Background Image Research Articles

Level2image: A Utility for Making 2D Tile-Based Level Images with Overlays

2D tile-based levels are a common format for video game research, particularly in procedural content generation. Often, tiles are represented as text characters. Here, we describe level2image, a utility that provides a flexible means for converting such levels into a variety of image formats. It can use text tiles or substitute image tiles or backgrounds. There is support for various geometry overlays, such as paths through the level, areas of interest, or boundaries. The utility is a Python script, intended to reduce duplicated work creating such converters within the game research community.

Enhancing wildfire detection: a novel algorithm for controllable generation of wildfire smoke images

Background The lack of wildfire smoke image data is one of the most important factors hindering the development of image-based wildfire detection. Smoke image generation based on image inpainting techniques is a solution worthy of study. However, it is difficult to generate smoke texture with context consistency in complex backgrounds with current image inpainting methods. Aims This work aims to provide a wildfire smoke image database for specific scenarios. Methods We designed an algorithm based on generative adversarial networks (GANs) to generate smoke images. The algorithm includes a multi-scale fusion module to ensure consistency between the generated smoke and backgrounds. Additionally, a local feature-matching mechanism in the discriminator guides the generator to capture real smoke’s feature distribution. Key results We generated 13,400 wildfire smoke images based on forest background images and early fire simulation from the Fire Dynamics Simulator (FDS). Conclusions A variety of advanced object detection algorithms were trained based on the generated data. The experimental results confirmed that the addition of the generated data to the real datasets can effectively improve model performance. Implications This study paves a way for generating object datasets to enhance the reliability of watchtower or satellite wildfire monitoring.

GUing: A Mobile GUI Search Engine using a Vision-Language Model

Graphical User Interfaces (GUIs) are central to app development projects. App developers may use the GUIs of other apps as a means of requirements refinement and rapid prototyping or as a source of inspiration for designing and improving their own apps. Recent research has thus suggested retrieving relevant GUI designs that match a certain text query from screenshot datasets acquired through crowdsourced or automated exploration of GUIs. However, such text-to-GUI retrieval approaches only leverage the textual information of the GUI elements, neglecting visual information such as icons or background images. In addition, retrieved screenshots are not steered by app developers and lack app features that require particular input data. To overcome these limitations, this paper proposes GUing, a GUI search engine based on a vision-language model called GUIClip, which we trained specifically for the problem of designing app GUIs. For this, we first collected from Google Play app introduction images which display the most representative screenshots and are often captioned (i.e. labelled) by app vendors. Then, we developed an automated pipeline to classify, crop, and extract the captions from these images. This resulted in a large dataset which we share with this paper: including 303k app screenshots, out of which 135k have captions. We used this dataset to train a novel vision-language model, which is, to the best of our knowledge, the first of its kind for GUI retrieval. We evaluated our approach on various datasets from related work and in a manual experiment. The results demonstrate that our model outperforms previous approaches in text-to-GUI retrieval achieving a Recall@10 of up to 0.69 and a HIT@10 of 0.91. We also explored the performance of GUIClip for other GUI tasks including GUI classification and sketch-to-GUI retrieval with encouraging results.

Automatic Defects Recognition of Lap Joint of Unequal Thickness Based on X-Ray Image Processing

It is difficult to automatically recognize defects using digital image processing methods in X-ray radiographs of lap joints made from plates of unequal thickness. The continuous change in the wall thickness of the lap joint workpiece causes very different gray levels in an X-ray background image. Furthermore, due to the shape and fixturing of the workpiece, the distribution of the weld seam in the radiograph is not vertical which results in an angle between the weld seam and the vertical direction. This makes automatic defect detection and localization difficult. In this paper, a method of X-ray image correction based on invariant moments is presented to solve the problem. In addition, a novel background removal method based on image processing is introduced to reduce the difficulty of defect recognition caused by variations in grayscale. At the same time, an automatic defect detection method combining image noise suppression, image segmentation, and mathematical morphology is adopted. The results show that the proposed method can effectively recognize the gas pores in an automatic welded lap joint of unequal thickness, making it suitable for automatic detection.

