View Images Research Articles

DPCN++: Differentiable Phase Correlation Network for Versatile Pose Registration.

Pose registration is critical in vision and robotics. This paper focuses on the challenging task of initialization-free pose registration up to 7DoF for homogeneous and heterogeneous measurements. While recent learning-based methods show promise using differentiable solvers, they either rely on heuristically defined correspondences or require initialization. Phase correlation seeks solutions in the spectral domain and is correspondence-free and initialization-free. Following this, we propose a differentiable solver and combine it with simple feature extraction networks, namely DPCN++. It can perform registration for homo/hetero inputs and generalizes well on unseen objects. Specifically, the feature extraction networks first learn dense feature grids from a pair of homogeneous/heterogeneous measurements. These feature grids are then transformed into a translation and scale invariant spectrum representation based on Fourier transform and spherical radial aggregation, decoupling translation and scale from rotation. Next, the rotation, scale, and translation are independently and efficiently estimated in the spectrum step-by-step. The entire pipeline is differentiable and trained end-to-end. We evaluate DCPN++ on a wide range of tasks taking different input modalities, including 2D bird's-eye view images, 3D object and scene measurements, and medical images. Experimental results demonstrate that DCPN++ outperforms both classical and learning-based baselines, especially on partially observed and heterogeneous measurements.

Maintaining the Quality of the Institution by Implementing the Education Services Mix

Nowadays, the challenges for educational institutions are starting to increase in competition, so that there are many educational institutions offering services. An educational institution must be able to maintain quality so as to market the institution, maintain the value of its existence or existence. Various interesting events occur when implementing education, when accepting new prospective students, people are now starting to be smart in choosing educational institutions for their sons/daughters. The institution must have a good image in view. To create a brand image for educational institutions, an appropriate marketing strategy for educational services is needed, namely the 7P marketing theory. The elements of the educational services marketing mix consist of seven educational services marketing tools, which are also known as 7p; Product, Price, Place, Promotion, People, Physical Evidence, Process, Field of analysis; school environment, school products and services, customers or stakeholders, and school competitors. School competitive strategy; 1. Reduction of education costs, 2. Quality of education. 3. Educational innovation strategy.

Exploration of street space architectural color measurement based on street view big data and deep learning-A case study of Jiefang North Road Street in Tianjin.

Urban space architectural color is the first feature to be perceived in a complex vision beyond shape, texture and material, and plays an important role in the expression of urban territory, humanity and style. However, because of the difficulty of color measurement, the study of architectural color in street space has been difficult to achieve large-scale and fine development. The measurement of architectural color in urban space has received attention from many disciplines. With the development and promotion of information technology, the maturity of street view big data and deep learning technology has provided ideas for the research of street architectural color measurement. Based on this background, this study explores a highly efficient and large-scale method for determining architectural colors in urban space based on deep learning technology and street view big data, with street space architectural colors as the research object. We conducted empirical research in Jiefang North Road, Tianjin. We introduced the SegNet deep learning algorithm to semantically segment the street view images, extract the architectural elements and optimize the edges of the architecture. Based on K-Means clustering model, we identified the colors of the architectural elements in the street view. The accuracy of the building color measurement results was cross-sectionally verified by means of a questionnaire survey. The validation results show that the method is feasible for the study of architectural colors in street space. Finally, the overall coordination, sequence continuity, and primary and secondary hierarchy of architectural colors of Jiefang North Road in Tianjin were analyzed. The results show that the measurement model can realize the intuitive expression of architectural color information, and also can assist designers in the analysis of architectural color in street space with the guidance of color characteristics. The method helps managers, planners and even the general public to summarize the characteristics of color and dig out problems, and is of great significance in the assessment and transformation of the color quality of the street space environment.

