Search Results (25,716)

The task of camouflaged object segmentation (COS) is a challenging endeavor that entails the identification of objects that closely blend in with their surrounding background. Furthermore, the camouflaged object’s obscure form and its subtle differentiation from the background present significant challenges during the feature extraction phase of the network. In order to extract more comprehensive information, thereby improving the accuracy of COS, we propose a diffusion model for a COS network that utilizes frequency domain information as auxiliary input, and we name it FreDiff. Firstly, we proposed a frequency auxiliary module (FAM) to extract frequency domain features. Then, we designed a Global Fusion Module (GFM) to make FreDiff pay attention to the global features. Finally, we proposed an Upsample Enhancement Module (UEM) to enhance the detailed information of the features and perform upsampling before inputting them into the diffusion model. Additionally, taking into account the specific characteristics of COS, we develop the specialized training strategy for FreDiff. We compared FreDiff with 17 COS models on the four challenging COS datasets. Experimental results showed that FreDiff outperforms or is consistent with other state-of-the-art methods under five evaluation metrics. Full article

Visually Impaired Persons (VIPs) have difficulty in recognizing vehicles used for navigation. Additionally, they may not be able to identify the bus to their desired destination. However, the bus bay in which the designated bus stops has not been analyzed in the existing literature. Thus, a guidance system for VIPs that identifies the correct bus for transportation is presented in this paper. Initially, speech data indicating the VIP’s destination are pre-processed and converted to text. Next, utilizing the Arctan Gradient-activated Recurrent Neural Network (ArcGRNN) model, the number of bays at the location is detected with the help of a Global Positioning System (GPS), input text, and bay location details. Then, the optimal bay is chosen from the detected bays by utilizing the Experienced Perturbed Bacteria Foraging Triangular Optimization Algorithm (EPBFTOA), and an image of the selected bay is captured and pre-processed. Next, the bus is identified utilizing a You Only Look Once (YOLO) series model. Utilizing the Sub-pixel Shuffling Convoluted Encoder–ArcGRNN Decoder (SSCEAD) framework, the text is detected and segmented for the buses identified in the image. From the segmented output, the text is extracted, based on the destination and route of the bus. Finally, regarding the similarity value with respect to the VIP’s destination, a decision is made utilizing the Multi-characteristic Non-linear S-Curve-Fuzzy Rule (MNC-FR). This decision informs the bus conductor about the VIP, such that the bus can be stopped appropriately to pick them up. During testing, the proposed system selected the optimal bay in 247,891 ms, which led to deciding the bus stop for the VIP with a fuzzification time of 34,197 ms. Thus, the proposed model exhibits superior performance over those utilized in prevailing works. Full article

(1) Background and Objectives: Flow assessment using cardiovascular magnetic resonance (CMR) provides important implications in determining physiologic parameters and clinically important markers. However, post-processing of CMR images remains labor- and time-intensive. This study aims to assess the validity and repeatability of fully automated segmentation of phase contrast velocity-encoded aortic root plane. (2) Materials and Methods: Aortic root images from 125 patients are segmented by artificial intelligence (AI), developed using convolutional neural networks and trained with a multicentre cohort of 160 subjects. Derived simple flow indices (forward and backward flow, systolic flow and velocity) and complex indices (aortic maximum area, systolic flow reversal ratio, flow displacement, and its angle change) were compared with those derived from manual contours. (3) Results: AI-derived simple flow indices yielded excellent repeatability compared to human segmentation (p < 0.001), with an insignificant level of bias. Complex flow indices feature good to excellent repeatability (p < 0.001), with insignificant levels of bias except flow displacement angle change and systolic retrograde flow yielding significant levels of bias (p < 0.001 and p < 0.05, respectively). (4) Conclusions: Automated flow quantification using aortic root images is comparable to human segmentation and has good to excellent repeatability. However, flow helicity and systolic retrograde flow are associated with a significant level of bias. Overall, all parameters show clinical repeatability. Full article