Lower DTI-ALPS index in patients with major depressive disorder: Correlation with fatigue

Background and purposeDiffusion tensor imaging along perivascular spaces (DTI-ALPS) is an index that may provide insights into intracranial waste clearance processes. Glymphatic system dysfunction has been suggested to play a role in the development of major depressive disorder (MDD). Additionally, fatigue—a common precursor of MDD—is also closely connected to the waste clearance function of the central nervous system (CNS), further underscoring the significance of efficient waste removal in MDD. However, evidence linking altered DTI-ALPS index to MDD remains limited. This study aims to investigate the changes in the DTI-ALPS index in patients with MDD and explore the potential interplay between DTI-ALPS index alterations, fatigue, and the presence of MDD. Material and methodsA total of 46 patients with MDD and 55 healthy controls (HC) were included in the study. All participants underwent diffusion tensor imaging using the same 3-T MRI (3-Tesla Magnetic Resonance Imaging) scanner. The DTI-ALPS index was assessed, and the Chalder Fatigue Scale (CFS) was used to evaluate fatigue levels in both groups, and the 17-item Hamilton Depression Rating Scale (HAMD-17) was used to evaluate the severity of depression in the patients. We compared the DTI-ALPS index and clinical characteristics between the MDD and HC group, and explored the relationship among the DTI-ALPS index, CFS scores, and the presence of MDD through mediation analysis. ResultsThe DTI-ALPS index in the right hemisphere (DTI-ALPS-R) is significantly lower in patients with MDD (t = 2.41, P = 0.02). The MDD patients exhibited significantly higher scores on the CFS scales compared with HCs (t = 13.12, P <.001). Mediation analysis showed that the CFS score plays a significant mediating role between DTI-ALPS-R and the presence of MDD, acting as a full mediator (indirect effect β = −0.230, 95 % CI: [-0.388, −0.059]). ConclusionOur study found that patients with MDD have a reduced DTI-ALPS index. This reduction appears to contribute to the development of MDD by facilitating the accumulation of fatigue symptoms. These findings may provide a new perspective on the pathogenesis of MDD, suggest a potential new biomarker for MDD, and offer new insights for its treatment.

An automated pheochromocytoma and paraganglioma lesion segmentation AI-model at whole-body 68Ga- DOTATATE PET/CT.

Somatostatin receptor (SSR) targeting radiotracer 68Ga-DOTATATE is used for Positron Emission Tomography (PET)/Computed Tomography (CT) imaging to assess patients with Pheochromocytoma and paraganglioma (PPGL), rare types of Neuroendocrine tumor (NET) which can metastasize thereby becoming difficult to quantify. The goal of this study is to develop an artificial intelligence (AI) model for automated lesion segmentation on whole-body 3D DOTATATE-PET/CT and to automate the tumor burden calculation. 132 68Ga-DOTATATE PET/CT scans from 38 patients with metastatic and inoperable PPGL, were split into 70, and 62 scans, from 20, and 18 patients for training, and test sets, respectively. The training set was further divided into patient-stratified 5 folds for cross-validation. 3D-full resolution nnUNet configuration was trained with 5-fold cross-validation. The model's detection performance was evaluated at both scan and lesion levels for the PPGL test set and two other clinical cohorts with NET (n = 9) and olfactory neuroblastoma (ONB, n = 5). Additionally, quantitative statistical analysis of PET parameters including SUVmax, total lesion uptake (TLU), and total tumor volume (TTV), was conducted. The nnUNet AI model achieved an average 5-fold validation dice similarity coefficient of 0.84 at the scan level. The model achieved dice similarity coefficients (DSC) of 0.88, 0.6, and 0.67 at the scan level, the sensitivity of 86%, 61.13%, and 61.64%, and a positive predictive value of 89%, 74%, and 86.54% at the lesion level for the PPGL test, NET and ONB cohorts, respectively. For PPGL cohorts, smaller lesions with low uptake were missed by the AI model (p < 0.001). Anatomical region-based failure analysis showed most of the false negative and false positive lesions within the liver for all the cohorts, mainly due to the high physiologic liver background activity and image noise on 68Ga- DOTATATE PET scans. The developed deep learning-based AI model showed reliable performance for automated segmentation of metastatic PPGL lesions on whole-body 68Ga-DOTATATE-PET/CT images, which may be beneficial for tumor burden estimation for objective evaluation during therapy follow-up. https://www. gov/study/NCT03206060 , https://www. gov/study/NCT04086485 , https://www. gov/study/NCT05012098 .

