Small Details Research Articles

TMLNet: Triad Multitask Learning Network for multiobjective based change detection

Change detection is an essential computer vision task in remote sensing applications. It faces challenges of image registration errors, variation in image capturing conditions, clouds, etc. It is observed that an accurate change detection task requires effective feature learning, which needs a noiseless representation. Moreover, change detection improves with error-free image reconstruction as an auxiliary task, which needs joint feature representation by a feature extractor. In this work, we proposed a novel triad (which is a combination of input images and its difference) learning based multiresolution architecture TMLNet for effective change detection. We also proposed a multi-context local self-attention module to efficiently calculate long-range pixel relations with multiple contexts. An enhanced backbone module with top-down connections and multi-scale channel and spatial attention is utilized for change map generation. It provides less noisy features extracted through the backbone. Laplacian pyramid loss is used to preserve small details in feature reconstruction. A set of comprehensive experimentations reveals that the proposed scheme achieved the state-of-the-art result for the F1 Score, intersection over union, and overall accuracy values in seven benchmark datasets. Our model code is available at https://github.com/chouhan-avinash/TMLNet.

Image Detection Network Based on Enhanced Small Target Recognition Details and Its Application in Fine Granularity

Image detection technology is of paramount importance across various fields. This significance is not only seen in general images with everyday scenes but also holds substantial research value in the field of remote sensing. Remote sensing images involve capturing images from aircraft or satellites. These images typically feature diverse scenes, large image formats, and varying imaging heights, thus leading to numerous small-sized targets in the captured images. Accurately identifying these small targets, which may occupy only a few pixels, is a challenging and active research area. Current methods mainly fall into two categories: enhancing small target features by improving resolution and increasing the number of small targets to bolster training datasets. However, these approaches often fail to address the core distinguishing features of small targets in the original images, thus resulting in suboptimal performance in fine-grained classification tasks. To address this situation, we propose a new network structure DDU (Downsample Difference Upsample), which is based on differential and resolution changing methods in the Neck layer of deep learning networks to enhance the recognition features of small targets, thus further improving the feature richness of recognition and effectively solving the problem of low accuracy in small target object recognition. At the same time, in order to take into account the recognition effect of targets of other sizes in the image, a new attention mechanism called PNOC (protecting the number of channels) is proposed, which integrates small target features and universal object features without losing the number of channels, thereby increasing the accuracy of recognition. And experimental verification was conducted on the PASCAL-VOC dataset. At the same time, it was applied to the testing of the fine-grained MAR20 dataset and found that the performance was better than other classic algorithms. At the same time, because the proposed framework belongs to a one-stage detection method, it has good engineering applicability and scalability, and universality in scientific research applications are good. Through comparative experiments, it was found that our algorithm improved the performance of the mAP by 0.7% compared to the original YOLOv8 algorithm.

P‐248: Late‐News Poster: Image Upscaling Techniques for IT Products; Evaluation and Analysis

This paper introduces an upscaler technology which is proven to be effective in improving power reduction for IT products at the system level. Proposed upscaler algorithm utilizes Lanczos and bilinear interpolation both of which are widely used in currently existing upscaler designs on the industry, and carefully analyzes the image distortion caused by 2x scaling of the input image and manipulates the possible artifacts by considering the key performance criteria in the up‐scaling. Our experimental results show that image quality distortion due to the proposed scaling algorithm is significantly improved while the small details in the images are preserved at most thanks to the combination of “blended upscale” and “edge enhancement” sub‐functions in the proposed upscaling design. According to our subjective evaluation, we got the average score of 4.39 over 15 persons with 21 images from well‐known image benchmark suites.