Accessibility analyses are important for public transit (PT) planning, as they reveal possible deficits in PT services. Since such accessibility analyses are highly time-dependent, the Modifiable Temporal Unit Problem (MTUP) should be considered. In this context, one approach is to calculate accessibility continuously for an entire day and aggregate it appropriately. However, this approach is complex and computationally intensive and is therefore rarely, if ever, applied. Instead, practitioners and researchers rely on simplified methods without considering temporal effects in detail. This paper bridges this gap by developing a simple yet representative method to account for the temporal variability of PT services. For this purpose, we calculate and compare PT accessibility for different time windows and through different aggregation methods for Germany. The results show that PT analyses between 9–11 a.m. were most representative. Alternatively, the time windows between 7–9 a.m. and between 1–3 p.m. adequately reflected accessibility. The median was suitable for aggregating individual time intervals into a representative value for the PT service throughout a day, while the maximum or mean value distorted the results. For practical planning purposes, we therefore recommend using the 9–11 a.m. time window. Full article

High-quality video object segmentation is a challenging visual computing task. Interactive segmentation can improve segmentation results. This paper proposes a multi-round interactive dynamic propagation instance-level video object segmentation network based on click interaction. The network consists of two parts: a user interaction segmentation module and a bidirectional dynamic propagation module. A prior segmentation network was designed in the user interaction segmentation module to better segment objects of different scales that users click on. The dynamic propagation network achieves high-precision video object segmentation through the bidirectional propagation and fusion of segmentation masks obtained from multiple rounds of interaction. Experiments on interactive segmentation datasets and video object segmentation datasets show that our method achieves state-of-the-art segmentation results with fewer click interactions. Full article

The integration of satellite data with deep learning has revolutionized various tasks in remote sensing, including classification, object detection, and semantic segmentation. Cloud segmentation in high-resolution satellite imagery is a critical application within this domain, yet progress in developing advanced algorithms for this task has been hindered by the scarcity of specialized datasets and annotation tools. This study addresses this challenge by introducing CloudLabel, a semi-automatic annotation technique leveraging region growing and morphological algorithms including flood fill, connected components, and guided filter. CloudLabel v1.0 streamlines the annotation process for high-resolution satellite images, thereby addressing the limitations of existing annotation platforms which are not specifically adapted to cloud segmentation, and enabling the efficient creation of high-quality cloud segmentation datasets. Notably, we have curated the Annotated Dataset for Training Cloud Segmentation (ADTCS) comprising 32,065 images (512 × 512) for cloud segmentation based on CloudLabel. The ADTCS dataset facilitates algorithmic advancement in cloud segmentation, characterized by uniform cloud coverage distribution and high image entropy (mainly 5–7). These features enable deep learning models to capture comprehensive cloud characteristics, enhancing recognition accuracy and reducing ground object misclassification. This contribution significantly advances remote sensing applications and cloud segmentation algorithms. Full article

Background: The accurate segmentation of the appendix with well-defined boundaries is critical for diagnosing conditions such as acute appendicitis. The manual identification of the appendix is time-consuming and highly dependent on the expertise of the radiologist. Method: In this study, we propose a fully automated approach to the detection of the appendix using deep learning architecture based on the U-Net with specific training parameters in CT scans. The proposed U-Net architecture is trained on an annotated original dataset of abdominal CT scans to segment the appendix efficiently and with high performance. In addition, to extend the training set, data augmentation techniques are applied for the created dataset. Results: In experimental studies, the proposed U-Net model is implemented using hyperparameter optimization and the performance of the model is evaluated using key metrics to measure diagnostic reliability. The trained U-Net model achieved the segmentation performance for the detection of the appendix in CT slices with a Dice Similarity Coefficient (DSC), Volumetric Overlap Error (VOE), Average Symmetric Surface Distance (ASSD), Hausdorff Distance 95 (HD95), Precision (PRE) and Recall (REC) of 85.94%, 23.29%, 1.24 mm, 5.43 mm, 86.83% and 86.62%, respectively. Moreover, our model outperforms other methods by leveraging the U-Net’s ability to capture spatial context through encoder–decoder structures and skip connections, providing a correct segmentation output. Conclusions: The proposed U-Net model showed reliable performance in segmenting the appendix region, with some limitations in cases where the appendix was close to other structures. These improvements highlight the potential of deep learning to significantly improve clinical outcomes in appendix detection. Full article