Towards safer streets: A framework for unveiling pedestrians’ perceived road safety using street view imagery

Road safety has become a global concern but its impact in low- and middle-income countries is widespread mainly due to lack of appropriate crash database system and under-reporting. In this context, the primary objective of this paper is to provide a scalable framework for unveiling pedestrians’ perceived road safety that can also be applied in regions where accessible crash data are limited or near-crashes are left unreported. In the first step of our methodology, a deep learning architecture-based semantic segmentation model (HRNet+OCR) is trained using labeled Google Street View (GSV) images from specific study areas in Dhaka, Bangladesh, which facilitates the identification of both man-made components (such as roads, sidewalks, buildings, and vehicles) and natural elements (including trees and sky). The developed model showed excellent performance in identifying different features in an image by achieving high precision (0.95), recall (0.97), F1-score (0.96), and intersection over union (IoU) (91.86). Secondly, a group of trained raters scored the perceived road safety on an ordinal scale from 0 to 10 (extremely unsafe to extremely safe to walk in terms of road crashes) by assessing the GSV images. Then, several regression models have been used on features extracted from GSV images, and socio-demographic factors (i.e., population density, and relative wealth index) to estimate the perceived road safety, and random forest regression model was found to perform the best. Further, Shapley Additive Explanations (SHAP), a model-agnostic technique has been used for examining feature importance by computing the contribution of each feature to the random forest regression model output. The results show that sidewalk, road, population density, wall, and relative wealth index have higher impact on determining the perceived road safety rating. Additionally, the results of t-tests between the average perceived road safety scores for crash-prone and non crash-prone areas revealed the existence of significant differences. This study also provides perceived road safety rating map on a neighborhood scale, which can be a useful visualization tool for policy-makers and practitioners to identify the road safety deficiencies at specific locations, and formulate appropriate and strategic countermeasures to improve pedestrians’ road safety.

Crosswalk Detection from Satellite Imagery for Pedestrian Network Completion

Pedestrian network creation and completion for various tasks in transportation planning engineering generally requires laborious manual labeling of sidewalks and crosswalks. In this study, we employed the deep learning-based object detection model YOLO v5 for detecting crosswalks using Google Maps satellite imagery data of four North American cities. We also tested for two additional scenarios involving two image-processing techniques to exploit the special properties of crosswalks. Observations showed that the original images performed better than with binary thresholding and nearly the same even when noisy regions outside the right-of-way were removed. We proposed an algorithm to assign the detected crosswalks to intersection legs. We demonstrated the effectiveness of this technique for Washington, D.C. and Los Angeles, CA, showing classification accuracy rates of 71% and 89%, respectively. We also showed the influence of increasing distance threshold, a tolerance radius of prediction accuracy, in degrading the classification performance. This method, when combined with existing methods for sidewalk detection from aerial and street view images, could be reasonably used to help complete urban pedestrian networks in cities where high-resolution satellite images are available.

Advances in estimating pedestrian measures through artificial intelligence: From data sources, computer vision, video analytics to the prediction of crash frequency

Data are essential for planning walkable cities that are comfortable, convenient and safe to pedestrians. Yet, in contrast to massive vehicular traffic data, data on pedestrian traffic have not been systematically collected by municipal governments. Nowadays, geospatial big data provide rich information related to human activities and, hence, can capture street scenes in an innovative way. Using bus dashcam videos (on 244.36 km of roads covered by 33 bus routes in Hong Kong) and deep learning methods (Fast R-CNN and Deepsort), this study proposes a new method for estimating pedestrian volume from this data source. In comparison, we generate two alternative measures from household travel surveys and Google Street View images. The estimates are validated by manual counts at selected locations on a main road. Using five different modelling approaches (including three variants of negative binomial and two variants of random forest models), the pedestrian volume estimates are used for predicting pedestrian-vehicle crashes. The results show that pedestrian volumes calculated from bus dashcam videos consistently show comparable, if not better, performance in explaining crash frequency. In the future, different data sources should be used to supplement each other so that a more complete picture of pedestrian flows at the city level can be obtained.

