanks&Temples dataset.

Low-Resolution Target Detection with High-Frequency Information Preservation

In the absence of high-frequency visual observation, low-resolution (LR) targets (e.g., objects, human body keypoints) are intrinsically difficult to detect in unconstrained images. This challenge can be further exasperated by typical downsampling operations (e.g., pooling, stride) of existing deep networks (e.g., CNNs). To tackle this challenge, in this work, we introduce a generic, High-Frequency Information Preservation (HFIP) block as a replacement for existing downsampling operations. It is composed of two key components: (1) the decoupled high-frequency learning component, which extracts the high-frequency information along the vertical and horizontal directions separately, and (2) the dilated frequency-aware channel correlation component, which decomposes the input feature map into multiple smaller ones in a dilated manner, concatenates them by channel, and then correlates the combined channels in the frequency space. Our module can generally be integrated into existing network architectures for target detection (e.g., YOLO, HRNet). Extensive experiments on low-resolution human pose estimation and object detection tasks show that our HFIP technique can generally boost the performance of state-of-the-art detection models significantly, e.g., improving the object detection accuracy of YOLOv5s by an absolute margin of 3.30% in mAP under a resolution of 640 × 640 compared to the COCO benchmark.

QAGA-Net: enhanced vision transformer-based object detection for remote sensing images

PurposeVision transformers (ViT) detectors excel in processing natural images. However, when processing remote sensing images (RSIs), ViT methods generally exhibit inferior accuracy compared to approaches based on convolutional neural networks (CNNs). Recently, researchers have proposed various structural optimization strategies to enhance the performance of ViT detectors, but the progress has been insignificant. We contend that the frequent scarcity of RSI samples is the primary cause of this problem, and model modifications alone cannot solve it.Design/methodology/approachTo address this, we introduce a faster RCNN-based approach, termed QAGA-Net, which significantly enhances the performance of ViT detectors in RSI recognition. Initially, we propose a novel quantitative augmentation learning (QAL) strategy to address the sparse data distribution in RSIs. This strategy is integrated as the QAL module, a plug-and-play component active exclusively during the model’s training phase. Subsequently, we enhanced the feature pyramid network (FPN) by introducing two efficient modules: a global attention (GA) module to model long-range feature dependencies and enhance multi-scale information fusion, and an efficient pooling (EP) module to optimize the model’s capability to understand both high and low frequency information. Importantly, QAGA-Net has a compact model size and achieves a balance between computational efficiency and accuracy.FindingsWe verified the performance of QAGA-Net by using two different efficient ViT models as the detector’s backbone. Extensive experiments on the NWPU-10 and DIOR20 datasets demonstrate that QAGA-Net achieves superior accuracy compared to 23 other ViT or CNN methods in the literature. Specifically, QAGA-Net shows an increase in mAP by 2.1% or 2.6% on the challenging DIOR20 dataset when compared to the top-ranked CNN or ViT detectors, respectively.Originality/valueThis paper highlights the impact of sparse data distribution on ViT detection performance. To address this, we introduce a fundamentally data-driven approach: the QAL module. Additionally, we introduced two efficient modules to enhance the performance of FPN. More importantly, our strategy has the potential to collaborate with other ViT detectors, as the proposed method does not require any structural modifications to the ViT backbone.

Option Pricing Based on REGARCH Model with High-Frequency Information

As a prevalent tool for hedging risk, the trading volume of options has been growing increasingly in the derivatives market. Precision in the estimation of volatility leads to accurate option pricing. Since volatility is time-varying and has a clustering effect, GARCH class of volatility models is effective in modeling volatility precisely. This paper utilizes the realized EGARCH (REGARCH) model combined with Monte Carlo simulation to investigate the role of high-frequency information in option pricing. The parameter estimates of the REGARCH model are obtained via joint maximum likelihood estimation using observations on returns and realized measure. Applying the model to S&P options market, the empirical results show that the REGARCH model that using high-frequency data is more efficient than the model that only use daily closing prices, including the EGARCH, NGARCH and GJR-GARCH models. This paper demonstrates that incorporating realized measures into volatility models can improve the accuracy of option pricing. The REGARCH model contained more intraday trading information from high-frequency data, can measure the additional risk premiums and specific volatility shocks.

