Loss Of Detail Research Articles

An Intelligent Traffic Analysis and Prediction System Using Deep Learning Technique

Accurate identification of vehicles and estimating density in traffic surveillance systems is a challenging task, particularly in scenarios with closely spaced lanes. Single Shot MultiBox Detector (SSD) is introduced in vehicle detection and classification due to its speed and accuracy. It utilizes a transfer learning technique that enables them to utilize features from pretrained Convolutional Neural Networks (CNNs). Although the utilization of multi-scale feature maps in SSD has achieved efficient results in vehicle detection, it struggles to identify key features of vehicles due to its one-stage detection approach. Furthermore, the use of VGG16 as the backbone network in these approaches leads to the loss of fine-grained details, posing challenges in accurately localizing and classifying small vehicles. Also, there is no interaction between high- and low-level features, which restricts the networks ability to effectively integrate and utilize both features for accurate vehicle detection. To overcome these limitations, an improved SSD approach is introduced in this paper, leveraging interactive multi-scale attention characteristics to accurately detect and classify vehicles. This approach utilizes ResNet50 as the network backbone to overcome the limitations of traditional SSDs in detecting small vehicles. Also, an attention block is incorporated into it to focus on key details and assign higher attention to relevant pixels within the feature map. Also, the network employs a parallel detection framework and shares multi-scale layers (both high and low), enabling efficient detection of vehicles with various sizes. Then, traffic density is estimated based on the weights assigned to the categorized vehicles obtained from the improved SSD and the recorded area. Finally, traffic is classified as high, low, or moderate by comparing the estimated density to a threshold value. By adding attention characteristics of different scales to the original detection branch and replacing the VGG16 with ResNet50 of the SSD technique using our method, the feature representation capability and detection accuracy are both significantly improved.

Dense Connected Edge Feature Enhancement Network for Building Edge Detection from High Resolution Remote Sensing Imagery

Deep-learning-based methods for building-edge-detection have been widely researched and applied in the field of image processing. However, these methods often emphasis the analysis of deep features, which may result in neglecting crucial shallow information representation. Furthermore, abstract features in the deep layers can potentially interfere with the accuracy of edge extraction. To address these challenges, we propose a densely connected edge-detection enhancement network (DCEFE-Net) for building-edge-detection in high-resolution remote sensing images. Firstly, by introducing spatial land channel attention modules, we effectively captured low-level spatial information and high-level semantic information from the input image. Secondly, the proposed edge-aware feature enhancement (EAFE) module emphasis the representation of informative edge features. By alliteratively generating multiple layers of edge-detection maps, it addresses the issue of edge detail loss and enhances edge-detection accuracy. Finally, the dense connectivity blocks strengthen the connections between the convolutional layers, thereby preventing the loss of edge features. Experimental results on the WHU and the Inria Aerial Image Labeling datasets validate the effectiveness of DCEFE-Net, as it consistently produces clear and reliable building-edge results.

NeXtResUNet: A neural network for industrial CT image denoising

Image denoising is a critical issue in industrial computed tomography (CT) inspection. Most existing noise reduction algorithms are based on synthetic data, resulting in the loss of fine details, local region distortions, and other problems that arise during application in practical scenarios. As such, a fusion network, combining ConvNeXt, ResNet, and UNet is proposed in this study to address these issues. This algorithm was applied to a self-constructed industrial CT image denoising dataset, achieving a peak signal-to-noise ratio (PSNR) of 47.20 dB, which is 6% (44.48 dB) higher than that of the benchmark DnCnn. This also represents a 0.5% improvement over the current state-of-the-art (SOTA) SCUNet (46.94 dB). More importantly, the proposed network is superior to other existing models in terms of the overall visual effects and degree of image detail preservation. The network structure is also concise yet simple to expand. As such, it can be applied not only to image denoising, but also to CT image segmentation, super resolution, and related tasks.

