Search Results (603)

The advancement of Transformer models in computer vision has rapidly spurred numerous Transformer-based object detection approaches, such as DEtection TRansformer. Although DETR’s self-attention mechanism effectively captures the global context, it struggles with fine-grained detail detection, limiting its efficacy in small object detection where noise can easily obscure or confuse small targets. To address these issues, we propose Fuzzy System DNN-DETR involving two key modules: Fuzzy Adapter Transformer Encoder and Fuzzy Denoising Transformer Decoder. The fuzzy Adapter Transformer Encoder utilizes adaptive fuzzy membership functions and rule-based smoothing to preserve critical details, such as edges and textures, while mitigating the loss of fine details in global feature processing. Meanwhile, the Fuzzy Denoising Transformer Decoder effectively reduces noise interference and enhances fine-grained feature capture, eliminating redundant computations in irrelevant regions. This approach achieves a balance between computational efficiency for medium-resolution images and the accuracy required for small object detection. Our architecture also employs adapter modules to reduce re-training costs, and a two-stage fine-tuning strategy adapts fuzzy modules to specific domains before harmonizing the model with task-specific adjustments. Experiments on the COCO and AI-TOD-V2 datasets show that FSDN-DETR achieves an approximately 20% improvement in average precision for very small objects, surpassing state-of-the-art models and demonstrating robustness and reliability for small object detection in complex environments. Full article

Multistage text-to-image generation algorithms have shown remarkable success. However, the images produced often lack detail and suffer from feature loss. This is because these methods mainly focus on extracting features from images and text, using only conventional residual blocks for post-extraction feature processing. This results in the loss of features, greatly reducing the quality of the generated images and necessitating more resources for feature calculation, which will severely limit the use and application of optical devices such as cameras and smartphones. To address these issues, the novel High-Detail Feature-Preserving Network (HDFpNet) is proposed to effectively generate high-quality, near-realistic images from text descriptions. The initial text-to-image generation (iT2IG) module is used to generate initial feature maps to avoid feature loss. Next, the fast excitation-and-squeeze feature extraction (FESFE) module is proposed to recursively generate high-detail and feature-preserving images with lower computational costs through three steps: channel excitation (CE), fast feature extraction (FFE), and channel squeeze (CS). Finally, the channel attention (CA) mechanism further enriches the feature details. Compared with the state of the art, experimental results obtained on the CUB-Bird and MS-COCO datasets demonstrate that the proposed HDFpNet achieves better performance and visual presentation, especially regarding high-detail images and feature preservation. Full article

In recent years, deep-network-based hashing has gained prominence in image retrieval for its ability to generate compact and efficient binary representations. However, most existing methods predominantly focus on high-level semantic features extracted from the final layers of networks, often neglecting structural details that are crucial for capturing spatial relationships within images. Achieving a balance between preserving structural information and maximizing retrieval accuracy is the key to effective image hashing and retrieval. To address this challenge, we introduce Multiscale Deep Feature Fusion for Supervised Hashing (MDFF-SH), a novel approach that integrates multiscale feature fusion into the hashing process. The hallmark of MDFF-SH lies in its ability to combine low-level structural features with high-level semantic context, synthesizing robust and compact hash codes. By leveraging multiscale features from multiple convolutional layers, MDFF-SH ensures the preservation of fine-grained image details while maintaining global semantic integrity, achieving a harmonious balance that enhances retrieval precision and recall. Our approach demonstrated a superior performance on benchmark datasets, achieving significant gains in the Mean Average Precision (MAP) compared with the state-of-the-art methods: 9.5% on CIFAR-10, 5% on NUS-WIDE, and 11.5% on MS-COCO. These results highlight the effectiveness of MDFF-SH in bridging structural and semantic information, setting a new standard for high-precision image retrieval through multiscale feature fusion. Full article

Addressing the common issues of low brightness, poor contrast, and blurred details in images captured under conditions such as night, backlight, and adverse weather, we propose a zero-reference dual-path network based on multi-scale depth curve estimation for low-light image enhancement. Utilizing a no-reference loss function, the enhancement of low-light images is converted into depth curve estimation, with three curves fitted to enhance the dark details of the image: a brightness adjustment curve (LE-curve), a contrast enhancement curve (CE-curve), and a multi-scale feature fusion curve (MF-curve). Initially, we introduce the TCE-L and TCE-C modules to improve image brightness and enhance image contrast, respectively. Subsequently, we design a multi-scale feature fusion (MFF) module that integrates the original and enhanced images at multiple scales in the HSV color space based on the brightness distribution characteristics of low-light images, yielding an optimally enhanced image that avoids overexposure and color distortion. We compare our proposed method against ten other advanced algorithms based on multiple datasets, including LOL, DICM, MEF, NPE, and ExDark, that encompass complex illumination variations. Experimental results demonstrate that the proposed algorithm adapts better to the characteristics of images captured in low-light environments, producing enhanced images with sharp contrast, rich details, and preserved color authenticity, while effectively mitigating the issue of overexposure. Full article