Improving the prediction of Spreading Through Air Spaces (STAS) in primary lung cancer with a dynamic dual-delta hybrid machine learning model: a multicenter cohort study

BackgroundReliable pre-surgical prediction of spreading through air spaces (STAS) in primary lung cancer is essential for precision treatment and surgical decision-making. We aimed to develop and validate a dual-delta deep-learning and radiomics model based on pretreatment computed tomography (CT) image series to predict the STAS in patients with lung cancer.MethodSix hundred seventy-four patients with pre-surgery CT follow-up scans (with a minimum interval of two weeks) and primary lung cancer diagnosed by surgery were retrospectively recruited from three Chinese hospitals. The training cohort and internal validation cohort, comprising 509 and 76 patients respectively, were selected from Shanghai Chest Hospital; the external validation cohorts comprised 36 and 53 patients from two other centers, respectively. Four imaging signatures (classic radiomics features and deep learning [DL] features, delta-radiomics and delta-DL features) reflecting the STAS status were constructed from the pretreatment CT images by comprehensive methods including handcrafting, 3D views extraction, image registration and subtraction. A stepwise optimized three-step procedure, including feature extraction (by DL and time-base radiomics slope), feature selection (by reproducibility check and 45 selection algorithms), and classification (32 classifiers considered), was applied for signature building and methodology optimization. The interpretability of the proposed model was further assessed with Grad-CAM for DL-features and feature ranking for radiomics features.ResultsThe dual-delta model showed satisfactory discrimination between STAS and non-STAS and yielded the areas under the receiver operating curve (AUCs) of 0.94 (95% CI, 0.92–0.96), 0.84 (95% CI, 0.82–0.86), and 0.84 (95% CI, 0.83–0.85) in the internal and two external validation cohorts, respectively, with interpretable core feature sets and feature maps.ConclusionThe coupling of delta-DL model with delta-radiomics features enriches information such as anisotropy of tumor growth and heterogeneous changes within the tumor during the radiological follow-up, which could provide valuable information for STAS prediction in primary lung cancer.

E- Commerce Practices with Special Reference on Amazon Product

Online shopping is nowadays used everywhere or in every corner of the world and it is happening only because of customers directly make an order something they need through online shopping in Amazon. The meaning of online shopping is the process of buying goods and service from merchant over the internet. Amazon is one the site that customer used for purchasing many products. Online shopping makes easier for customer to choose different variety of product which they want, that the reason online shopping is popularly used nowadays. This paper made an research in online shopping through Amazon. Online shopping is a form of electronic commerce which allows consumers to directly buy goods and services from a seller over the internet using a web browser or a mobile app. Customer find a product of interest by visiting the website of the retailer directly or by searching among alternative vendors using a shopping search engine which display the same products availability and pricing at different e- retailer. A typical online store enables the customer to browse the firm’s range of product and services view photos or images of the product, along with information about the product specifications, features and prices. Online stores typically enable shoppers to use “search “features to find specific model, brand or items. It aims to find that the online shopping is easy to use and identify the offers and replacement of product factors.

Assessing the Effects of Subjective and Objective Measures on Housing Prices with Street View Imagery: A Case Study of Suzhou

The price of a house is affected by both the subjective and objective factors of the street environment in a neighborhood. However, the relationships between these factors and housing prices are not fully understood. Street view imagery (SVI) has recently emerged as a new data source for housing price studies. The SVI contains both objective and subjective information and can be used to extract objective measurements describing the physical environment and subjective measurements depicting human perceptions. Compared to conventional methods, there is consistency between subjective and objective information extracted from SVIs, and the two types of information are acquired from the perspective of the human visual perceptual system. Therefore, using both objective and subjective information extracted from street view images to study their relationship with housing prices has several advantages. In this study, focusing on the city of Suzhou, China, we extracted subjective perception and objective view indices from SVIs and systematically assessed their effects on housing prices. The global ordinary least squares (OLS) regression model and the local geographically weighted regression (GWR) model were used to model the correlations between these measures and housing prices. The OLS reveals that overall objective measures have stronger explanatory power, and built environment factors have a greater impact on housing prices. GWR shows that subjective factors can explain more variance in housing prices on the local scale and that home buyers care more about the subjective perceptions of the neighborhood’s surroundings. The map of the GWR local coefficients demonstrates that the perception indicators have both positive and negative effects on housing prices in different places. In addition, a Monte Carlo test was performed to verify the spatially varying relationships between these measures. Our findings provide important references for urban designers and guide various applications, such as safe neighborhood design and sustainable city planning.