An adaptive framework of De-noising and enhancement for fundus imaging using enhanced guided filter and non-illumination correction method for diabetic retinopathy

Retinal fundus imaging has been used in the diagnosis of many cardiovascular and retinal-related diseases like age-related macular degeneration, pathological myopia, diabetic retinopathy, and glaucoma. In recent years, the computerized technique is adapted to fundus image analysis and processing for quick diagnosis. However, low-quality retinal pictures are produced when certain eye disorders and photographic circumstances exist. Poor-quality retinal pictures of this kind are not helpful for diagnosis, particularly when using automated image analysis software. The causes for degradation include low illumination, high illumination, image blurring, uneven illumination, low contrast, and color distortion. Several image-enhancing techniques have been developed to address this; however, they may result in sudden changes in color levels, false borders, and the loss of picture detail, particularly in retinal images. To prevent these negative consequences, this paper proposed a novel enhancement algorithm EGF_AGC (Enhanced Guided Filter+Adaptive Gamma Correction) for color fundus retinal images. The proposed method is worked is described in four steps. The first stage of the recommended procedure is to utilize a hybrid filter to enhance the retinal pictures’ appearance details. Next, applied enhanced guided filter method to the detailed retina images to enhance the image quality. Moreover, the suggested technique also worked to retain uneven illumination issues raised in retina images using the Adaptive Gamma Correction method. The final improved retinal pictures are created by combining the phases before utilizing the HSV color approach. The databases of retinal images, which include the STARE, the DRIVE, and the CHBASE_DB1 database, were utilized to determine image enhancement effects. The findings were compared with CLAHE, Bilateral Filter, Guided Filter, and Gamma Correction retinal enhancement methods. Compared to similar enhancement approaches, experiments showed that our method could give a competitive outcome and eliminate degradation caused by low-quality fundus pictures.

Infrared and visible image fusion method based on full convolutional network (FCN)

In the IP sector, the combination of visible image fusion (VIF) with infrared (IR) gives a more comprehensive and accurate description of a target image. To get over the problems of detail and energy loss during the fusion process caused by current deep learning fusion approaches, it is proposed to use a fusion strategy of IR and visible pictures based on full convolutional network (FCN) applying transfer learning. FCN model can take any size of the input and generate constant size of the output with desired rules. Through effective inference and learning procedure, the ability of features extraction and energy conservation can be enhanced a lot. Experimental results demonstrate that the suggested method succeeds in improving IF quality over the other two comparable methods by preserving high light intensity and retrieving detail information. This also confirms its dominance across five different objective assessment indices: mutual information (MI), entropy (EN), edge-based similarity measure (Qabf), sum of correlations of differences (SCD), and multi-scale structural similarity for image (MS-SSIM).

Strategically Acquired Gradient Echo (STAGE) Imaging, part IV: Constrained Reconstruction of White Noise (CROWN) Processing as a Means to Improve Signal-to-Noise in STAGE Imaging at 3 Tesla

Increasing the signal-to-noise ratio (SNR) has always been of critical importance for magnetic resonance imaging. Although increasing field strength provides a linear increase in SNR, it is more and more costly as field strength increases. Therefore, there is a major effort today to use signal processing methods to improve SNR since it is more efficient and economical. There are a variety of methods to improve SNR such as averaging the data at the expense of imaging time, or collecting the data with a lower resolution, all of these methods, including imaging processing methods, usually come at the expense of loss of image detail or image blurring. Therefore, we developed a new mathematical approach called CROWN (Constrained Reconstruction of White Noise) to enhance SNR without loss of structural detail and without affecting scanning time. In this study, we introduced and tested the concept behind CROWN specifically for STAGE (strategically acquired gradient echo) imaging. The concept itself is presented first, followed by simulations to demonstrate its theoretical effectiveness. Then the SNR improvement on proton spin density (PSD) and R2⁎ maps was investigated using brain STAGE data acquired from 10 healthy controls (HCs) and 10 patients with Parkinson's disease (PD). For the PSD and R2* maps, the SNR and CNR between white matter and gray matter were improved by a factor of 1.87 ± 0.50 and 1.72 ± 0.88, respectively. The white matter hyperintensity lesions in PD patients were more clearly defined after CROWN processing. Using these improved maps, simulated images for any repeat time, echo time or flip angle can be created with improved SNR. The potential applications of this technology are to trade off the increased SNR for higher resolution images and/or faster imaging.

Zoom Text Detector.