Tensor restoration finds applications in various fields, including data science, image processing, and machine learning, where the global low-rank property is a crucial prior. As the convex relaxation to the tensor rank function, the traditional tensor nuclear norm is used by directly adding all the singular values of a tensor. Considering the variations among singular values, nonconvex regularizations have been proposed to approximate the tensor rank function more effectively, leading to improved recovery performance. In addition, the local characteristics of the tensor could further improve detail recovery. Currently, the gradient tensor is explored to effectively capture the smoothness property across tensor dimensions. However, previous studies considered the gradient tensor only within the context of the nuclear norm. In order to better simultaneously represent the global low-rank property and local smoothness of tensors, we propose a novel regularization, the Tensor-Correlated Total Variation (TCTV), based on the nonconvex Geman norm and total variation. Specifically, the proposed method minimizes the nonconvex Geman norm on singular values of the gradient tensor. It enhances the recovery performance of a low-rank tensor by simultaneously reducing estimation bias, improving approximation accuracy, preserving fine-grained structural details and maintaining good computational efficiency compared to traditional convex regularizations. Based on the proposed TCTV regularization, we develop TC-TCTV and TRPCA-TCTV models to solve completion and denoising problems, respectively. Subsequently, the proposed models are solved by the Alternating Direction Method of Multipliers (ADMM), and the complexity and convergence of the algorithm are analyzed. Extensive numerical results on multiple datasets validate the superior recovery performance of our method, even in extreme conditions with high missing rates. Full article

Optical image sequences of spacecraft acquired by space-based monocular cameras are typically imaged through exposure bracketing. The spacecraft feature deformable alignment network for multi-exposure image fusion (SFDA-MEF) aims to synthesize a High Dynamic Range (HDR) spacecraft image from a set of Low Dynamic Range (LDR) images with varying exposures. The HDR image contains details of the observed target in LDR images captured within a specific luminance range. The relative attitude of the spacecraft in the camera coordinate system undergoes continuous changes during the orbital rendezvous, which leads to a large proportion of moving pixels between adjacent frames. Concurrently, subsequent tasks of the In-Orbit Servicing (IOS) system, such as attitude estimation, are highly sensitive to variations in multi-view geometric relationships, which means that the fusion result should preserve the shape of the spacecraft with minimal distortion. However, traditional methods and unsupervised deep-learning methods always exhibit inherent limitations in dealing with complex overlapping regions. In addition, supervised methods are not suitable when ground truth data are scarce. Therefore, we propose an unsupervised learning framework for the multi-exposure fusion of optical spacecraft image sequences. We introduce a deformable convolution in the feature deformable alignment module and construct an alignment loss function to preserve its shape with minimal distortion. We also design a feature point extraction loss function to render our output more conducive to subsequent IOS tasks. Finally, we present a multi-exposure spacecraft image dataset. Subjective and objective experimental results validate the effectiveness of SFDA-MEF, especially in retaining the shape of the spacecraft. Full article

Evaluating the results of image denoising algorithms in Computed Tomography (CT) scans typically involves several key metrics to assess noise reduction while preserving essential details. Full Reference (FR) quality evaluators are popular for evaluating image quality in denoising CT scans. There is limited information about using Blind/No Reference (NR) quality evaluators in the medical image area. This paper shows the previously utilized Natural Image Quality Evaluator (NIQE) in CT scans; this NIQE is commonly used as a photolike image evaluator and provides an extensive assessment of the optimum NIQE setting. The result was obtained using the library of good images. Most are also part of the Convolutional Neural Network (CNN) training dataset against the testing dataset, and a new dataset shows an optimum patch size and contrast levels suitable for the task. This evidence indicates a possibility of using the NIQE as a new option in evaluating denoised quality to find improvement or compare the quality between CNN models. Full article