Association of the Genotype of TNFα with the Efficiency of the Intensive Phase of Chemotherapy and Formation of the Sizes of Destruction Cavities in Patients with Pulmonary Tuberculosis.

The interrelation between the TNFα gene polymorphism and the effectiveness of the intensive phase of chemotherapy in patients with pulmonary tuberculosis, as well as the formation of diameter of the sizes of destruction cavities was studied. It was revealed that the most adverse course of the disease and a high frequency of formation of destruction cavities are associated with the GG genotype of the TNFα gene.

CNCAN: Contrast and normal channel attention network for super-resolution image reconstruction of crops and weeds

Numerous studies have been performed to apply camera vision technologies in robot-based agriculture and smart farms. In particular, to obtain high accuracy, it is essential to procure high-resolution (HR) images, which requires a high-performance camera. However, due to high costs it is difficult to widely apply the camera in agricultural robots. To overcome this limitation, we propose contrast and normal channel attention network (CNCAN) for super-resolution reconstruction (SR), which is the first research for the accurate semantic segmentation of crops and weeds even with low-resolution (LR) images captured by low-cost and LR camera. Attention block and activation function that considers high frequency and contrast information of images are used in CNCAN, and the residual connection method is applied to improve the learning stability.As a result of experimenting with three open datasets, namely, Bonirob, rice seedling and weed, and crop/weed field image (CWFID) datasets, the mean intersection of union (MIOU) results of semantic segmentation for crops and weeds with SR images through CNCAN were 0.7685, 0.6346, and 0.6931 in the Bonirob, rice seedling and weed, and CWFID datasets, respectively, confirming higher accuracy than other state-of-the-art methods for SR.

Give me a hint: an explicit prior based image denoising

The performance of an image denoising method highly depends on its ability to produce smoothed homogeneous regions and preserve fine-grained texture details in the restored image. Existing denoising methods rely on implicit learning of Convolutional Neural Networks (CNNs) to restore an image. We show that the implicitly learned representations are limited in capacity, resulting in a sub-optimal pixel estimation of homogeneous regions and loss in texture details. In this work, we argue that explicitly introducing low and high frequency information enhances the representational capacity of denoising CNNs. To this end, we propose XPrior that provides the network with two explicit priors, i.e., a Gaussian smoothed image and an Edge map. This simple yet effective integration of priors with the noisy image enhances the model’s ability to restore images with high perceived quality. The proposed approach can easily be plugged into the existing denoising methods. Comprehensive experimental results show the superiority of the proposed approach that achieves a remarkable performance gain of 1.23 dB and 1.49 dB average PSNR on BSD68 dataset over DRUNet (Zhang in IEEE Trans Pattern Anal Mach Intell 10.1109/TPAMI.2021.3088914, 2021) and DnCNN (Zhang in IEEE Trans Image Process 26:3142–3155, 2017), respectively.

A Review of CiteSpace Visualisation and Analysis of High-Temperature Corrosion Inhibitors for Oil and Gas Fields

With the increasing global energy demand, the exploitation depth of oil and gas fields is gradually increasing. At the same time, importing many high-acid crude oil makes it increasingly heavy and inferior, and its acid value is constantly improving. This change has led to the increasingly severe corrosion problem of oil and gas field equipment, which has become an essential factor affecting the national economy and society's sustainable development. As an economical and practical anticorrosion measure, corrosion inhibitors play a crucial role in the anticorrosion of oil and gas field equipment. Through the research method of biorientation, retrieve the Web of Science database in 2008-2024 about oil and gas field high-temperature corrosion inhibitor research literature information, using CiteSpace measurement analysis software visual analysis in the literature keywords, publications, high citation frequency, cooperation and word clustering information change trend, analyze the research situation of oil and gas field in recent years, summarizes the research hotspot of oil field high-temperature corrosion inhibitor, prominent authors and institutions, reference relationship and development trend. The number of documents on high-temperature corrosion inhibitors in oil and gas fields is smaller, and the authors, institutions, and countries are not closely related. There is less research cooperation on high-temperature corrosion inhibitors in oil and gas fields, and more independent studies exist. China University of Petroleum has the first number of publications, and China is the country with the first number of publications. The study of mass spectrometry for corrosion inhibitors in oil and gas fields is prominent, while the literature on high-temperature corrosion inhibitors is rare.