To pursue comprehensive performance, recent text detectors improve detection speed at the expense of accuracy. They adopt shrink-mask-based text representation strategies, which leads to a high dependence of detection accuracy on shrink-masks. Unfortunately, three disadvantages cause unreliable shrink-masks. Specifically, these methods try to strengthen the discrimination of shrink-masks from the background by semantic information. However, the feature defocusing phenomenon that coarse layers are optimized by fine-grained objectives limits the extraction of semantic features. Meanwhile, since both shrink-masks and the margins belong to texts, the detail loss phenomenon that the margins are ignored hinders the distinguishment of shrink-masks from the margins, which causes ambiguous shrink-mask edges. Moreover, false-positive samples enjoy similar visual features with shrink-masks. They aggravate the decline of shrink-masks recognition. To avoid the above problems, we propose a zoom text detector (ZTD) inspired by the zoom process of the camera. Specifically, zoomed-out view module (ZOM) is introduced to provide coarse-grained optimization objectives for coarse layers to avoid feature defocusing. Meanwhile, zoomed-in view module (ZIM) is presented to enhance the margins recognition to prevent detail loss. Furthermore, sequential-visual discriminator (SVD) is designed to suppress false-positive samples by sequential and visual features. Experiments verify the superior comprehensive performance of ZTD.

Underwater image denoising based on curved wave filtering and two-dimensional variational mode decomposition

Underwater image denoising technology is of great significance in underwater operation. Underwater operations (such as offshore oil drilling, undersea tunnels, pipeline construction, underwater archaeology, biological research, and lifesaving) require stable and clear underwater images to aid analysis. Due to the scattering and absorption of light by water bodies, obtaining high-quality underwater images is a challenging task. Underwater images are prone to low contrast, low resolution and edge distortion. Therefore, it is difficult to accurately separate the effective signal when removing the underwater image noise, which leads to the image contrast reduction. Also the edge contour is not clear, and the detail loss is serious. Therefore, we propose a novel underwater image denoising method based on curved wave filter and two-dimensional variational mode decomposition. Firstly, the noisy image is decomposed by two-dimensional variational mode decomposition, and a series of modal components with different center frequencies are obtained. The effective modal components are selected by correlation coefficient and structural similarity. And the effective modal components are processed by the curve-wave filter. Finally, the filtered modal components are reconstructed to remove the noise in the image. The experimental results show that, compared with other state-of-the-art methods, the proposed method has clearer denoising results, less mean square error, and better denoising effect.

JPEG Quantized Coefficient Recovery via DCT Domain Spatial-Frequential Transformer.

JPEG compression adopts the quantization of Discrete Cosine Transform (DCT) coefficients for effective bit-rate reduction, whilst the quantization could lead to a significant loss of important image details. Recovering compressed JPEG images in the frequency domain has recently garnered increasing interest, complementing the multitude of restoration techniques established in the pixel domain. However, existing DCT domain methods typically suffer from limited effectiveness in handling a wide range of compression quality factors or fall short in recovering sparse quantized coefficients and the components across different colorspaces. To address these challenges, we propose a DCT domain spatial-frequential Transformer, namely DCTransformer, for JPEG quantized coefficient recovery. Specifically, a dual-branch architecture is designed to capture both spatial and frequential correlations within the collocated DCT coefficients. Moreover, we incorporate the operation of quantization matrix embedding, which effectively allows our single model to handle a wide range of quality factors, and a luminance-chrominance alignment head that produces a unified feature map to align different-sized luminance and chrominance components. Our proposed DCTransformer outperforms the current state-of-the-art JPEG artifact removal techniques, as demonstrated by our extensive experiments.

Prior Knowledge-guided Triple-Domain Transformer-GAN for Direct PET Reconstruction from Low-Count Sinograms.

To obtain high-quality positron emission tomography (PET) images while minimizing radiation exposure, numerous methods have been dedicated to acquiring standard-count PET (SPET) from low-count PET (LPET). However, current methods have failed to take full advantage of the different emphasized information from multiple domains, i.e., the sinogram, image, and frequency domains, resulting in the loss of crucial details. Meanwhile, they overlook the unique inner-structure of the sinograms, thereby failing to fully capture its structural characteristics and relationships. To alleviate these problems, in this paper, we proposed a prior knowledge-guided transformer-GAN that unites triple domains of sinogram, image, and frequency to directly reconstruct SPET images from LPET sinograms, namely PK-TriDo. Our PK-TriDo consists of a Sinogram Inner-Structure-based Denoising Transformer (SISD-Former) to denoise the input LPET sinogram, a Frequency-adapted Image Reconstruction Transformer (FaIR-Former) to reconstruct high-quality SPET images from the denoised sinograms guided by the image domain prior knowledge, and an Adversarial Network (AdvNet) to further enhance the reconstruction quality via adversarial training. Specifically tailored for the PET imaging mechanism, we injected a sinogram embedding module that partitions the sinograms by rows and columns to obtain 1D sequences of angles and distances to faithfully preserve the inner-structure of the sinograms. Moreover, to mitigate high-frequency distortions and enhance reconstruction details, we integrated global-local frequency parsers (GLFPs) into FaIR-Former to calibrate the distributions and proportions of different frequency bands, thus compelling the network to preserve high-frequency details. Evaluations on three datasets with different dose levels and imaging scenarios demonstrated that our PK-TriDo outperforms the state-of-the-art methods.