The accurate segmentation of cracks in structural materials is crucial for assessing the safety and durability of infrastructure. Although conventional segmentation models based on deep learning techniques have shown impressive detection capabilities in these tasks, their performance can be restricted by small amounts of training data. Data augmentation techniques have been proposed to mitigate the data availability issue; however, these systems often have limitations in texture diversity, scalability over multiple physical structures, and the need for manual annotation. In this paper, a novel generative artificial intelligence (GAI)-driven data augmentation framework is proposed to overcome these limitations by integrating a projected generative adversarial network (ProjectedGAN) and a multi-crack texture transfer generative adversarial network (MCT2GAN). Additionally, a novel metric is proposed to evaluate the quality of the generated data. The proposed method is evaluated using three datasets: the bridge crack library (BCL), DeepCrack, and Volker. From the simulation results, it is confirmed that the segmentation performance can be improved by the proposed method in terms of intersection over union (IoU) and Dice scores across three datasets. Full article

In recent years, active research has been conducted using deep learning to evaluate damage to aging bridges. However, this method is inappropriate for practical use because its performance deteriorates owing to numerous classifications, and it does not use photos of actual sites. To this end, this study used image data from an actual bridge management system as training data and employed a combined learning model for each member among various instance segmentation models, including YOLO, Mask R-CNN, and BlendMask. Meanwhile, techniques such as hyperparameter tuning are widely used to improve the accuracy of deep learning, and this study aimed to improve the accuracy of the existing model through this. The hyperparameters optimized in this study are DEPTH, learning rate (LR), and iterations (ITER) of the neural network. This technique can improve the accuracy by tuning only the hyperparameters while using the existing model for bridge damage identification as it is. As a result of the experiment, when DEPTH, LR, and ITER were set to the optimal values, mAP was improved by approximately 2.9% compared to the existing model. Full article

Monitoring of the vital signs or environment of disabled people is currently very popular because it increases their safety, improves their quality of life and facilitates remote care. The article proposes a system for automatic protection against burns based on the detection of thermal threats intended for blind or visually impaired people. Deep learning methods and CNNs were used to analyze images recorded by mobile thermal cameras. The proposed algorithm analyses thermal images covering the field of view of a user for the presence of objects with high or very high temperatures. If the user’s hand appears in such an area, the procedure warning about the possibility of burns is activated and the algorithm generates an alarm. To achieve this effect, the thermal images were analyzed using the 15-layered convolutional neural network proposed in the article. The proposed solution provided the efficiency of detecting threat situations of over 99% for a set of more than 21,000 images. Tests were carried out for various network configurations, architecture and both the accuracy and precision of hand detection was 99.5%, whereas sensitivity reached 99.7%. The effectiveness of burn risk detection was 99.7%—a hot object—and the hand appeared simultaneously in the image. The presented method allows for quick, effective and automatic warning against thermal threats. The optimization of the model structure allows for its use with mobile devices such as smartphones and mobile thermal imaging cameras. Full article

Colorectal polyps, precursors to colorectal cancer (CRC), require precise identification for appropriate diagnosis and therapy. This study aims to investigate the differences in hematological and inflammatory markers, specifically the CALLY index, HALP score, and immuno-inflammatory indexes, between neoplastic and nonneoplastic polyps. A retrospective cross-sectional study was conducted on 758 patients aged 61.0 ± 11.8 who underwent polypectomy between June 2021 and May 2024. Patients with colorectal adenocarcinoma (n = 22) were excluded. The polyps were classified into neoplastic and nonneoplastic categories based on histopathological evaluation. The study compared the CALLY index, HALP score, and various inflammatory indexes between neoplastic and nonneoplastic polyps. Out of 758 polyps analyzed, 514 were neoplastic, and 244 were nonneoplastic. Neoplastic polyps exhibited significantly lower CALLY and HALP scores (p < 0.05) and higher immuno-inflammatory indexes (p < 0.05) compared to nonneoplastic polyps. Dysplasia status, polyp diameter, and sigmoid colon localization were significant factors in determining neoplastic growth potential. No significant differences were observed in polyp localization in the proximal and distal colon segments or in solitary versus multiple polyps. The CALLY and HALP scores and immuno-inflammatory indexes can serve as valuable markers for distinguishing neoplastic from nonneoplastic polyps. These indexes reflect underlying inflammatory and immune responses, highlighting their potential utility in the early detection and risk stratification of colorectal polyps. Integrating these markers into clinical practice may enhance diagnostic accuracy and improve patient management, leading to timely interventions and better outcomes for individuals at risk of CRC. Full article