A modified Fuvar fusion algorithm based on adaptive end-member selection for hyperspectral remote sensing images

ABSTRACT A significant strategy for achieving a balance between spatial and spectral resolution is to combine hyperspectral remote sensing images with low spatial resolution and multispectral remote sensing images with high spatial resolution. For the issues of spectrum distortion and lack of the ability of gaining high frequency information when the Fuvar algorithm employs the Vertex Component Analysis (VCA) technique, this thesis presents a modified Fuvar (MFuvar) algorithm that utilizes the Maximum Distance Analysis (MDA) method instead of the VCA approach. The two subsets from the hyperspectral remote sensing image of GF-5 and the multispectral remote sensing image of Sentinel-2A, representing different land cover types were used as test data. The spectral fidelity and the ability of gaining high frequency information were assessed by using visual and statistical analysis. Fused images are compared with eight fusion methods, including SFIMHS, GLPHS, MAPSMM, CNMF, Hysure, SpaFusion, LTTR, and Fuvar, respectively. The results show that the MFuvar algorithm can keep the best balance between spectral fidelity and the ability of gaining high frequency information, and it is generally better than the compared eight algorithms. And it fulfils the automatic selection of end elements without manual intervention and increases the efficiency of algorithm operation.

Establishing a user demand hierarchy model driven by a mental model for complex operating systems

In complex human-computer interactions, issues such as task failures, system malfunctions, and frequent accidents caused by user errors are common. Therefore, it is necessary to study complex system interactions to enhance overall efficiency. This study focuses on the task interface of a digital twin system for computer numerical control (CNC) machine tools and examines the relationship between the user mental model and interface design elements. Key mental information is obtained through questionnaires and interviews, forming the basis for establishing a user mental model. High-frequency information words are extracted, experimental samples are designed, and importance rating surveys are conducted. Quantitative analysis methods, including factor analysis and weight calculation, are utilized to analyze the needs of target users. Consequently, a user demand hierarchy model is constructed. This approach aims to effectively reduce user errors in the human-computer interaction process within complex systems and enhance cognitive efficiency.

3D Roughness Prediction Modeling and Evaluation of Textured Liner of Piston Component-Cylinder System

In this study, a machine vision method is proposed to characterize 3D roughness of the textured surface on cylinder liner processed by plateau honing. The least absolute value (L∞) regression robust algorithm and Levenberg-Marquardt (LM) algorithm are employed to reconstruct image reference plane. On this basis, a single-hidden layer feedforward neural network (SLFNN) based on the extreme learning machine (ELM) is employed to model the relationship between high frequency information and 3D roughness. The characteristic parameters of Abbott-Firestone curve and 3D roughness measured by a confocal microscope are used to construct ELM-SLFNN prediction model for 3D roughness. The results indicate that the proposed method can effectively characterize 3D roughness of the textured surface of cylinder liner.

Enhancement of Underwater Images through Parallel Fusion of Transformer and CNN

Ocean exploration is crucial for utilizing its extensive resources. Images captured by underwater robots suffer from issues such as color distortion and reduced contrast. To address the issue, an innovative enhancement algorithm is proposed, which integrates Transformer and Convolutional Neural Network (CNN) in a parallel fusion manner. Firstly, a novel transformer model is introduced to capture local features, employing peak-signal-to-noise ratio (PSNR) attention and linear operations. Subsequently, to extract global features, both temporal and frequency domain features are incorporated to construct the convolutional neural network. Finally, the image’s high and low frequency information are utilized to fuse different features. To demonstrate the algorithm’s effectiveness, underwater images with various levels of color distortion are selected for both qualitative and quantitative analyses. The experimental results demonstrate that our approach outperforms other mainstream methods, achieving superior PSNR and structural similarity index measure (SSIM) metrics and yielding a detection performance improvement of over ten percent.

High-frequency and low-frequency dual-channel graph attention network

Most existing graph convolution layers use learnable or fixed weights to sum up neighbor features to aggregate neighbor information. Since the attention values are always positive, these graph convolution layers perform as low-pass filters, which may result in their poor performance on heterophilic graphs. In this paper, two graph convolutional layers are proposed, HLGAT and NGAT. NGAT is a convolution network using only negative attention values, which only make the aggregation of high-frequency information of neighbor nodes. HLGAT makes the aggregation of low-frequency and high-frequency information by two channels, respectively, and fuses two outputs by using a learnable way. On node-classification task, both NGAT and HLGAT offer significant performance improvement compared to existing methods. The results clearly show that: (1) High-frequency information of neighborhoods plays a decisive role in heterophilic graphs. (2) The aggregation of low-frequency and high-frequency information of neighbor nodes can significantly improve the performance on heterophilic graphs.