Prior function designs for X-ray reconstructions from limited angular views

The objectivity of three-dimensional imaging has placed X-ray Computed Tomography (CT) into an indispensable diagnostic method in clinical and industrial applications. Despite the advances in the X-ray CT imaging, high- resolution three-dimensional reconstruction of small details in objects with large in-plane dimensions has been a fundamental challenge for the conventional X-ray CT method. An alternative scanning geometry, where X-ray source and a detector move synchronously irradiating a specimen at an inclined angle, has been proposed to address this challenge. This work presents new designs of prior (or penalty) functions developed for reconstructing the radiography data obtained under the limited angular conditions. The new prior targets reduction of shape distortion artifacts, common for laminographic reconstructions, as well as preservation of small features, such as voids, edge details, and high-density inclusions. The novel prior function designs are demonstrated on the computational phantom of a specimen scanned under tilted planar CT geometry simulating large planar dimensions of a specimen. To accommodate physical CT scanning of such specimens, a novel microfocus X-ray CT scanning device that rotates X-ray source and moves the detector synchronously while staying on the same side of the test specimen (Inclined CT), is introduced. This device allows scanning of critical areas of up to 3-meter-wide and potentially unlimited length specimens with high resolution required for identification of manufacturing and structural defects typical for modern composite structural elements used in the aerospace industry.

Hi-gMISnet: generalized medical image segmentation using DWT based multilayer fusion and dual mode attention into high resolution pGAN

Objective. Automatic medical image segmentation is crucial for accurately isolating target tissue areas in the image from background tissues, facilitating precise diagnoses and procedures. While the proliferation of publicly available clinical datasets led to the development of deep learning-based medical image segmentation methods, a generalized, accurate, robust, and reliable approach across diverse imaging modalities remains elusive. Approach. This paper proposes a novel high-resolution parallel generative adversarial network (pGAN)-based generalized deep learning method for automatic segmentation of medical images from diverse imaging modalities. The proposed method showcases better performance and generalizability by incorporating novel components such as partial hybrid transfer learning, discrete wavelet transform (DWT)-based multilayer and multiresolution feature fusion in the encoder, and a dual mode attention gate in the decoder of the multi-resolution U-Net-based GAN. With multi-objective adversarial training loss functions including a unique reciprocal loss for enforcing cooperative learning in pGANs, it further enhances the robustness and accuracy of the segmentation map. Main results. Experimental evaluations conducted on nine diverse publicly available medical image segmentation datasets, including PhysioNet ICH, BUSI, CVC-ClinicDB, MoNuSeg, GLAS, ISIC-2018, DRIVE, Montgomery, and PROMISE12, demonstrate the proposed method’s superior performance. The proposed method achieves mean F1 scores of 79.53%, 88.68%, 82.50%, 93.25%, 90.40%, 94.19%, 81.65%, 98.48%, and 90.79%, respectively, on the above datasets, surpass state-of-the-art segmentation methods. Furthermore, our proposed method demonstrates robust multi-domain segmentation capabilities, exhibiting consistent and reliable performance. The assessment of the model’s proficiency in accurately identifying small details indicates that the high-resolution generalized medical image segmentation network (Hi-gMISnet) is more precise in segmenting even when the target area is very small. Significance. The proposed method provides robust and reliable segmentation performance on medical images, and thus it has the potential to be used in a clinical setting for the diagnosis of patients.

Effective three-step method for efficient correction of stripe noise and non-uniformity in infrared remote sensing images

The non-uniformity inherent in infrared detectors’ readout circuits often manifests as stripe noise, significantly impacting the interpretability and utility of infrared images in remote sensing applications. This paper introduces a novel three-step approach designed to overcome the challenges posed by stripe noise, offering a balance between real-time performance, detail preservation, and noise suppression. The proposed method involves subtracting the average of image columns from the noisy image, adding the wavelet denoised average signal to the subtraction result, and finally correcting the resulting image using an image-guidance mechanism. This unique three-step process ensures effective noise removal while preserving image details. The incorporation of wavelet transform leverages the sparsity of noise in the wavelet domain, enhancing denoising without introducing blurring. In a further refinement, the third step utilizes an image-guidance mechanism to recover small details with increased precision. This comprehensive approach addresses both stripe noise and non-uniformity, offering an easy, efficient, and fast technique for image correction. A comprehensive set of experiments, which involves comparisons with state-of-the-art algorithms, serves to substantiate the effectiveness and superior performance of the proposed method in real-world remote sensing and infrared images. Various examples, encompassing both real and artificial noise, are presented to showcase the robustness and applicability of our approach.