Beneficial associations between outdoor visible greenness at the workplace and metabolic syndrome in Chinese adults

BackgroundGreenness surrounding residential places has been found to significantly reduce the risk of diseases such as hypertension, obesity, and metabolic syndrome (MetS). However, it is unclear whether visible greenness exposure at the workplace has any impact on the risk of MetS. MethodsVisible greenness exposure was assessed using a Green View Index (GVI) based on street view images through a convolutional neural network model. We utilized logistic regression to examine the cross-sectional association between GVI and MetS as well as its components among 51,552 adults aged 18–60 in the city of Hangzhou, China, from January 2018 to December 2021. Stratified analyses were conducted by age and sex groups. Furthermore, a scenario analysis was conducted to investigate the risks of having MetS among adults in different GVI scenarios. ResultsThe mean age of the participants was 40.1, and 38.5% were women. We found a statistically significant association between GVI and having MetS. Compared to the lowest quartile of GVI, participants in the highest quartile of GVI had a 17% (95% CI: 11–23%) lower odds of having MetS. The protective association was stronger in the males, but we did not observe such differences in different age groups. Furthermore, we found inverse associations between GVI and the odds of hypertension, low high-density lipoprotein cholesterol, obesity, and high levels of FPG. ConclusionsHigher exposure to outdoor visible greenness in the workplace environment might have a protective effect against MetS.

Comparing multiple machine learning models to investigate the relationship between urban morphology and PM2.5 based on mobile monitoring

The development of compact cities has deteriorated the air quality and threatened public health. To improve air quality, it is necessary to systematically analyze the complex relationship between urban morphology and air pollution to propose spatial optimization strategies. This study selected Nanshan District in Shenzhen, China as a case study, and used the street-level exposure to PM2.5 obtained from mobile monitoring as the dependent variable. Nine urban morphology parameters extracted from 3D building dataset and street view images, as well as other environmental parameters, were used as independent variables. A multiple linear regression model and six machine learning models were constructed for investigating the optimal one. For the study area, the results indicated that the RF and the XGBoost were the best-fit models with the R2 of 0.95 and 0.96 respectively. The frontal area index at 20 m height, the standard deviation of volume obstruction ratio, the building coverage ratio at 20 m height and the standard deviation of building height are the key urban form factors influencing PM2.5. In urban renewal and design practices, it is recommended to maintain the frontal area index within the range of 0.10–0.35, the standard deviation of volume obstruction ratio within the range of 4 × 10−5–8 × 10−5, and the standard deviation of building height below 48 m or above 64 m within the 1000 m buffer zone. These findings provide a generalizable analytical framework for understanding PM2.5 dispersion, and the differentiated applicability of models and parameters is recommended to be considered for different cities.

Salary Satisfaction of Employees at Workplace on a Large Area of Planted Land

Salary satisfaction (SS) perception by employees can be affected by psychological impacts from the workplace setting. Landscape attributes of green and blue spaces (GBS) may account for this effect, but relevant evidence is rarely verified. In this study, a total of 56 Chinese industrial parks were chosen as study sites, where employee satisfaction was assessed by rating facial expression scores (happy, sad, and neutral emotions) in photos obtained from social networks (Sina Weibo and Douyin). The structures of the GBSs were characterized remotely by largeness of size, height, and visible ratio of green view (GVI) in a 2 km radius buffer area around the workplace. Street view images from Baidu map were selected for estimating GVI using a pre-trained deep learning model and botanical experts evaluating woody plants’ diversity. The results indicated that SS can be estimated with the maximum likelihood analysis model against the happy score, which ranged within 8.37–18.38 (average: 13.30 ± 2.32) thousand RMB. A regression model indicated SS was lowered by a larger green space area in agreement with a reduced happy score. Further, sad scores in highland areas with tall plants and a strong depression on the happy score was associated with a greater plant diversity. Interesting from this study, the designed apparent size of green space should be considered in green space construction near a workplace to prevent perceptional decline towards SS, while blue space is irrelevant in this relationship. Similarly, the diversity of woody plants should be planned to control its negative impact on the perception of positive emotions, with plant diversity beyond a comfortable level perhaps further decreasing SS.

