Learning spatial relations for object-specific segmentation using Bayesian network model

Scene recognition aims to find a semantic explanation of a scene, i.e., it helps intelligent machines to know where they are. It can be widely applied into various tasks in computer vision and robotics. Most of pioneer methods extracted a set of low-level features and put them into classifier directly to identify scene category. But it has been proved that low-level features do not work well. Currently researchers aim to overcome the semantic gap between the low-level vision features and high-level semantic categories to improve the recognition performance. Therefore, much attention has been put on transforming low-level descriptors into richer intermediate representations. This paper proposed a novel method based on intermediate feature representation to solve the problem of recognizing the semantic category of scene image. This proposed method uses sparse coding on SIFT features and presents a spatial layout sensitive pooling method. The space layout for pooling is based on three rectangles with size of 1∗1,1∗4 and 4∗1 in each image. They are derived from inherent characteristics of the scene images by regularly dividing the image in horizontal and vertical direction. This spatial pooling strategy is easier and it can get optimal representation of scene images. Extreme learning machine (ELM) is used as a classifier. ELM has shown great ability to fit nonlinear classification boundaries. Experimental results have shown that this proposed method not only extracts lower dimension image feature but also outperforms other similar state-of-the-art methods in terms of recognition performance.

The influence of architectural design on well-being is increasingly recognised, underscoring the need to understand how humans evaluate their surroundings. This study examines the impact of low-level visual features and high-level architectural elements in images of house facades generated by a generative adversarial network (GAN). We introduce a novel method to create controlled image datasets using the style mixing component of StyleGAN2-ADA, enabling the combination of specific architectural features with styles representing low-level image properties, such as brightness or colour. Our dataset consists of 900 images encompassing five high-level features (house size, garage door visibility, number of windows, entrance door visibility, and roof shape) and 16 low-level visual features. These GAN-generated images were rated in an online experiment by 303 participants on six dimensions: facelikeness, hominess, relaxation, invitingness, safety, and price. Linear mixed-effects models identified significant predictors, including the number of floors as a high-level feature and green pixel percentage, saturation (SD), brightness (SD), and contrast as low-level features. This indicated that houses with two floors and images with an overall higher amount of green pixel, as well as diversity of saturation and brightness and a lower contrast were rated as, e.g., more inviting and safer. The analysis showed that high- and low-level features explained up to 54% of the variance in house price ratings, with high-level features accounting for the larger share and highlighting their relevance for house facade evaluations. This work advances the use of GANs and offers a method to modify visual elements that shape architectural evaluations.

Minimum training sample size requirements for achieving high prediction accuracy with the BN model: A case study regarding seismic liquefaction

In content based image retrieval (CIBR), low-level visual features are usually extracted from the highest quality and resolutions of visual contents. When content is scalable coded, their bit streams can be adapted at various nodes in multimedia usage chains to cater the variations in network bandwidths, display device resolutions and resources and usage preferences by just discarding insignificant resolution-quality layers. This can result in the existence of different version of the same content with dissimilar low-level visual features. Therefore, mapping of low level visual descriptors into content resolution- quality spaces is important in order to obtain low-level visual features that are resilient to content adaptations. A new scalable domain feature extraction using the compression modes and decisions is presented and its CBIR performance is evaluated. The proposed scheme outperforms MPEG-7 visual descriptors in both the original image and scaled resolution-quality space domains.

Previous work has proposed the ontology-based semi-automatic generation of antipattern Bayesian Network(BN) models. The generated BN model can be used to illustrate the effects of uncertainty on antipatterns using Bayesian propagation. This can guide users in detecting particular antipattern attributes of importance based on uncertain ontological information. However, the proposed approach has been implemented in the Protege ontology editor environment and requires human intervention to specify how the BN model will be generated. The fully automated generation of ontology-based antipattern BN models still remains an open issue. SPARSE is an OWL ontology based intelligent system that assists software project managers in the antipattern detection process. In this paper, we propose the use of the resulting detected antipatterns of SPARSE, their attributes (i.e. causes, symptoms, consequences) and the ontological relationships between these attributes, in order to automatically generate BN models of the detected antipatterns. We illustrate how this approach can be implemented using an example of 8 antipattern attributes of 6 inter-related antipatterns detected using SPARSE.