CTA-Net: A gaze estimation network based on dual feature aggregation and attention cross fusion

Recent work has demonstrated the Transformer model is effective for computer vision tasks. However, the global self-attention mechanism utilized in Transformer models does not adequately consider the local structure and details of images, which may result in the loss of information and local details, causing decreased estimation accuracy in gaze estimation tasks when compared to convolution or sequential stacking methods. To address this issue, we propose a parallel CNNs-Transformer aggregation network (CTA-Net) for gaze estimation, which fully leverages the advantages of the Transformer model in modeling global context while the convolutional neural networks (CNNs) model in retaining local details. Specifically, Transformer and ResNet are deployed to extract facial and eye information, respectively. Additionally, an attention cross fusion (ACFusion) Block is embedded with CNN branch, which decomposes features in space and channels to supplement lost features, suppress noise, and help extract eye features more effectively. Finally, a dual-feature aggregation (DFA) module is proposed to effectively fuse the output features of both branches with the help feature a selection mechanism and a residual structure. Experimental results on the MPIIGaze and Gaze360 datasets demonstrate that our CTA-Net achieves state-of-the-art results.

Morphology, Phylogeny, and Evolution of the Rarely Known Genus Admetella McIntosh, 1885 (Annelida, Polynoidae) with Recognition of Four New Species from Western Pacific Seamounts

The polynoid genus Admetella constitutes a deep‐sea assemblage of polychaetes, notable for their large bodies adorned with antennal scales positioned dorsally to the bases of lateral antennae. Furthermore, the genus exhibits swimming proficiencies facilitated by elongated parapodia and flattened chaetae. Despite the frequent encounters with Admetella members during various deep‐sea explorations, a substantial gap in our comprehension of their diversity, phylogeny, and evolutionary trajectories still exists. Our thorough morphological and phylogenetic investigations of specimens obtained from three seamounts located in the tropical western Pacific have unveiled six species belonging to the genus Admetella, four of these being newly identified as Admetella multiseta sp. nov., A. levensteini sp. nov., A. nanhaiensis sp. nov., and A. undulata sp. nov. The other two species of Admetella remain unidentifiable at the species level due to the loss of crucial details. Our phylogenetic analysis, grounded on 13 mitochondrial protein‐coding genes and the inclusion of 12S, 16S, 18S, 28S rRNA, and ITS1–ITS2 genes, substantiates the monophyly of Admetella. Admetella is positioned at an intermediate node within the phylogenetic tree, situated between representative shallow‐water and deep‐sea subfamilies. The independent evolution of antennal scales within Admetella among polynoids constitutes a synapomorphy for this genus. Ancestral state reconstruction (ASR) analyses suggest that deep‐sea polynoids evolved from shallow‐water ancestors that possessed lateral antennae, which were subsequently lost in members inhabiting extreme marine environments, such as deep‐sea hydrothermal vents and anchialine caves. The analysis further implicates that swimming ability independently evolved at least four times within the Polynoidae family.

A new extraction method of underglaze brown decorative pattern based on the coupling of single scale gamma correction and gray sharpening.

In order to solve the problem of image quality and morphological characteristics of primary underglaze brown decorative pattern extraction, this paper proposes a method of primary underglaze brown decorative pattern extraction based on the coupling of single scale gamma correction and gray sharpening. The single-scale gamma correction is combined with the gray sharpening method. The single-scale gamma correction improves the contrast and brightness of the image by nonlinear transformation, but may lead to the loss of image detail. Gray sharpening can enhance the high frequency component and improve the clarity of the image, but it will introduce noise. Combining these two technologies can compensate for their shortcomings. The experimental results show that this method can improve the efficiency of last element underglaze brown decorative pattern extraction by enhancing the image retention detail and reducing the influence of noise. The experimental results showed that F1Score, Miou(%), Recall, Precision and Accuracy(%) were 0.92745, 0.82253, 0.97942, 0.92458 and 0.92745, respectively.