Semantic Guided Matting Net

Abstract Human matting refers to extracting human parts from natural images with high quality, including human detail information such as hair, glasses, hats, etc. This technology plays an essential role in image synthesis and visual effects in the film industry. When the green screen is not available, the existing human matting methods need the help of additional inputs (such as trimap, background image, etc.), or the model with high computational cost and complex network structure, which brings great difficulties to the application of human matting in practice. To alleviate such problems, we use a segmentation network as the foundation and use multiple branches to achieve human segmentation, contour detail extraction, and information fusion. We also propose a foreground probability map module, which uses the feature maps in the segmentation network to pre-estimate the foreground probabilities of each pixel and obtain Semantic Guided Matting Net. Under the condition that only a single image is needed as the input, the human matting task can be realized by making full use of the semantic information in the image. We validate our method on the P3M-10k dataset. Compared with the benchmark, our method has made significant improvements in various evaluation indicators.

High-speed in vivo calcium recording using structured illumination with self-supervised denoising

High-speed widefield fluorescence imaging of neural activity in vivo is fundamentally limited by fluctuations in recorded signal due to background contamination and stochastic noise. In this study, we show background and shot noise-reduced imaging of the ultrafast genetically encoded Ca2+ indicator GCaMP8f in CA1 pyramidal neurons using periodic structured illumination (SI) with computational image reconstruction. We implement what we believe to be a novel reconstruction method for data acquired using periodic structured illumination, termed pseudo-HiLo (pHiLo), that combines a pseudo-widefield (pWF) reconstruction with individual SI frames to perform a HiLo reconstruction. We compare this new technique to interleaved optical sectioning structured illumination microscopy (OS-SIM) and pWF reconstruction. We quantify the performance of each reconstruction by evaluating contrast, transient peak-to-noise ratio (PNR), pairwise correlation coefficients between ΔF/F time courses extracted from individual in-focus cells, and correlation coefficients between each cell with surrounding cell-free background pixels. We additionally incorporate a self-supervised deep learning method for real-time noise suppression (DeepCAD-RT) into our data preprocessing pipeline. At 500 Hz frame rates, we demonstrate a 75% increase in PNR using the denoised pHiLo reconstruction compared to pWF. Utilizing DeepCAD-RT, we show significant PNR improvements using both structured illumination (SI) reconstruction methods with OS-SIM showing a 59% increase in PNR after denoising. Both pHiLo and OS-SIM reconstructions result in a ≈65% decrease in the mean correlation coefficient of the ΔF/F time courses between ROIs in comparison with pWF, indicating the potential to remove background fluorescent transients from out-of-focus cells.

A Small-Scale Object Detection Algorithm in Intelligent Transportation Scenarios

In response to the problem of poor detection ability of object detection models for small-scale targets in intelligent transportation scenarios, a fusion method is proposed to enhance the features of small-scale targets, starting from feature utilization and fusion methods. The algorithm is based on the YOLOv4 tiny framework and enhances the utilization of shallow and mid-level features on the basis of Feature Pyramid Network (FPN), improving the detection accuracy of small and medium-sized targets. In view of the problem that the background of the intelligent traffic scene image is cluttered, and there is more redundant information, the Convolutional Block Attention Module (CBAM) is used to improve the attention of the model to the traffic target. To address the problem of data imbalance and prior bounding box adaptation in custom traffic data sets that expand traffic images in COCO and VOC, we propose a Copy-Paste method with an improved generation method and a K-means algorithm with improved distance measurement to enhance the model’s detection ability for corresponding categories. Comparative experiments were conducted on a customized 260-thousand traffic data set containing public traffic images, and the results showed that compared to YOLOv4 tiny, the proposed algorithm improved mAP by 4.9% while still ensuring the real-time performance of the model.