Analysis for warning factors of type 2 diabetes mellitus complications with Markov blanket based on a Bayesian network model

Weakly supervised video anomaly detection aims to identify abnormal events in video sequences without requiring frame-level supervision, which is a challenging task in computer vision. Traditional methods typically rely on low-level visual features with weak supervision from a single backbone branch, which often struggles to capture the distinctive characteristics of different categories. This limitation reduces their adaptability to real-world scenarios. In real-world situations, the boundary between normal and abnormal events is often unclear and context-dependent. For example, running on a track may be considered normal, but running on a busy road could be deemed abnormal. To address these challenges, RelVid is introduced as a novel framework that improves anomaly detection by expanding the relative feature gap between classes extracted from a single backbone branch. The key innovation of RelVid lies in the integration of auxiliary tasks, which guide the model to learn more discriminative features, significantly boosting the model’s performance. These auxiliary tasks—including text-based anomaly detection and feature reconstruction learning—act as additional supervision, helping the model capture subtle differences and anomalies that are often difficult to detect in weakly supervised settings. In addition, RelVid incorporates two other components, which include class activation feature learning for improved feature discrimination and a temporal attention module for capturing sequential dependencies. This approach enhances the model’s robustness and accuracy, enabling it to better handle complex and ambiguous scenarios. Evaluations on two widely used benchmark datasets, UCF-Crime and XD-Violence, demonstrate the effectiveness of RelVid. Compared to state-of-the-art methods, RelVid achieves superior performance in both detection accuracy and robustness.

ObjectiveThis study aims to identify relevant risk factors, assess the interactions between variables, and establish a predictive model for ischemic stroke (IS) in patients with cardiac myxoma (CM) using the Bayesian network (BN) approach.MethodsData of patients with CM were collected from three tertiary comprehensive hospitals in Beijing from January 2002 to January 2022. Age, sex, medical history, and information related to CM were extracted from the electronic medical record system. The BN model was constructed using the tabu search algorithm, and the conditional probability of each node was calculated using the maximum likelihood estimation method. The probability of each node of the network and the interrelationship between IS and its related factors were qualitatively and quantitatively analyzed. A receiver operating characteristic (ROC) curve was also plotted. Sensitivity, specificity, and area under the curve (AUC) values were calculated and compared between the BN and logistic regression models to evaluate the efficiency of the predictive model.ResultsA total of 416 patients with CM were enrolled in this study, including 61 with and 355 without IS. The BN model found that cardiac symptoms, systemic embolic symptoms, platelet counts, and tumor with high mobility were directly associated with the occurrence of IS in patients with CM. The BN model for predicting CM-IS achieved higher scores on AUC {0.706 [95% confidence interval (CI), 0.639–0.773]} vs. [0.697 (95% CI, 0.629–0.766)] and sensitivity (99.44% vs. 98.87%), but lower scores on accuracies (85.82% vs. 86.06%) and specificity (6.56% vs. 11.48%) than the logistic regression model.ConclusionCardiac symptoms, systemic embolic symptoms, platelet counts, and tumor with high mobility are candidate predictors of IS in patients with CM. The BN model was superior or at least non-inferior to the traditional logistic regression model, and hence is potentially useful for early IS detection, diagnosis, and prevention in clinical practice.

Stroke is a major chronic non-communicable disease with high incidence, high mortality, and high recurrence. To comprehensively digest its risk factors and take some relevant measures to lower its prevalence is of great significance. This study aimed to employ Bayesian Network (BN) model with Max-Min Hill-Climbing (MMHC) algorithm to explore the risk factors for stroke. From April 2019 to November 2019, Shanxi Provincial People's Hospital conducted opportunistic screening for stroke in ten rural areas in Shanxi Province. First, we employed propensity score matching (PSM) for class balancing for stroke. Afterwards, we used Chi-square testing and Logistic regression model to conduct a preliminary analysis of risk factors for stroke. Statistically significant variables were incorporated into BN model construction. BN structure learning was achieved using MMHC algorithm, and its parameter learning was achieved with Maximum Likelihood Estimation. After PSM, 748 non-stroke cases and 748 stroke cases were included in this study. BN was built with 10 nodes and 12 directed edges. The results suggested that age, fasting plasma glucose, systolic blood pressure, and family history of stroke constitute direct risk factors for stroke, whereas sex, educational levels, high density lipoprotein cholesterol, diastolic blood pressure, and urinary albumin-to-creatinine ratio represent indirect risk factors for stroke. BN model with MMHC algorithm not only allows for a complicated network relationship between risk factors and stroke, but also could achieve stroke risk prediction through Bayesian reasoning, outshining traditional Logistic regression model. This study suggests that BN model boasts great prospects in risk factor detection for stroke.