Deep-FG-DSM: Fine-Grained DSM for High-resolution UAV Imagery Based on Deep Implicit Occupancy Networks

Abstract. Nowadays, DSM (Digital Surface Model) is one of most important products that has been widely applied in digital city or smart city. Over the last decades, the common way is to follow the conventional photogrammetric pipeline for generating DSM, which is often not efficient and yields noise with small geometric details lost. Inspired by the development of deep implicit occupancy network, in this paper, we presented a learning-based method for obtaining fine-grained DSM, i.e., fine Deep-FG-DSM. In particular, high-resolution UAV imagery together with the corresponding original point cloud are employed to improve DSMs preserving higher details, two heads that embed the features of images and 3D points are deployed, and a MLP (Muti-layer Perceptron) is appended to decode these embedding into continuous occupancy probabilities for predicting the existence (or not) of surface point. Our experimental results demonstrate the robustness of our model against both sparse and noisy point clouds. While generating DSMs, it retains high-frequency details from high-resolution UAV images while maintaining relatively high accuracy. For point cloud obtained after simplification with average sampling resolution of d=5m, the MAE (Mean Absolute Residual Error) is 2.15m.

Segmentation of dermatoscopic images of skin lesions. Comparison of methods

The work discusses a number of techniques for segmenting dermoscopic images of skin lesions to identify the areas occupied by these lesions. Segmentation is necessary as the first stage of most methods of computer diagnostics of malignancy of neoplasms. A number of techniques, such as ABCDE, use the shape of the tumor as one of the criteria for making a diagnosis; for others, such as the use of convolutional neural networks, identifying the tumor allows one to increase the accuracy of the results obtained. The work discusses three methods of segmentation: thresholding using Otsu's method to calculate the threshold value, a convolutional neural network built on the U-net architecture, and a similar convolutional neural network with an added attention mechanism. The advantages and disadvantages of each method are considered, as well as the possibility of using them together to obtain the best segmentation results.The paper considers the application of an algorithm based on a morphological projector for determining structural differences for comparing dermoscopic images. This will allow to identify changes that have occurred in skin lesions over time, for a more accurate diagnosis of their malignancy. The proposed algorithm makes it possible to detect differences in images even if there is a significant difference in the brightness and color levels of the compared images, and also ignores small insignificant details, such as noise, dermatoscope optics marks, hair, etc. A method for correcting the desynchronization of images using the structural similarity index as a similarity metric, and the sine-cosine algorithm as an optimization algorithm is proposed. The proposed algorithms were tested on dermatoscopic images and the possibility of their application was demonstrated.

MFMENet: multi-scale features mutual enhancement network for change detection in remote sensing images