The apparent diffusion coefficient can serve as a predictor of survival in patients with gliomas

Background and purposeMagnetic resonance imaging is indispensable for the preoperative diagnosis of glioma. This study aimed to investigate the role of the apparent diffusion coefficient values as predictors of survival in patients with gliomas.Methods and materialsA retrospective analysis was conducted on 101 patients with gliomas who underwent surgery between 2015 and 2020. Diffusion-weighted MRI was performed before the surgery. The regions of interest were categorized into parenchymal area, non-enhancing peritumoral area, and necrotic or cystic area. All the patients were divided into three subgroups: the parenchyma group, the non-enhancing peritumoral signal abnormality group, and the necrosis or cyst group. Univariate and multivariate analyses were performed using COX regression.ResultsIn the parenchymal group, Ki67, P53, IDH, and the high or low ADC values were identified as independent prognosticators for disease-free survival, while Ki67, IDH, and the high or low ADC values for overall survival. In the non-enhancing peritumoral signal abnormality group, Ki67, P53, IDH, and the ADC parenchymal area/ADC non−enhancing peritumoral area ratio were identified as independent prognostic factors for disease-free survival, while Ki67, IDH, and the ADC parenchymal area/ADC non−enhancing peritumoral area ratio for overall survival. In the necrosis or cyst group, Ki67 was significantly associated with disease-free survival, while Ki67 and the ADC value of the necrotic or cystic area for overall survival.ConclusionsThe ADC values, including the ADC value in the parenchymal area, the ADC parenchymal area/ADC non−enhancing peritumoral area ratio, and the ADC value in the necrotic or cystic area, can serve as an efficient and potential index for predicting the survival of patients with glioma.

ILN-SSR: Improved Logarithmic Norm and Sparse Structure Refinement for Infrared Small Target Detection

The effective discrimination of targets from backgrounds in environments characterized by a low signal-to-clutter ratio (SCR) is paramount for the advancement of infrared small target detection (IRSTD). In this work, we propose a novel detection framework predicated on low-rank sparse decomposition (LRSD), incorporating an improved logarithmic norm and a mechanism for sparse structure refinement, herein referred to as the improved logarithmic norm and sparse structure refinement (ILN-SSR). The ILN-SSR framework more precisely characterizes the sparse properties of both the background and the target, enabling a more effective distinction between the target and its background. Initially, our approach entails the utilization of an improved logarithmic norm to precisely estimate the low-rank attributes of the infrared image background. This is followed by the employment of a linear sparse regularization term alongside a target-traits-based sparse regularization term aimed at meticulously identifying targets within sparse regions and refining the sparse structure. Subsequently, we combine these components into the ILN-SSR framework, which formulates IRSTD as an optimization problem. The resolution of this framework is achieved through the implementation of the alternating direction method of multipliers (ADMM). The efficacy of the proposed framework is corroborated through the analysis of six image sequences. Comprehensive experimental assessments affirmed the framework’s substantial robustness in navigating various complex backgrounds.

Towards robust crop disease detection for complex real field background images

Most of the work done in image processing-based crop disease detection focuses on images with plain background. This paper presents a technique for crop disease detection for complex real field background images. A segmentation technique is presented to extract leaf patches from the entire image. Transform domain cepstral analysis is proposed for obtaining cepstral coefficients, to attain two level classifications. The first level classifies the crop species while the second level classifies the species into healthy leaf or leaf with specific type of disease. The work is tested on three crops Banana, Soybean and Grape and is checked on plain background laboratory images and on complex real field images. Suggested technique give species level accuracy of 94.33 %, 94.11 % and 98.44 % and disease level average accuracy of 97.75 %, 96.66 % and 97.95 % for Banana, Soybean and Grape, respectively. Comparison with standard features like texture and shape indicate that the presented technique gives the best results for both plain and complex background images suggesting its utilization in crop disease detection to reduce the agricultural and economic losses.