Multi-level feature representations for video semantic concept detection

To investigate form-related activity in motion-sensitive cortical areas, we recorded cell responses to animate implied motion in macaque middle temporal (MT) and medial superior temporal (MST) cortex and investigated these areas using fMRI in humans. In the single-cell studies, we compared responses with static images of human or monkey figures walking or running left or right with responses to the same human and monkey figures standing or sitting still. We also investigated whether the view of the animate figure (facing left or right) that elicited the highest response was correlated with the preferred direction for moving random dot patterns. First, figures were presented inside the cell's receptive field. Subsequently, figures were presented at the fovea while a dynamic noise pattern was presented at the cell's receptive field location. The results show that MT neurons did not discriminate between figures on the basis of the implied motion content. Instead, response preferences for implied motion correlated with preferences for low-level visual features such as orientation and size. No correlation was found between the preferred view of figures implying motion and the preferred direction for moving random dot patterns. Similar findings were obtained in a smaller population of MST cortical neurons. Testing human MT+ responses with fMRI further corroborated the notion that low-level stimulus features might explain implied motion activation in human MT+. Together, these results suggest that prior human imaging studies demonstrating animate implied motion processing in area MT+ can be best explained by sensitivity for low-level features rather than sensitivity for the motion implied by animate figures.

The present paper describes a query-by-sketch image retrieval system aimed at reducing the semantic gap by adopting relevance feedback. To reduce the semantic gap between low-level visual features and high-level semantics, in this content-based image retrieval system, users' sketches play an important role in relevance feedback. When users mark similar images of output images with "relevant" labels, the "relevant" images are relevant to the sketch image in positive feedback. This method was applied to 5,500 images in Corel Photo Gallery. Experimental results show that the proposed method is effective in retrieving images.

The semantic gap between low-level visual features and high-level semantics is a well-known challenge in content-based multimedia information retrieval. With the rapid popularization of social media, which allows users to assign tags to describe images and videos, attention is naturally drawn to take advantage of these metadata in order to bridge the semantic gap. This paper proposes a sparse linear integration (SLI) model that focuses on integrating visual content and its associated metadata, which are referred to as the content and the context modalities, respectively, for semantic concept retrieval. An optimization problem is formulated to approximate an instance using a sparse linear combination of other instances and minimize the difference between them. The prediction score of a concept for a test instance measures how well it can be reconstructed by the positive instances of that concept. Two benchmark image data sets and their associated tags are used to evaluate the SLI model. Experimental results show promising performance by comparing with the approaches based on a single modality and approaches based on popular fusion methods.

Climate and land-use changes are posing increasing threats to freshwater-related ecosystem services, acting both on the supply and demand sides. These changes disrupt critical processes such as nutrient cycling, sediment transport, and water flow regulation, leading to declining water quality and reduced ecosystem resilience. There is an urgent need for a deeper understanding of the dynamics of these threats, which can help enhance water management, environmental protection, and human well-being. To effectively tackle these risks, it is essential to quantitatively combine physical hazards and vulnerabilities by pinpointing hotspots where multiple stressors greatly increase the risk of water quality degradation. In response to this challenge, Bayesian network models offer a promising decision-support tool for evaluating adaptation options for water resource management, as they can integrate both quantitative and qualitative data. Building upon this approach, we developed a Spatial Bayesian Network (SBN) model to predict the probability of potential risks to water quality at the river basin scale in Italy and support the goal of achieving good chemical and ecological status according to the Water Framework Directive. This integrated model incorporates the complex relationships between land use change, climate change indicators (e.g., flood and drought intensity), and their combined impacts on water degradation. First, the baseline model uses historical patterns of climate change metrics from the CMCC DSS dataset and land use indicators from the Corine Land Cover as inputs to generate probabilistic predictions of potential risks to water quality. Then, the relationships between these variables are captured in their conditional probabilities, allowing for quantifying interactions and identifying key stressors, paving the way for scenario analysis. Finally, different future scenarios will be developed to predict the changes in water quality, considering projected climate data and socio-economic conditions. The outcome of this analysis contributes to developing an integrated management strategy that will help water managers make decisions and ultimately improve the resilience of freshwater ecosystems while supporting the implementation of adaptation strategies to address such problems.Keywords: Water Framework Directive, Water quality, Climate change, Land-use/ land-cover change Bayesian Network (BN) model

Survival prediction for gallbladder carcinoma after curative resection: Comparison of nomogram and Bayesian network models