This study proposes a charging efficiency calculation model based on an equivalent internal resistance framework. A data-driven neural network model is developed to predict the charging efficiency of lithium titanate (LTO) batteries for 5% state of charge (SOC) segments under various charging conditions. By considering the impact of entropy change on the open-circuit voltage (OCV) during the charging process, the accuracy of energy efficiency calculations is improved. Incorporating battery data under various charging conditions, and comparing the predictive accuracy and computational complexity of different hyperparameter configurations, we establish a backpropagation neural network model designed for implementation in embedded systems. The model predicts the energy efficiency of subsequent 5% SOC segments based on the current SOC and operating conditions. The results indicate that the model achieves a prediction error of only 0.29% under unknown charging conditions while also facilitating the deployment of the neural network model in embedded systems. In future applications, the relevant predictive data can be transmitted in real time to the cooling system for thermal generation forecasting and predictive control of battery systems, thereby enhancing temperature control precision and improving cooling system efficiency. Full article

Background: The skin prick test (SPT) is used to diagnose sensitization to antigens. This study proposes a deep learning approach to infer wheal dimensions, aiming to reduce dependence on human interpretation. Methods: A dataset of SPT images (n = 5844) was used to infer a convolutional neural network for wheal segmentation (ML model). Three methods for inferring wheal dimensions were evaluated: the ML model; the standard protocol (MA1); and approximation of the area as an ellipse using diameters measured by an allergist (MA2). The results were compared with assisted image segmentation (AIS), the most accurate method. Bland–Altman analysis, distribution analyses, and correlation tests were applied to compare the methods. This study also compared the percentage deviation among these methods in determining the area of wheals with regular geometric shapes (n = 150) and with irregular shapes (n = 150). Results: The Bland–Altman analysis showed that the difference between methods was not correlated with the absolute area. The ML model achieved a segmentation accuracy of 85.88% and a strong correlation with the AIS method (ρ = 0.88), outperforming all other methods. Additionally, MA1 showed significant error (13.44 ± 13.95%) for pseudopods. Conclusions: The ML protocol can potentially automate the reading of SPT, offering greater accuracy than the standard protocol. Full article

Based on the approximation spaces, the interval-valued intuitionistic fuzzy rough set (IVIFRS) plays an essential role in coping with the uncertainty and ambiguity of the information obtained whenever human opinion is modeled. Moreover, a family of flexible t-norm (TNrM) and t-conorm (TCNrM) known as the Aczel–Alsina t-norm (AATNrM) and t-conorm (AATCNrM) plays a significant role in handling information, especially from the unit interval. This article introduces a novel clustering model based on IFRS using the AATNrM and AATCNrM. The developed clustering model is based on the aggregation operators (AOs) defined for the IFRS using AATNrM and AATCNrM. The developed model improves the level of accuracy by addressing the uncertain and ambiguous information. Furthermore, the developed model is applied to the segmentation problem, considering the information about the income and spending scores of the customers. Using the developed AOs, suitable customers are targeted for marketing based on the provided information. Consequently, the proposed model is the most appropriate technique for the segmentation problems. Furthermore, the results obtained at different values of the involved parameters are studied. Full article

The Aegean region and its graben system constitute one of Turkey’s most significant seismic zones. The faults within the Aegean graben generate numerous earthquakes, leading to various human and economic losses. To better understand the seismicity of western Anatolia, it is necessary to obtain concrete findings regarding the seismic history of earthquake-producing graben faults. This can be achieved through paleoseismological studies and other relevant disciplines. This study focuses on paleoseismological investigations along the northern boundary fault of the Acıgöl graben, located east of the Aegean graben system. The Maymundağı fault zone has been examined in two separate segments: east and west. The Dazkırı segment to the east shows evidence of movement dating back at least 10,000 years, with subsequent intensified activity observed later on the western Bozkurt segment. An earthquake occurred approximately 2370 years ago east of the Bozkurt segment, followed by movements migrating westward, resulting in earthquakes approximately 1322 and 598 years ago. Further analysis of the western segment indicates an average recurrence interval of 724 years for earthquakes, with a slip rate of 0.58 mm/year. Based on these findings, a future earthquake can be expected in this region around 2028–2129 AD. Full article