ABSTRACT Change detection, an important task in remote sensing image analysis, has been extensively studied in recent years. However, change detection still faces problems such as difficulty in detecting small targets and incomplete edge detection. Furthermore, pseudo changes such as seasonal changes can also lead to many false detections. In response to these challenges, we propose the Multi-scale Features Mutual Enhancement Network (MFMENet), a simple yet efficient network. MFMENet maximizes feature utilization through mutual guidance and supplementation, enhancing the detection capabilities for small targets and edges. First, we use a lightweight feature extraction network to extract features, which mitigates the information loss caused by continuous downsampling of an overly deep network structure. Then, we design a Context Adaptive Interaction Module (CAIM) to realize the complementarity of feature information at different levels. This facilitates shallow features in gaining more semantic information and deep features in acquiring more texture information, thereby enhancing the model’s capability to capture more comprehensive edge features while effectively mitigating interference from pseudo changes. Finally, we introduce a Feature Aggregation Comparison Module (FACM), which uses a combination of aggregation and comparison methods to refine and enhance features. FACM can not only highlight the changed features but also retain more details, improving the model’s detection ability of small targets and edge details. The full utilization of features and effective mutual enhancement of information ensure the improvement of MFMENet’s performance in small target and edge detection. Extensive experiments on three publicly available datasets (LEVIR, DSIFN, and CDD) demonstrate that our approach achieves superior performance with fewer parameters compared to state-of-the-art methods in recent years. In comparison to these baseline methods, our proposed approach achieves improvements of 0.98%, 12.24%, and 2.03% in the IOU metric on the LEVIR, DSIFN, and CDD datasets, respectively, while utilizing only 1.1 M parameters.

Microfacet rendering with diffraction compensation

AbstractThe traditional microfacet rendering models usually only consider the straight propagation of light and do not take into account the diffraction effect when calculating the radiance of outgoing light. However, ignoring the energy generated by diffraction can lead to darker rendering results when the object's surface has many small details. To address this issue, we introduce a diffraction energy term in the microfacet model to compensate for the energy loss caused by diffraction. Starting from the Fresnel‐Kirchhoff diffraction theorem, we combine it with the Cook‐Torrance model. By incorporating the computed diffraction radiance into the outgoing radiance of the microfacet, we obtain a diffraction‐compensated BRDF (Bidirectional Reflectance Distribution Function) model. Experimental results demonstrate that our proposed method has a significant effect in compensating for outgoing light and produces more realistic rendering results.

Drone photogrammetry as a tool for modern conflict archaeology: a case study of a Second World War armament bunker in Bavaria

ABSTRACT Within this paper, a case study on the application of drone photogrammetry at the Second World War armament bunker Weingut I is presented. The aim of the project was the mapping of the preserved remains in advance to the installation of a memorial site. In particular, the preserved concrete arch and one of the ventilation shafts were mapped by this method in high quality. Therefore, even small details of the building structure as well as several damage spots needing remediation became visible. One important advantage of drone prospecting is an enhanced occupational safety and a cost and time reduction as no further manual resources are necessary. In areas without finalized clearing for Unexploded Ordnance, other manual survey method would be especially dangerous. By comparing the photogrammetric model of the bunker Weingut I with corresponding ALS Digital Surface Models (DSM) of varying resolution, the increase in information is shown.

South Cuidrach

A new site in north Skye has revealed evidence of human occupation from the Late Upper Palaeolithic as well as the Early and Late Mesolithic periods, suggesting repeated visits by humans for at least 5,000 years from the Late Glacial Interstadial through the Loch Lomond Readvance and up to the mid-Holocene, a time of dynamic climatic and sea-level change. South Cuidrach is one of several sites in the north of the Isle of Skye and around the Inner Sound with lithic (stone tool) evidence diagnostic of a human presence before the mid-Holocene. These sites currently represent a unique concentration of pre-Late Mesolithic sites in Scotland. The lithic evidence which forms the backbone of the cultural and temporal characterisation of the site at South Cuidrach is outlined here. This includes characteristic points and microliths, burins, cores and other items diagnostic of specific cultural groups. Additionally, it has become evident that Rum bloodstone, a raw material well known from the Late Mesolithic period in the region, may be absent before the Early/Late Mesolithic transition, with baked mudstone from Staffin in North Skye being the primary raw material used before this time. The need for detailed knowledge of the region’s lithic technology and typology is well illustrated here, as small details frequently reflect important chronological differences.

Infertile Jordanian Women's Self-Perception About Societal Violence: An Interpretative Phenomenological Study.