Improved algorithm for fracture-dissolution pore detection in resistivity imaging logging based on dung beetle optimization

Abstract Resistivity imaging logging has become a direct and precise method for visualizing the structural complexities of reservoir fractures and dissolution pores. The current use of Otsu's thresholding for segmentation results in poor segmentation quality and significant noise. Accurate segmentation of sub-images containing fracture and dissolution pore targets is essential for automated structure identification and subsequent parameter calculation. This study leverages the rapid convergence and robust global optimization capabilities of the Dung Beetle Optimizer to develop enhanced image segmentation approaches. Specifically, it introduces a refined K-Means algorithm for multi-category image segmentation and an Otsu algorithm for multi-threshold image segmentation, both optimized by the Dung Beetle Optimizer. Compared to conventional binary segmentation algorithms, this new algorithm effectively isolates noise and extracts multi-category information. Using the segmented sub-images, this paper integrates mathematical morphology techniques to compute parameters such as area, perimeter, tortuosity length, and pore shape factor for identified targets. Additionally, Principal Component Analysis is used to derive recognition factors for fractures and dissolution pores. Applications show that this factor can identify matrix, fracture, and dissolution pore targets in complex background images. By combining parameter information of the target area, the method effectively removes false information in resistivity imaging and segments sub-images of fractures and dissolution pores, calculating fracture area ratio, dissolution pore area ratio, and total area ratio.

Effects of Packaging on Consumer Buying Decision: A Study of African Black Soap (Alata Samina)

Product packaging plays a key role in consumer purchasing decision as packaging promote brand identity and customer loyalty. However, the African black soap prepared in Ghana has not seen the best of packaging, hence, its rejection in international marketplace. It is in this vein that the researcher sought to examine effect of packaging on consumer buying decision of African black soap (Alata Samina). Correlation and cross-sectional design were used. Cluster and simple random were used to select 245 respondents. Findings revealed that best packaging influence consumers’ perception to buy local Alata Samina soap recorded a cumulative mean of 4.69198 and standard deviation of 3.31842, suggesting respondents’ agreement that best packaging induce consumer buying decision. In terms of relationship, it was clear that strong positive link exist between packaging elements and consumer buying decision; Packaging colour had r(245) = 0.882, p&lt;0.05, Background image had r(245) = 0.881, p&lt;0.05, Packaging materials had r(245) = 0.882, p&lt;0.05, Font style had r(245) = 0.876, p&lt;0.05. Printed information had r(245) = 0.772, p&lt;0.05, Labelling, wrapper design and creativity had r(245) = 0.882, p&lt;0.05. Standard Multiple Regression was used to analyzed the effect of packaging elements on consumer buying decision and the result were packaging colour had 53.8% and statistically significant effect of r(245) = 0.538, p&lt;0.05. Brand image had 85.3 % and statistically significant effect of r(245) = 0.853, p&lt;0.05. Printed information had 73.8% and statistically significant effect of r(245) = 0.738, p&lt;0.05. Packaging materials had 62.6% and statistically significant effect of r(245) = 0.626, p&lt;0.05. Innovation had 92.3 % and statistically significant effect of r(245) = 0.923, p&lt;0.05. Labelling quality had 76.5% and statistically significant effect of r(245) = 0.765, p&lt;0.05. Design of wrapper had 44.2% and statistically significant effect of r(245) = 0.442, p&lt;0.05. However, font style had only 1.2% and this effect is not statistically significant; r(245) = 0.012, p&lt;0.05. This means that most consumers do not believe that font style influences consumers’ purchase decision. The study recommends that Institute of Packaging Ghana, Federation of Ghanaian Exporters, Ministry of Trade and Industry, Ghana Exports Promotions Authority and Association of Ghana Industries should collaborate to improve the packaging skills of the local African black soap producer. The essence is to enable them to package their products to meet the export market standards.