Classification of tree forms in aerial LiDAR point clouds using CNN for 3D tree modelling

ABSTRACT Three-dimensional models of trees that correspond to the real-world forms of the trees on the ground are used in urban planning, solar power estimation, and other disciplines. Previous studies have focused on generating 3D tree models from high-density point cloud data such as Terrestrial Laser Scanning (TLS) data, which is expensive and limited to small spatial extents. However, there has been limited exploration of inexpensive solutions to model trees over large spatial extents. The goal of this study is to use widely available discrete return Airborne Laser Scanning (ALS) data along with field-captured tree photographs and Google Street View (GSV) images to develop 3D equivalents of trees over larger spatial extents. To this end, we designed a process to assign representative 3D models for individual trees in discrete return ALS point clouds. This study demonstrates the use of a Convolutional Neural Network (CNN) model and 3D models generated with Structure from Motion (SfM) for the realistic modelling of deciduous non-overlapping trees from discrete return ALS data. We classified and labelled the crown shapes of deciduous trees in a study area into four classes based on GSV images of trees. We delineated and segmented non-overlapping deciduous trees from ALS data and reduced them to 2D images using voxel point counts. Next, we trained a CNN architecture to match the 2D images to the corresponding classes observed from GSV images. For each class, we created a representative 3D tree model using field-captured circumnavigational photos of trees and SfM. To demonstrate 3D visualization using the 3D tree models, we created a 3D visualization of the trees surrounding a parking lot. The trained CNN model from this study can be used to classify non-overlapping deciduous trees from discrete return ALS data and subsequently visualize near-realistic 3D tree models of trees.

Data Insights for Sustainable Cities: Associations between Google Street View-Derived Urban Greenspace and Google Air View-Derived Pollution Levels.

Unprecedented levels of urbanization have escalated urban environmental health issues, including increased air pollution in cities globally. Strategies for mitigating air pollution, including green urban planning, are essential for sustainable and healthy cities. State-of-the-art research investigating urban greenspace and pollution metrics has accelerated through the use of vast digital data sets and new analytical tools. In this study, we examined associations between Google Street View-derived urban greenspace levels and Google Air View-derived air quality, where both have been resolved in extremely high resolution, accuracy, and scale along the entire road network of Dublin City. Particulate matter of size fraction less than 2.5 μm (PM2.5), nitrogen dioxide, nitric oxide, carbon monoxide, and carbon dioxide were quantified using 5,030,143 Google Air View measurements, and greenspace was quantified using 403,409 Google Street View images. Significant (p < 0.001) negative associations between urban greenspace and pollution were observed. For example, an interquartile range increase in the Green View Index was associated with a 7.4% [95% confidence interval: -13.1%, -1.3%] decrease in NO2 at the point location spatial resolution. We provide insights into how large-scale digital data can be harnessed to elucidate urban environmental interactions that will have important planning and policy implications for sustainable future cities.

Three-Dimensional Reconstruction of Railway Bridges Based on Unmanned Aerial Vehicle–Terrestrial Laser Scanner Point Cloud Fusion

To address the incomplete image data collection of close-to-ground structures, such as bridge piers and local features like the suspension cables in bridges, obtained from single unmanned aerial vehicle (UAV) oblique photography and the difficulty in acquiring point cloud data for the top structures of bridges using single terrestrial laser scanners (TLSs), as well as the lack of textural information in TLS point clouds, this study aims to establish a high-precision, complete, and realistic bridge model by integrating UAV image data and TLS point cloud data. Using a particular large-scale dual-track bridge as a case study, the methodology involves aerial surveys using a DJI Phantom 4 RTK for comprehensive image capture. We obtain 564 images circling the bridge arches, 508 images for orthorectification, and 491 images of close-range side views. Subsequently, all images, POS data, and ground control point information are imported into Context Capture 2023 software for aerial triangulation and multi-view image dense matching to generate dense point clouds of the bridge. Additionally, ground LiDAR scanning, involving the placement of six scanning stations both on and beneath the bridge, was conducted and the point cloud data from each station are registered in Trimble Business Center 5.5.2 software based on identical feature points. Noise point clouds are then removed using statistical filtering techniques. The integration of UAV image point clouds with TLS point clouds is achieved using the iterative closest point (ICP) algorithm, followed by the creation of a TIN model and texture mapping using Context Capture 2023 software. The effectiveness of the integrated modeling is verified by comparing the geometric accuracy and completeness of the images with those obtained from a single UAV image-based model. The integrated model is used to generate cross-sectional profiles of the dual-track bridge, with detailed annotations of boundary dimensions. Structural inspections reveal honeycomb surfaces and seepage in the bridge piers, as well as painted rust and cracks in the arch ribs. The geometric accuracy of the integrated model in the X, Y, and Z directions is 1.2 cm, 0.8 cm, and 0.9 cm, respectively, while the overall 3D model accuracy is 1.70 cm. This method provides technical reference for the reconstruction of three-dimensional point cloud bridge models. Through 3D reconstruction, railway operators can better monitor and assess the condition of bridge structures, promptly identifying potential defects and damages, thus enabling the adoption of necessary maintenance and repair measures to ensure the structural safety of the bridges.