Infertility affects one in six couples globally and is compounded by stigma and violence, particularly for women, in Jordan's culture. While existing research has illuminated societal pressures faced by infertile women, there is, yet no comprehensive understanding of the violence they encounter in their daily lives. This Interpretative Phenomenological study seeks to unravel the experiences of infertile women regarding societal violence in Jordan. By focusing on different types of community violence - physical, psychological, and emotional - The study aims to provide nuanced insights into the challenges these women confront. It also endeavors to identify contributing factors, including societal attitudes, cultural beliefs, and individual encounters, while informing policy and practice to mitigate this issue. Employing a qualitative approach, this study conducted semi-structured interviews with purposively sampled infertile women. Thematic analysis was utilized to uncover recurring patterns and themes, facilitating a comprehensive exploration of their experiences. Five main themes were identified: How the surrounding people view me as an infertile woman; I am suffocated by their questions; they interfere in the smallest details; I got burned and turned to ashes, and I have no right to complain; The problem of childbearing and the treatment plan is a matter for me and my husband only; and who supports me and what do I want from those around me? This study's implications are significant for policy and practice. By foregrounding the prevalent violence faced by infertile women, it underscores the urgency of interventions. Raising awareness, providing education, and extending support can counteract societal stigma and violence. Creating a more compassionate societal fabric can ensure a safer, more inclusive environment for these women.

An Improved Data Interpolating Empirical Orthogonal Function Method for Data Reconstruction: A Case Study of the Chlorophyll-a Concentration in the Bohai Sea, China

Chlorophyll-a (chl-a) serves as a key indicator in water quality and harmful algal blooms (HABs) research. While satellite ocean color data have greatly advanced chl-a research and HABs monitoring, missing data caused by cloud cover and other factors limit the spatiotemporal continuity and the utility of remote sensing data products. The Data Interpolating Empirical Orthogonal Function (DINEOF) method, widely used to reconstruct missing values in remote sensing datasets, is open to improvement in terms of computational accuracy and efficiency. We propose an improved method called Concentration-Stratified DINEOF (CS-DINEOF), which uses a coordinate–value correlative data division strategy to stratify the study area into several subregions based on annual average chl-a concentration. The proposed method clusters data points with similar spatiotemporal patterns, allowing for more targeted and effective reconstruction in each sub-dataset. The feasibility and advantage of the proposed method are tested and evaluated in the experiments of chl-a data reconstruction in the water of the Bohai Sea. Compared with the ordinary DINEOF method, the CS-DINEOF method improves the reconstruction accuracy, with an average Root Mean Square Error (RMSE) reduction of 0.0281 mg/m3, and saves computational time by 228.9%. Furthermore, the gap-free images generated from CS-DINEOF are able to illustrate small variations and details of the chl-a distribution in local areas. We can conclude that the proposed CS-DINEOF method is superior in providing significant insights for water quality and HABs studies in the Bohai Sea region.

Statistical Analysis of Measurement Processes Using Multi-Physic Instruments: Insights from Stitched Maps

Stitching methods allow one to measure a wider surface without the loss of resolution. The observation of small details with a better topographical representation is thus possible. However, it is not excluded that stitching methods generate some errors or aberrations on topography reconstruction. A device including confocal microscopy (CM), focus variation (FV), and coherence scanning interferometry (CSI) instrument modes was used to chronologically follow the drifts and the repositioning errors on stitching topographies. According to a complex measurement plan, a wide measurement campaign was performed on TA6V specimens that were ground with two neighboring SiC FEPA grit papers (P#80 and P#120). Thanks to four indicators (quality, drift, stability, and relevance indexes), no measurement drift in the system was found, indicating controlled stitching and repositioning processes for interferometry, confocal microscopy, and focus variation. Measurements show commendable stability, with interferometric microscopy being the most robust, followed by confocal microscopy, and then focus variation. Despite variations, robustness remains constant for each grinding grit, minimizing interpretation biases. A bootstrap analysis reveals time-dependent robustness for confocal microscopy, which is potentially linked to human presence. Despite Sa value discrepancies, all three metrologies consistently discriminate between grinding grits, highlighting the reliability of the proposed methodology.