TrajectoryViz: Interactive visualization of treatment trajectories

Background and objectivesWith the proliferation of real-world or observational health data, there is increasing interest in studying treatment trajectories. The real-life treatment trajectories can be complex, and one has to simplify the patterns to draw any conclusions; however, oversimplification will cause the loss of essential details. Thus, the visualization challenge is to strike a balance between the two extremes. MethodsWe have implemented the observation of treatment trajectories starting from cohort definitions in cooperation with medical specialists, data processing, and then generating the interactive visualizations and detailed data tables derived from input data within an open-source R package as a Shiny dashboard. The created R package called TrajectoryViz (https://github.com/HealthInformaticsUT/TrajectoryViz) enables reproducible visual analysis and visual content generation for various data investigations and explanations. ResultsWe illustrate the use of the tool by assessing the sequence of events present within the data of cervical cancer prevention pathways, as well as the proportions of timely follow-up procedure events. ConclusionBuilding a toolset to access, manage, and analyze observational health data enables more accessible visual analysis of complicated data, adding time dimension to otherwise simplified event sequences that make up trajectories.

EWRD: Entropy‐Weighted Low‐Light Image Enhancement via Reverse Diffusion Model

Low‐light enhancement significantly aids vision tasks under poor illumination conditions. Existing methods primarily focus on enhancing severely degraded low‐light areas, improving illumination accuracy, or noise suppression, yet they often overlook the loss of additional details and color distortion during calculation. In this paper, we propose an innovative entropy‐weighted low‐light image enhancement method via the reverse diffusion model, aiming at addressing the limitations of the traditional Retinex decomposition model in preserving local pixel details and handling excessive smoothing issues. This method integrates an entropy‐weighting mechanism for improved image quality and entropy, along with a reverse diffusion model to address the detail loss in total variation regularization and refine the enhancement process. Furthermore, we utilize long short‐term memory networks for the learning reverse process and the simulation of image degradation, based on a thermodynamics‐based nonlinear anisotropic diffusion model. Comparative experiments reveal the superiority of our method over conventional Retinex‐based approaches in terms of detail preservation and visual quality. Extensive tests across diverse datasets demonstrate the exceptional performance of our method, evidencing its potential as a robust solution for low‐light image enhancement.

Common-Unique Decomposition Driven Diffusion Model for Contrast-Enhanced Liver MR Images Multi-Phase Interconversion.

All three contrast-enhanced (CE) phases (e.g., Arterial, Portal Venous, and Delay) are crucial for diagnosing liver tumors. However, acquiring all three phases is constrained due to contrast agents (CAs) risks, long imaging time, and strict imaging criteria. In this paper, we propose a novel Common-Unique Decomposition Driven Diffusion Model (CUDD-DM), capable of converting any two input phases in three phases into the remaining one, thereby reducing patient wait time, conserving medical resources, and reducing the use of CAs. 1) The Common-Unique Feature Decomposition Module, by utilizing spectral decomposition to capture both common and unique features among different inputs, not only learns correlations in highly similar areas between two input phases but also learns differences in different areas, thereby laying a foundation for the synthesis of remaining phase. 2) The Multi-scale Temporal Reset Gates Module, by bidirectional comparing lesions in current and multiple historical slices, maximizes reliance on previous slices when no lesions and minimizes this reliance when lesions are present, thereby preventing interference between consecutive slices. 3) The Diffusion Model-Driven Lesion Detail Synthesis Module, by employing a continuous and progressive generation process, accurately captures detailed features between data distributions, thereby avoiding the loss of detail caused by traditional methods (e.g., GAN) that overfocus on global distributions. Extensive experiments on a generalized CE liver tumor dataset have demonstrated that our CUDD-DM achieves state-of-the-art performance (improved the SSIM by at least 2.2% (lesions area 5.3%) comparing the seven leading methods). These results demonstrate that CUDD-DM advances CE liver tumor imaging technology.

A novel image enhancement method using retinex-based illumination map weighted guided filtering

Halo artifact, edge detail loss and noise amplification are the main problems in low illumination image enhancement, an image enhancement algorithm combining Retinex and illumination map weighted guided filtering is proposed. The traditional defogging physical models only enhance the images based on dark channels prior, resulting in different depths of field in local areas, and it can lead to some problems such as image overexposure and halo artifacts. To solve this problem, the method of combining light and dark channels is adopted to calculate the atmospheric light value and transmittance. For the problem that edge information is easily lost, the illumination gradient domain weighted guided filtering is utilized to improve the thinning transmittance. Experimental results with the proposed method have obvious improvement in denoising, halo elimination, brightness adjustment and edge preservation in the low-illumination image under different conditions.