Token-word mixer meets object-aware transformer for referring image segmentation

Referring image segmentation aims to generate a binary mask of the target object according to a referring expression. Some recent works argue that post-fusion paradigm may result in inconsistency and insufficiency issue and propose to integrate textual features during the visual encoding process. Although effective, they do fusion in a single way at each stage of encoder, e.g. utilizing cross attention mechanisms. This single fusion method ignores local and detailed image information correlated with language due to the incapability of attention in capturing high-frequencies information. To address this issue, we propose a Token-Word Mixer, which takes into consideration the characteristics of convolution and attention, and achieves more comprehensive interactions and alignments of multi-modal features through a mix operation. Furthermore, existing methods that rely solely on grid features lack perception of the target object and inference of relationships between objects, making it difficult to associate and align semantic information of target objects during multi-modal fusion when referring expressions or image scenes are complex. Therefore, we propose to incorporates object-level information by exploiting a DETR-based detector to provide region features, and the Object-Aware Transformer encoder with an additional learnable token is proposed to perceive effective information associated with the target object. Based on the enhanced cross-modal features and the aggregated token, we adopt query-based mask generation method instead of pixel classification framework for referring image segmentation. Extensive experiments and ablation studies indicate the effectiveness of our proposed methods.

Enforcing high frequency enhancement in deep networks for simultaneous depth estimation and dehazing

Single image dehazing and monocular depth estimation algorithms are crucial for environmental perception for autonomous driving. However, few algorithms can simultaneously perform single-image dehazing and depth estimation to solve the problems of perceiving surrounding information for unmanned vehicles in haze. The current algorithm based on the Transformer and Convolutional Neural Networks cannot effectively extract high frequency information and low frequency information in the image, resulting in poor image dehazing and depth estimation ability. Therefore, we design the DADENet network to solve these problems. We enhanced sensitivity to high-frequency information by incorporating Sobel operators and Gaussian blur into the residual unit structure, resulting in the ResNet-Edge-Encoder block. Furthermore, we improved the capability to extract both high-frequency and low-frequency information from feature maps by effectively combining depthwise convolutions with attention structures in the Transformer-Edge-Decoder block. Additionally, we designed the Depth-Transmission loss function to leverage the relationship between depth maps and transmission maps, thereby enhancing the accuracy of both single-image dehazing and depth estimation tasks. Extensive experiments on the NYU, KITTI and Citycapes datasets demonstrate that our DADENet algorithm effectively performs single-image dehazing and depth estimation, surpassing classical and state-of-the-art algorithms.

Far-Field Super-Resolution Microscopy Using Evanescent Illumination: A Review

The resolution of conventional optical microscopy is restricted by the diffraction limit. Light waves containing higher-frequency information about the sample are bound to the sample surface and cannot be collected by far-field optical microscopy. To break the resolution limit, researchers have proposed various far-field super-resolution (SR) microscopy imaging methods using evanescent waves to transfer the high-frequency information of samples to the low-frequency passband of optical microscopy. Optimization algorithms are developed to reconstruct a SR image of the sample by utilizing the high-frequency information. These techniques can be collectively referred to as spatial-frequency-shift (SFS) SR microscopy. This review aims to summarize the basic principle of SR microscopy using evanescent illumination and introduce the advances in this research area. Some current challenges and possible directions are also discussed.