Elastic slip in printing tasks under pressure

In the printing industry, the most common are offset printing and flexography. These types of printing have inevitable distortions in the reproduced image. Constant, inevitable distortion occurs due to the slippage of elastic surfaces in the contact strip during the printing process. In order to solve this problem, a full-scale experiment was conducted on printing register marks using technology developed and patented by Techkon. The experiment was aimed at studying the distortions that can occur when the plate and offset cylinders come into contact. As the surface of the plate and offset cylinders of printing machines, a rigid metal mold and an offset rubber-fabric fabric, which is a textolite, were used. As a result of the experiment, it was discovered that the distortion on these surfaces causes permanent and inevitable deformation of the raster dot. To describe this deformation, the classical Hertz contact problem was used, which is a widely used model in the field of contact mechanics. By approximating the obtained numerical values, a simple equation was obtained. This equation is practically useful for engineering calculations that can be used to introduce pre-emphasis on a rasterized original file. This, in turn, allows you to achieve a more accurate reproduced image in terms of matching colors, shapes and sizes of small details. This is especially important when products require high color and detail accuracy to ensure aesthetic appeal and functionality. High quality is in demand in the production of decorative elements such as wallpaper, laminate, paintings, and security printing -banknotes, stamps, excise taxes..

Description of a new genus and two new species of Darwinulidae (Crustacea, Ostracoda), from Christmas Island (Indian Ocean) with some considerations on the morphological evolution of ancient asexuals

Darwinulidae is believed to be one of the few metazoan taxa in which fully asexual reproduction might have persisted for millions of years. Although rare males in a single darwinulid species have recently been found, they may be non-functional atavisms. The representatives of this family are characterized by a slow evolutionary rate, resulting in a conservative morphology in the different lineages over long time frames and across wide geographic ranges. Differences between species and genera, although often based on small details of valve morphology and chaetotaxy, are nevertheless well-recognizable. Five Recent genera (Darwinula, Alicenula, Vestalenula, Penthesilenula and Microdarwinula) and about 35 living species, including also those left in open nomenclature, are included in this family. Previous phylogenetic analyses using both morphological characters and molecular data confirmed that the five genera are good phyletic units. Here, we report on the results of a study on darwinulid ostracods from Christmas Island (Indian Ocean). The taxonomic investigation led to the description of a new genus and two new species, namely Isabenula humphreysi gen. nov. sp. nov. and Vestalenula sp. E, this latter being left in open nomenclature, because only one individual was available. A new phylogeny of the family, using 30 morphological characters of 23 described species, is here presented. A key to the genera of Recent Darwinulidae, based on the morphology of valves and soft parts in adult females, is also provided. The discovery of Isabenula gen. nov., which shares intermediate morphological features with the genera Vestalenula and Penthesilenula, opens new opportunities to critically reconsider previous hypotheses on tempo and mode of evolution in the family Darwinulidae and on the phylogenetic relationships between its representatives.

Improved reconstruction for highly accelerated propeller diffusion 1.5T clinical MRI.