ScatterX： A software for fast processing of high-throughput small-angle scattering data

Abstract Scattering experiments become increasingly popular in modern scientific research, including the areas of materials, biology, chemistry, physics, etc. Besides, various types of scattering facilities have been developed recently, such as lab-based X-ray scattering equipment, national synchrotron facilities and large neutron facilities. These above-mentioned trends bring up fast-increasing data amounts of scattering data, as well as different scattering types (X-ray, neutron, laser and even microwaves). This struggles with data analysis software of the last generation, where data are usually analyzed assuming certain first-principle models and given data formats. To help researchers process and analyze scattering data more efficiently, we developed a general and model-free scattering data analysis software based on matrix operation, which has the unique advantage of high throughput scattering data processing, analysis and visualization. To maximize generality and efficiency, data processing is performed based on a three-dimensional matrix, where scattering curves are saved as matrices or vectors, rather than the traditional definition of paired values. It can not only realize image batch processing, background subtraction and correction, but also analyze data according to scattering theory and model, such as radius of gyration, fractal dimension and other physical quantities. In the aspect of visualization, the software allows the modify the color maps of two-dimensional scattering images and the gradual color variation of one-dimensional curves to suit efficient data communications. In all, this new software can work as a standalone platform for researchers to process, analyze and visualize scattering data from different research facilities without considering different file types or formats. All codes in this manuscript are open-sourced and can be easily implemented in matrix-based software, such as MATLAB, Python and Igor.

Deep learning networks-based tomato disease and pest detection: a first review of research studies using real field datasets.

Recent advances in deep neural networks in terms of convolutional neural networks (CNNs) have enabled researchers to significantly improve the accuracy and speed of object recognition systems and their application to plant disease and pest detection and diagnosis. This paper presents the first comprehensive review and analysis of deep learning approaches for disease and pest detection in tomato plants, using self-collected field-based and benchmarking datasets extracted from real agricultural scenarios. The review shows that only a few studies available in the literature used data from real agricultural fields such as the PlantDoc dataset. The paper also reveals overoptimistic results of the huge number of studies in the literature that used the PlantVillage dataset collected under (controlled) laboratory conditions. This finding is consistent with the characteristics of the dataset, which consists of leaf images with a uniform background. The uniformity of the background images facilitates object detection and classification, resulting in higher performance-metric values for the models. However, such models are not very useful in agricultural practice, and it remains desirable to establish large datasets of plant diseases under real conditions. With some of the self-generated datasets from real agricultural fields reviewed in this paper, high performance values above 90% can be achieved by applying different (improved) CNN architectures such as Faster R-CNN and YOLO.

Improved Empirical Backgrounds for JWST NIRISS Image/Wide-field Slitless Spectroscopy Data Reduction

The Near Infrared Imager and Slitless Spectrograph (NIRISS) on the James Webb Space Telescope (JWST) is a versatile instrument for collecting imaging and wide-field slitless spectroscopy (WFSS) data for surveys of galaxy clusters, emission-line galaxies, stellar populations, and more. Dispersed zodiacal light imprints distinct structures on space-based near-infrared imaging and WFSS observations, necessitating careful subtraction during JWST NIRISS data reduction. As of 2024 September 24 NIRISS WFSS calibration backgrounds introduce significant spatially dependent artifacts, up to 5% of the overall background level, which can severely affect data quality and following astronomical analysis. Notably, there are no existing backgrounds for NIRISS imaging data, which also show systematic artifacts, such as the “light saber” effect. In this work, we present improved empirical JWST NIRISS imaging and WFSS backgrounds derived from all available public data in the F115W, F150W, and F200W filters. We demonstrate that our empirical backgrounds provide a more accurate representation of the background structure in NIRISS imaging and WFSS data than existing reference files, mitigating the impact of spatially dependent artifacts. Our empirical backgrounds are publicly available and can be used to improve the quality of JWST NIRISS imaging and WFSS data reduction.