Background/Objectives: Mummy studies allow to reconstruct the characteristic of a population in a specific spatiotemporal context, in terms of living conditions, pathologies and death. Radiology represents an efficient diagnostic technique able to establish the preservation state of mummified organs and to estimate the patient's pathological conditions. However, the radiological approach shows some limitations. Although bone structures are easy to differentiate, soft tissue components are much more challenging, especially when they overlap. For this reason, computed tomography, a well-established approach that achieves optimal image contrast and three-dimensional reconstruction, has been introduced. This original article focuses attention on the role of virtual dissection as a promising technology for exploring human mummy anatomy and considers the potential of environmental scanning electron microscopy and X-ray spectroscopy as complementary approaches useful to understand the state of preservation of mummified remains. Methods: Ancient mummy corps have been analyzed through Anatomage Table 10 and environmental scanning electron microscope equipped with X-ray spectrometer; Results: Anatomage Table 10 through various volumetric renderings allows us to describe spine alteration due to osteoarthritis, dental state, and other clinical-pathological characteristics of different mummies. Environmental scanning electron microscope, with the advantage of observing mummified samples without prior specimen preparation, details on the state of tissue fragments. Skin, tendon and muscle show a preserved morphology and keratinocytes, collagen fibers and tendon structures are easily recognizable. Furthermore, X-ray spectrometer reveals in our tissue remains, the presence of compounds related to soil contamination. This investigation identifies a plethora of organic and inorganic substances where the mummies were found, providing crucial information about the mummification environment. Conclusions: These morphological and analytical techniques make it possible to study mummified bodies and describe their anatomical details in real size, in a non-invasive and innovative way, demonstrating that these interdisciplinary approaches could have great potential for improving knowledge in the study of ancient corpses. Full article

This paper presents a Regeneration filter for reducing near Salt-and-Pepper (nS&P) noise in images, designed for selective noise removal while simultaneously preserving structural details. Unlike conventional methods, the proposed filter eliminates the need for median or other filters, focusing exclusively on restoring noise-affected pixels through localized contextual analysis in the immediate surroundings. Our approach employs an iterative processing method, where additional iterations do not degrade the image quality achieved after the first filtration, even with high noise densities up to 97% spatial distribution. To ensure the results are measurable and comparable with other methods, the filter’s performance was evaluated using standard image quality assessment metrics. Experimental evaluations across various image databases confirm that our filter consistently provides high-quality results. The code is implemented in the R programming language, and both data and code used for the experiments are available in a public repository, allowing for replication and verification of the findings. Full article

Novel view synthesis aims to generate new perspectives from a limited number of input views. Neural Radiance Field (NeRF) is a key method for this task, and it produces high-fidelity images from a comprehensive set of inputs. However, a NeRF’s performance drops significantly with sparse views. To mitigate this, depth information can be used to guide training, with coarse depth maps often readily available in practical settings. We propose an improved sparse view NeRF model, ATGANNeRF, which integrates an enhanced U-Net generator with a dual-discriminator framework, CBAM, and Multi-Head Self-Attention mechanisms. The symmetric design enhances the model’s ability to capture and preserve spatial relationships, ensuring a more consistent generation of novel views. Additionally, local depth ranking is employed to ensure depth consistency with coarse maps, and spatial continuity constraints are introduced to synthesize novel views from sparse samples. SSIM loss is also added to preserve local structural details like edges and textures. Evaluation on LLFF, DTU, and our own datasets shows that ATGANNeRF significantly outperforms state-of-the-art methods, both quantitatively and qualitatively. Full article

Skull base tumors such as meningiomas and schwannomas are often pathologically benign. However, surgery for these tumors poses significant challenges because of their proximity to critical structures such as the brainstem, cerebral arteries, veins, and cranial nerves. These structures are compressed or encased by the tumor as they grow, increasing the risk of unintended injury to these structures, which can potentially lead to severe neurological deficits. Preoperative imaging is crucial for assessing the tumor size, location, and its relationship with adjacent vital structures. This study reviews advanced imaging techniques that allow detailed visualization of vascular structures and cranial nerves. Contrast-enhanced computed tomography and digital subtraction angiography are optimal for evaluating vascular structures, whereas magnetic resonance imaging (MRI) with high-resolution T2-weighted images and diffusion tensor imaging are optimal for evaluating cranial nerves. These methods help surgeons plan tumor resection strategies, including surgical approaches, more precisely. An accurate preoperative assessment can contribute to safe tumor resection and preserve neurological function. Additionally, we report the MRI contrast defect sign in skull base meningiomas, which suggests cranial nerve penetration through the tumor. This is an essential finding for inferring the course of cranial nerves completely encased within the tumor. These preoperative imaging techniques have the potential to improve the outcomes of patients with skull base tumors. Furthermore, this study highlights the importance of multimodal imaging approaches and discusses future directions for imaging technology that could further develop preoperative surgical simulations and improve the quality of complex skull base tumor surgeries. Full article