Visible outside view as a facilitation tool to evaluate view quality and shading systems through building openings

Viewing the natural environment from inside homes and workplaces has been recognized by a range of scholars as having an impact on improving health and well-being. Research has shown that a combination of outdoor elements – such as blue sky, sea view and greenery – is highly preferred as these elements are therapeutic for human wellbeing. However, installing shading systems is an important strategy for passive building cooling but it could affect our sense of connection to the outside environment. Most researchers evaluate view quality using qualitative questionnaires or quantitative methods by analyzing the geometry outside using 2D and 3D software which needs the outdoor environment to be fully built in the simulation accurately takes more time and may cause a system crash to run. This paper presents a new facilitation tool to quantify the visible outside view (VOV) by analyzing the outside view image by converting the view content into red, blue, and green (RGB) pixels using image processing technique. VOV measures the occupant's ray tracking percentage to the visible outside view content taking into consideration the blind factor of shading. An indicator start from 0 % to 100 % is given to quantify the outside view content including shading systems which then the overall VOV is related to the visible outside view quality as a factor of well-being potential (WP). The study found that the shading strategy should not be the same at all levels and shading devices in primary design stages considering the view to the natural elements positively affects occupants' wellbeing potential. These findings suggest that the proposed algorithm needs to be implemented with building energy and daylight simulation to produce more holistic systems. This will be the only way to get efficient and sustainable buildings highly connected with the human dimension.

An Improved New YOLOv7 Algorithm for Detecting Building Air Conditioner External Units from Street View Images.

Street view images are emerging as new street-level sources of urban environmental information. Accurate detection and quantification of urban air conditioners is crucial for evaluating the resilience of urban residential areas to heat wave disasters and formulating effective disaster prevention policies. Utilizing street view image data to predict the spatial coverage of urban air conditioners offers a simple and effective solution. However, detecting and accurately counting air conditioners in complex street-view environments remains challenging. This study introduced 3D parameter-free attention and coordinate attention modules into the target detection process to enhance the extraction of detailed features of air conditioner external units. It also integrated a small target detection layer to address the challenge of detecting small target objects that are easily missed. As a result, an improved algorithm named SC4-YOLOv7 was developed for detecting and recognizing air conditioner external units in street view images. To validate this new algorithm, we extracted air conditioner external units from street view images of residential buildings in Guilin City, Guangxi Zhuang Autonomous Region, China. The results of the study demonstrated that SC4-YOLOv7 significantly improved the average accuracy of recognizing air conditioner external units in street view images from 87.93% to 91.21% compared to the original YOLOv7 method while maintaining a high speed of image recognition detection. The algorithm has the potential to be extended to various applications requiring small target detection, enabling reliable detection and recognition in real street environments.

Anomalies of the Corpus Callosum in Prenatal Ultrasound: A Narrative Review for Diagnosis and Further Counseling.

The corpus callosum is the major interhemispheric tract that plays an important role in neurological function. Understanding the etiology and embryology development helps the ultrasound diagnosis for disorders of the corpus callosum and further counseling. The nonvisualization of cavum septum pellucidum or dysmorphic cavum septum pellucidum in axial view are indirect signs for beginners to diagnose complete agenesis of corpus callosum (cACC) and partial agenesis of the corpus callosum (pACC). Further coronal view, sagittal view, and fetal magnetic resonance imaging are also important for evaluation. Genetic testing plays an essential tool in anomalies of corpus callosum by revealing the underlying genetic pathophysiology, such as chromosomal anomalies and numerous monogenetic disorders in 30%-45% of ACC. Diagnosis and prediction of prognosis for hypoplasia or hyperplasia of the corpus callosum are more difficult compared to cACC and pACC because of the limited reports in the literature. However, the complex types often had poorer prognostic outcomes compared to the isolated types. Hence, it is important to evaluate and follow fetal conditions thoroughly to rule out intracranial or extracranial anomalies in other systems.