Propeller fast-spin-echo diffusion magnetic resonance imaging (FSE-dMRI) is essential for the diagnosis of Cholesteatoma. However, at clinical 1.5T MRI, its signal-to-noise ratio (SNR) remains relatively low. To gain sufficient SNR, signal averaging (number of excitations, NEX) is usually used with the cost of prolonged scan time. In this work, we leveraged the benefits of Locally Low Rank (LLR) constrained reconstruction to enhance the SNR. Furthermore, we enhanced both the speed and SNR by employing Convolutional Neural Networks (CNNs) for the accelerated PROPELLER FSE-dMRI on a 1.5T clinical scanner. Residual U-Net (RU-Net) was found to be efficient for propeller FSE-dMRI data. It was trained to predict 2-NEX images obtained by Locally Low Rank (LLR) constrained reconstruction and used 1-NEX images obtained via simplified reconstruction as the inputs. The brain scans from healthy volunteers and patients with cholesteatoma were performed for model training and testing. The performance of trained networks was evaluated with normalized root-mean-square-error (NRMSE), structural similarity index measure (SSIM), and peak SNR (PSNR). For 4 × under-sampled with 7 blades data, online reconstruction appears to provide suboptimal images-some small details are missing due to high noise interferences. Offline LLR enables suppression of noises and discovering some small structures. RU-Net demonstrated further improvement compared to LLR by increasing 18.87% of PSNR, 2.11% of SSIM, and reducing 53.84% of NRMSE. Moreover, RU-Net is about 1500 × faster than LLR (0.03 vs. 47.59s/slice). The LLR remarkably enhances the SNR compared to online reconstruction. Moreover, RU-Net improves propeller FSE-dMRI as reflected in PSNR, SSIM, and NRMSE. It requires only 1-NEX data, which allows a 2 × scan time reduction. In addition, its speed is approximately 1500 times faster than that of LLR-constrained reconstruction.

Dual-Band Image Fusion Approach Using Regional Weight Analysis Combined with a Multi-Level Smoothing Filter

Image fusion is an effective and efficient way to express the feature information of an infrared image and abundant detailed information of a visible image in a single fused image. However, obtaining a fused result with good visual effect, while preserving and inheriting those characteristic details, seems a challenging problem. In this paper, by combining a multi-level smoothing filter and regional weight analysis, a dual-band image fusion approach is proposed. Firstly, a series of dual-band image layers with different details are obtained using smoothing results. With different parameters in a bilateral filter, different smoothed results are achieved at different levels. Secondly, regional weight maps are generated for each image layer, and then we fuse the dual-band image layers with their corresponding regional weight map. Finally, by imposing proper weights, those fused image layers are synthetized. Through comparison with seven excellent fusion methods, both subjective and objective evaluations for the experimental results indicate that the proposed approach can produce the best fused image, which has the best visual effect with good contrast, and those small details are preserved and highlighted, too. In particular, for the image pairs with a size of 640 × 480, the algorithm could provide a good visual effect result within 2.86 s, and the result has almost the best objective metrics.

Spectrum learning for super-resolution tomographic reconstruction.

Objective. Computed Tomography (CT) has been widely used in industrial high-resolution non-destructive testing. However, it is difficult to obtain high-resolution images for large-scale objects due to their physical limitations. The objective is to develop an improved super-resolution technique that preserves small structures and details while efficiently capturing high-frequency information.Approach. The study proposes a new deep learning based method called spectrum learning (SPEAR) network for CT images super-resolution. This approach leverages both global information in the image domain and high-frequency information in the frequency domain. The SPEAR network is designed to reconstruct high-resolution images from low-resolution inputs by considering not only the main body of the images but also the small structures and other details. The symmetric property of the spectrum is exploited to reduce weight parameters in the frequency domain. Additionally, a spectrum loss is introduced to enforce the preservation of both high-frequency components and global information.Main results. The network is trained using pairs of low-resolution and high-resolution CT images, and it is fine-tuned using additional low-dose and normal-dose CT image pairs. The experimental results demonstrate that the proposed SPEAR network outperforms state-of-the-art networks in terms of image reconstruction quality. The approach successfully preserves high-frequency information and small structures, leading to better results compared to existing methods. The network's ability to generate high-resolution images from low-resolution inputs, even in cases of low-dose CT images, showcases its effectiveness in maintaining image quality.Significance. The proposed SPEAR network's ability to simultaneously capture global information and high-frequency details addresses the limitations of existing methods, resulting in more accurate and informative image reconstructions. This advancement can have substantial implications for various industries and medical diagnoses relying on accurate imaging.