Edge-Oriented Compressed Video Super-Resolution.

Due to the proliferation of video data in Internet of Things (IoT) systems, in order to reduce the data burden, most social media platforms typically employ downsampling to reduce the resolution of high-resolution (HR) videos before video coding. Consequently, the loss of detail and the introduction of additional artifacts seriously compromise the quality of experience (QoE). Recently, the task of compressive video super-resolution (CVSR) has garnered significant attention, aiming to simultaneously eliminate compression artifacts and enhance the resolution of compressed videos. In this paper, we propose an edge-oriented compressed video super-resolution network (EOCVSR), which focuses on reconstructing higher-quality details, to effectively address the CVSR task. Firstly, we devised a motion-guided alignment module (MGAM) to achieve precise bi-direction motion compensation in a multi-scale manner. Secondly, we introduced an edge-oriented recurrent block (EORB) to reconstruct edge information by combining the merits of explicit and implicit edge extraction. In addition, benefiting from the recurrent structure, the receptive field of EOCVSR can be enhanced and the features can be effectively refined without introducing additional parameters. Extensive experiments conducted on benchmark datasets demonstrate that our method surpasses the performance of state-of-the-art (SOTA) approaches in both quantitative and qualitative evaluations. Our approach can provide users with high-quality and cost-effective HR videos by integrating with sensors and codecs.

Multi-Level Attention Interactive Network for Cloud and Snow Detection Segmentation

The ground is typically hidden by cloud and snow in satellite images, which have a similar visible spectrum and complex spatial distribution characteristics. The detection of cloud and snow is important for increasing image availability and studying climate change. To address the issues of the low classification accuracy and poor generalization effect by the traditional threshold method, as well as the problems of the misdetection of overlapping regions, rough segmentation results, and a loss of boundary details in existing algorithms, this paper designed a Multi-level Attention Interaction Network (MAINet). The MAINet uses a modified ResNet50 to extract features and introduces a Detail Feature Extraction module to extract multi-level information and reduce the loss of details. In the last down-sampling, the Deep Multi-head Information Enhancement module combines a CNN and a Transformer structure to make deep semantic features more distinct and reduce redundant information. Then, the Feature Interactive and Fusion Up-sampling module enhances the information extraction of deep and shallow information and, then, guides and aggregates each to make the learned semantic features more comprehensive, which can better recover remote sensing images and increase the prediction accuracy. The MAINet model we propose performed satisfactorily in handling cloud and snow detection and segmentation tasks in multiple scenarios. Experiments on related data sets also showed that the MAINet algorithm exhibited the best performance.

Pan-Sharpening Network of Multi-Spectral Remote Sensing Images Using Two-Stream Attention Feature Extractor and Multi-Detail Injection (TAMINet)

Achieving a balance between spectral resolution and spatial resolution in multi-spectral remote sensing images is challenging due to physical constraints. Consequently, pan-sharpening technology was developed to address this challenge. While significant progress was recently achieved in deep-learning-based pan-sharpening techniques, most existing deep learning approaches face two primary limitations: (1) convolutional neural networks (CNNs) struggle with long-range dependency issues, and (2) significant detail loss during deep network training. Moreover, despite these methods’ pan-sharpening capabilities, their generalization to full-sized raw images remains problematic due to scaling disparities, rendering them less practical. To tackle these issues, we introduce in this study a multi-spectral remote sensing image fusion network, termed TAMINet, which leverages a two-stream coordinate attention mechanism and multi-detail injection. Initially, a two-stream feature extractor augmented with the coordinate attention (CA) block is employed to derive modal-specific features from low-resolution multi-spectral (LRMS) images and panchromatic (PAN) images. This is followed by feature-domain fusion and pan-sharpening image reconstruction. Crucially, a multi-detail injection approach is incorporated during fusion and reconstruction, ensuring the reintroduction of details lost earlier in the process, which minimizes high-frequency detail loss. Finally, a novel hybrid loss function is proposed that incorporates spatial loss, spectral loss, and an additional loss component to enhance performance. The proposed methodology’s effectiveness was validated through experiments on WorldView-2 satellite images, IKONOS, and QuickBird, benchmarked against current state-of-the-art techniques. Experimental findings reveal that TAMINet significantly elevates the pan-sharpening performance for large-scale images, underscoring its potential to enhance multi-spectral remote sensing image quality.