Cloud and Cloud Shadow Detection of GF-1 Images Based on the Swin-UNet Method

Cloud and cloud shadow detection in remote sensing images is an important preprocessing technique for quantitative analysis and large-scale mapping. To solve the problems of cloud and cloud shadow detection based on Convolutional Neural Network models, such as rough edges and insufficient overall accuracy, cloud and cloud shadow segmentation based on Swin-UNet was studied in the wide field of view (WFV) images of GaoFen-1 (GF-1). The Swin Transformer blocks help the model capture long-distance features and obtain deeper feature information in the network. This study selects a public GF1_WHU cloud and cloud shadow detection dataset for preprocessing and data optimization and conducts comparative experiments in different models. The results show that the algorithm performs well on vegetation, water, buildings, barren and other types. The average accuracy of cloud detection is 98.01%, the recall is 96.84% and the F1-score is 95.48%. The corresponding results of cloud shadow detection are 84.64%, 83.12% and 97.55%. In general, compared to U-Net, PSPNet and DeepLabV3+, this model performs better in cloud and cloud shadow detection, with clearer detection boundaries and a higher accuracy in complex surface conditions. This proves that Swin-UNet has great feature extraction capability in moderate and high-resolution remote sensing images.

Artificial intelligence in pediatric echocardiography-Automated view classification and image anonymization in rare cardiac malformations on the example of borderline HLHS

Abstract Background/Introduction Artificial intelligence (AI) has already much potential in adult echocardiography. However, there is still few experience in how to effectively implement AI methods in pediatric echocardiography. Since children with congenital heart defects show a large anatomical variance, AI-based image analysis tools adapted to this group are necessary. For the fundamentally needed steps of an automatic image data anonymization and a view classification there are still no reliable tools available. Purpose We have developed an anonymization tool for pediatric echocardiography data. For effective automated analysis of echocardiographies in children, we established an algorithm using bHLHS as an example that requires only small numbers of patients for a good view classification accuracy. Thus, we expect that the algorithm will also be applicable to other cardiac malformations. Method Raw data were obtained from visage imaging platforms and echocardiography examinations were conducted on Siemens machines. An anonymization tool was developed using the Python programming language and the Tkinter library. It converts medical images and crops the embedded data from the non-anatomical top region. The classification neural network was evaluated with data obtained from 5 patients with bHLHS. After data cleaning, each patient's image is labelled according to different Echo views: 4CH, SA, and LA, creating a total of 6600 images of 3 separate views. Due to the limited echocardiography dataset, a data augmentation approach was employed to improve the generalization of the AI model. The transfer learning approach was used to develop the AI model, which utilized a pre-trained VGG 13 model that was modified to classify the three different views of the pediatric echocardiography dataset. The modified model was fine-tuned on the dataset with a GPU-enhanced Intel i9 platform and Pycharm 2022.2.4. We use 70% dataset for training, 20% dataset for testing, and 10% dataset for validating. Results Table 1 shows that the expert clinician validated the anonymization tool by comparing identified patient information in the raw image and anonymized image. The trained pediatric echocardiography view classification model achieved an overall accuracy of 0.776 with precision values of 0.725, 0.889, and 0.775 for LA, SA and 4CH classes, respectively. The F1 score values were 0.835, 0.690, and 0.778 for the same classes. ROC curve were plotted (Figure 1) to visualize the performance of the model. Conclusion This study shows the importance of data anonymization and view classification in automating the diagnosis of HLHS using echocardiography images. The developed deep learning algorithm showed promising performance in accurately classifying different views of echocardiography images for diagnosing HLHS. Future studies can build upon this work by developing a complete diagnostic system to automatically analyze cardiac structures of patients with cardiac malformations.