Image visibility is often degraded under challenging conditions such as low light, backlighting, and inadequate contrast. To mitigate these issues, techniques like histogram equalization, high dynamic range (HDR) tone mapping and near-infrared (NIR)–visible image fusion are widely employed. However, these methods have inherent drawbacks: histogram equalization frequently causes oversaturation and detail loss, while visible–NIR fusion requires complex and error-prone images. The proposed algorithm of a complementary cycle-consistent generative adversarial network (CycleGAN)-based training with visible and NIR images, leverages CycleGAN to generate fake NIR images by blending the characteristics of visible and NIR images. This approach presents tone compression and preserves fine details, effectively addressing the limitations of traditional methods. Experimental results demonstrate that the proposed method outperforms conventional algorithms, delivering superior quality and detail retention. This advancement holds substantial promise for applications where dependable image visibility is critical, such as autonomous driving and CCTV (Closed-Circuit Television) surveillance systems. Full article

Images captured under low-light conditions often suffer from serious degradation due to insufficient light, which adversely impacts subsequent computer vision tasks. Retinex-based methods have demonstrated strong potential in low-light image enhancement. However, existing approaches often directly design prior regularization functions for either illumination or reflectance components, which may unintentionally introduce noise. To address these limitations, this paper presents an enhancement method by integrating a Plug-and-Play strategy into an extended decomposition model. The proposed model consists of three main components: an extended decomposition term, an iterative reweighting regularization function for the illumination component, and a Plug-and-Play refinement term applied to the reflectance component. The extended decomposition enables a more precise representation of image components, while the iterative reweighting mechanism allows for gentle smoothing near edges and brighter areas while applying more pronounced smoothing in darker regions. Additionally, the Plug-and-Play framework incorporates off-the-shelf image denoising filters to effectively suppress noise and preserve useful image details. Extensive experiments on several datasets confirm that the proposed method consistently outperforms existing techniques. Full article

The Lichenotheca Veneta, published in 1869 by Vittore Trevisan (1818–1897), is one of the most relevant historic collections of exsiccatae of lichens in Italy. It contains a total of 268 specimens from 74 genera, 197 species, and 119 varieties and forms, organized into eight files and four volumes, and was probably conceived as the first of a series, even if no further volumes were published. To our knowledge, it is probably preserved in its complete and original form at the Natural History Museum of Venice only. Given its historical, cultural, and scientific relevance, it has been digitized and the resulting images and metadata have been published in a web portal. The digitization workflow comprised an initial digital imaging phase, followed by the extraction of specimens’ metadata from the specimens’ labels, and by a further digital imaging phase to capture specimens’ relevant details. The mobilization of metadata and images by means of digitization is widely recognized as an effective approach for enhancing the accessibility and usability of natural history collections. At the same time, since several inferences can be made without physically accessing the specimens, which, being biological objects, are intrinsically fragile, digitization contributes to reducing the risk of their deterioration. This contribution details the collection and its features, discussing the digitization process and its results. Full article

Pan-sharpening aims to generate high-resolution multispectral (HRMS) images by combining high-resolution panchromatic (PAN) images with low-resolution multispectral (LRMS) data, while maintaining the symmetry of spatial and spectral characteristics. Traditional convolutional neural networks (CNNs) struggle with global dependency modeling due to local receptive fields, and Transformer-based models are computationally expensive. Recent Mamba models offer linear complexity and effective global modeling. However, existing Mamba-based methods lack sensitivity to local feature variations, leading to suboptimal fine-detail preservation. To address this, we propose a Conditional Skipping Mamba Network (CSMN), which enhances global-local feature fusion symmetrically through two modules: (1) the Adaptive Mamba Module (AMM), which improves global perception using adaptive spatial-frequency integration; and (2) the Cross-domain Mamba Module (CDMM), optimizing cross-domain spectral-spatial representation. Experimental results on the IKONOS and WorldView-2 datasets demonstrate that CSMN surpasses existing state-of-the-art methods in achieving superior spectral consistency and preserving spatial details, with performance that is more symmetric in fine-detail preservation. Full article