Accurate quantification of dislocation loops in complex functional alloys enabled by deep learning image analysis

In-depth statistics of individual defects observed during transmission electron microscopy (TEM) experiments are essential for the thorough characterization of materials. In this study, we aim to quantitatively characterize the population of dislocation loops in ion-irradiated CrFeMnNi alloys. To this end, we propose an efficient guideline to prepare TEM micrographs dataset for deep learning analysis, adapted for accurate characterization of microstructures produced by thousands of overlapping defects, a very common situation in TEM images, unfeasible by previous existing methods. To reduce human effort, we annotate only a few images and complement the database through a two-step process: initially, singular value decomposition to normalize image background, followed by a controlled data augmentation. The performed analysis provides precise quantitative information about the number of loops of different types, as well as their spatial distribution, their size, and the inter-object distances. These characteristics provide insights into the nucleation, mobility, and growth of dislocation loops, as well as the elastic anisotropy of the material. Our results emphasize how accurate analysis of complex microstructures can provide insights into the physical properties of materials and open up many perspectives for attaining quantitative information on materials properties based solely on their image analysis.

X-ray coronary angiography background subtraction by adaptive weighted total variation regularized online RPCA

Objective.X-ray coronary angiograms (XCA) are widely used in diagnosing and treating cardiovascular diseases. Various structures with independent motion patterns in the background of XCA images and limitations in the dose of injected contrast agent have resulted in low-contrast XCA images. Background subtraction methods have been developed to enhance the visibility and contrast of coronary vessels in XCA sequences, consequently reducing the requirement for excessive contrast agent injections.Approach.The current study proposes an adaptive weighted total variation regularized online RPCA (WTV-ORPCA) method, which is a low-rank and sparse subspaces decomposition approach to subtract the background of XCA sequences. In the proposed method, the images undergo initial preprocessing using morphological operators to eliminate large-scale background structures and achieve image homogenization. Subsequently, the decomposition algorithm decomposes the preprocessed images into background and foreground subspaces. This step applies an adaptive weighted TV constraint to the foreground subspace to ensure the spatial coherency of the finally extracted coronary vessel images.Main results.To evaluate the effectiveness of the proposed background subtraction method, some qualitative and quantitative experiments are conducted on two clinical and synthetic low-contrast XCA datasets containing videos from 21 patients. The obtained results are compared with six state-of-the-art methods employing three different assessment criteria. By applying the proposed method to the clinical dataset, the mean values of the global contrast-to-noise ratio, local contrast-to-noise ratio, structural similarity index, and reconstruction error (RE) are obtained as5.976,3.173,0.987, and0.026, respectively. These criteria over the low-contrast synthetic dataset were4.851,2.942,0.958, and0.034, respectively.Significance.The findings demonstrate the superiority of the proposed method in improving the contrast and visibility of coronary vessels, preserving the integrity of the vessel structure, and minimizing REs without imposing excessive computational complexity.

PBNet: Combining Transformer and CNN in Passport Background Texture Printing Image Classification

Passport background texture classification has always been an important task in border checks. Current manual methods struggle to achieve satisfactory results in terms of consistency and stability for weakly textured background images. For this reason, this study designs and develops a CNN and Transformer complementary network (PBNet) for passport background texture image classification. We first design two encoders by Transformer and CNN to produce complementary features in the Transformer and CNN domains, respectively. Then, we cross-wisely concatenate these complementary features to propose a feature enhancement module (FEM) for effectively blending them. In addition, we introduce focal loss to relieve the overfitting problem caused by data imbalance. Experimental results show that our PBNet significantly surpasses the state-of-the-art image segmentation models based on CNNs, Transformers, and even Transformer and CNN combined models designed for passport background texture image classification.