Dual-space high-frequency learning for transformer-based MRI super-resolution

Background and ObjectiveMagnetic resonance imaging (MRI) can provide rich and detailed high-contrast information of soft tissues, while the scanning of MRI is time-consuming. To accelerate MR imaging, a variety of Transformer-based single image super-resolution methods are proposed in recent years, achieving promising results thanks to their superior capability of capturing long-range dependencies. Nevertheless, most existing works prioritize the design of transformer attention blocks to capture global information. The local high-frequency details, which are pivotal to faithful MRI restoration, are unfortunately neglected. MethodsIn this work, we propose a high-frequency enhanced learning scheme to effectively improve the awareness of high frequency information in current Transformer-based MRI single image super-resolution methods. Specifically, we present two entirely plug-and-play modules designed to equip Transformer-based networks with the ability to recover high-frequency details from dual spaces: 1) in the feature space, we design a high-frequency block (Hi-Fe block) paralleled with Transformer-based attention layers to extract rich high-frequency features; while 2) in the image intensity space, we tailor a high-frequency amplification module (HFA) to further refine the high-frequency details. By fully exploiting the merits of the two modules, our framework can recover abundant and diverse high-frequency information, rendering faithful MRI super-resolved results with fine details. ResultsWe integrated our modules with six Transformer-based models and conducted experiments across three datasets. The results indicate that our plug-and-play modules can enhance the super-resolution performance of all foundational models to varying degrees, surpassing the capabilities of existing state-of-the-art single image super-resolution networks. ConclusionComprehensive comparison of super-resolution images and high-frequency maps from various methods, clearly demonstrating that our module possesses the capability to restore high-frequency information, showing huge potential in clinical practice for accelerated MRI reconstruction.

Short Utterance Speaker Recognition Based on Speech High Frequency Information Compensation and Dynamic Feature Enhancement Methods

This work aims to further compensate for the weaknesses of feature sparsity and insufficient discriminative acoustic features in existing short-duration speaker recognition. To address this issue, we propose the Bark-scaled Gauss and the linear filter bank superposition cepstral coefficients (BGLCC), and the multidimensional central difference (MDCD) acoustic feature extracted method. The Bark-scaled Gauss filter bank focuses on low-frequency information, while linear filtering is uniformly distributed, therefore, the filter superposition can obtain more discriminative and richer acoustic features of short-duration audio signals. In addition, the multi-dimensional central difference method captures better dynamics features of speakers for improving the performance of short utterance speaker verification. Extensive experiments are conducted on short-duration text-independent speaker verification datasets generated from the VoxCeleb, SITW, and NIST SRE corpora, respectively, which contain speech samples of diverse lengths, and different scenarios. The results demonstrate that the proposed method outperforms the existing acoustic feature extraction approach by at least 10% in the test set. The ablation experiments further illustrate that our proposed approaches can achieve substantial improvement over prior methods.

Improved modeling of human vision by incorporating robustness to blur in convolutional neural networks

Whenever a visual scene is cast onto the retina, much of it will appear degraded due to poor resolution in the periphery; moreover, optical defocus can cause blur in central vision. However, the pervasiveness of blurry or degraded input is typically overlooked in the training of convolutional neural networks (CNNs). We hypothesized that the absence of blurry training inputs may cause CNNs to rely excessively on high spatial frequency information for object recognition, thereby causing systematic deviations from biological vision. We evaluated this hypothesis by comparing standard CNNs with CNNs trained on a combination of clear and blurry images. We show that blur-trained CNNs outperform standard CNNs at predicting neural responses to objects across a variety of viewing conditions. Moreover, blur-trained CNNs acquire increased sensitivity to shape information and greater robustness to multiple forms of visual noise, leading to improved correspondence with human perception. Our results provide multi-faceted neurocomputational evidence that blurry visual experiences may be critical for conferring robustness to biological visual systems.

The origins of frequency-difference and frequency-sum beamforming

Beamforming the signals recorded by an array enables the determination of sound source location(s) or the arrival directions of ray paths between a sound source and the receiving array. Frequency-difference and frequency-sum beamforming are beamforming techniques that provide out-of-band information from in-band signal frequencies. Interestingly, the out-of-band frequencies can be chosen by the user, within limits set by the signal recordings, to achieve desired properties of the beamformed output, such as: increased resolution, reduced sidelobes, or greater robustness to random scattering. Both techniques are general and are not limited to any particular acoustic environment, frequency range, or array geometry. Frequency-sum beamforming, generates higher-frequency information from lower frequency signal components, enhancing beamforming results in scenarios with random scattering between the source and the receivers. However, it is limited by artifacts arising from cross-terms when multiple source signals are present in the same bandwidth. Conversely, frequency-difference beamforming manufactures lower-frequency information from higher frequency signal components, effectively mitigating the impact of spatial aliasing in situations where the receiving array is sparse. This presentation delves into the origins of frequency-difference and frequency-sum beamforming, presents the fundamental mathematics underlying their algorithms, and showcases their performance via simulations and experimental results. [Work supported by ONR.]