Detection Paradigm Research Articles

SeFi-CD: A Semantic First Change Detection Paradigm That Can Detect Any Change You Want

The existing change detection (CD) methods can be summarized as the visual-first change detection (ViFi-CD) paradigm, which first extracts change features from visual differences and then assigns them specific semantic information. However, CD is essentially dependent on change regions of interest (CRoIs), meaning that the CD results are directly determined by the semantics changes in interest, making its primary image factor semantic of interest rather than visual. The ViFi-CD paradigm can only assign specific semantics of interest to specific change features extracted from visual differences, leading to the inevitable omission of potential CRoIs and the inability to adapt to different CRoI CD tasks. In other words, changes in other CRoIs cannot be detected by the ViFi-CD method without retraining the model or significantly modifying the method. This paper introduces a new CD paradigm, the semantic-first CD (SeFi-CD) paradigm. The core idea of SeFi-CD is to first perceive the dynamic semantics of interest and then visually search for change features related to the semantics. Based on the SeFi-CD paradigm, we designed Anything You Want Change Detection (AUWCD). Experiments on public datasets demonstrate that the AUWCD outperforms the current state-of-the-art CD methods, achieving an average F1 score 5.01% higher than that of these advanced supervised baselines on the SECOND dataset, with a maximum increase of 13.17%. The proposed SeFi-CD offers a novel CD perspective and approach.

Defect intelligent recognition of membrane product based on deep learning

Defect detection plays a crucial role in the manufacturing industry, ensuring the quality of industrial products. Despite advancements in this field, current defect detection methods face two primary challenges: (1) extracting visually similar features from the background poses difficult and (2) these methods struggle to identify tiny defects in the target objects. We present a feature enhancement module called the SE-CAR, which aims to handle the identified problems effectively. This module is designed to efficiently capture tiny defects in images, prioritize defect information features, and ultimately enhance the model’s predictive performance and accuracy in defect recognition tasks. In addition, Distance-IoU Non-Maximum Suppression is employed as a substitution for the original Non-Maximum Suppression. This enhances the recognition accuracy of bounding boxes, ensuring that the model maintains high detection accuracy even after complex scenarios. Moreover, the proposed methodology exhibits broad applicability across a wide spectrum of prevalent defect detection paradigms. In empirical experiments, we employ the YOLOv7 architecture as the foundation framework, integrating the proposed methodologies for the purpose of detecting defects in the membrane. The empirical evidence demonstrates a notable improvement in the performance of detecting defects in membrane products using the SE-CAR feature enhancement module and Distance-IoU Non-Maximum Suppression algorithm. In comparison to baseline networks, an increase in mAp by 6.29%, precision by 1.63%, and recall by 8.76%.

The Effects of the Complexity of 3D Virtual Objects on Visual Working Memory Capacity in AR Interface for Mobile Phones

The current study aims to investigate the effects of 3D virtual object complexity on visual working memory capacity in mobile augmented reality (MAR) interfaces. With the popularity of augmented reality technology in mobile applications, 3D virtual elements play a key role in interaction. However, prior studies ignored the correlation between virtual object presentation and visual working memory (VWM). Given that visual working memory capacity is closely related to overall cognitive ability, the current study explored the relationship between the complexity of 3D virtual objects and VWM capacity in AR interfaces. Sixty volunteers participated in a 5-point Likert scale rating to eliminate the interference factors of familiarity, concreteness and similarity in 3D virtual material objects. Then, we further conducted an MAR change detection paradigm experiment and successfully manipulated the complexity of 3D virtual objects and the set size. Thirty-one subjects completed the formal experiment. Three measurements (reaction time, proportion correct, and Cowan’s K) were analyzed for nine experimental conditions (three object complexity levels and three object set sizes). The results confirmed that the visual working memory capacity in mobile AR interfaces is modulated by the complexity of the 3D virtual objects and set size, which decreases with the increase in complexity and set size of the displayed 3D virtual objects. As the complexity of the 3D virtual objects increases, the amount of resources allocated to each item decreases, resulting in a decrease in memory capacity and memory accuracy. This study highlights the effectiveness of VWM capacity in MAR interface design research and provides valuable insights into determining the best way to present 3D virtual objects.

Feasibility of snapshot testing using wearable sensors to detect cardiorespiratory illness (COVID infection in India).

The COVID-19 pandemic has challenged the current paradigm of clinical and community-based disease detection. We present a multimodal wearable sensor system paired with a two-minute, movement-based activity sequence that successfully captures a snapshot of physiological data (including cardiac, respiratory, temperature, and percent oxygen saturation). We conducted a large, multi-site trial of this technology across India from June 2021 to April 2022 amidst the COVID-19 pandemic (Clinical trial registry name: International Validation of Wearable Sensor to Monitor COVID-19 Like Signs and Symptoms; NCT05334680; initial release: 04/15/2022). An Extreme Gradient Boosting algorithm was trained to discriminate between COVID-19 infected individuals (n = 295) and COVID-19 negative healthy controls (n = 172) and achieved an F1-Score of 0.80 (95% CI = [0.79, 0.81]). SHAP values were mapped to visualize feature importance and directionality, yielding engineered features from core temperature, cough, and lung sounds as highly important. The results demonstrated potential for data-driven wearable sensor technology for remote preliminary screening, highlighting a fundamental pivot from continuous to snapshot monitoring of cardiorespiratory illnesses.

Three-Dimensional Outdoor Object Detection in Quadrupedal Robots for Surveillance Navigations

Quadrupedal robots are confronted with the intricate challenge of navigating dynamic environments fraught with diverse and unpredictable scenarios. Effectively identifying and responding to obstacles is paramount for ensuring safe and reliable navigation. This paper introduces a pioneering method for 3D object detection, termed viewpoint feature histograms, which leverages the established paradigm of 2D detection in projection. By translating 2D bounding boxes into 3D object proposals, this approach not only enables the reuse of existing 2D detectors but also significantly increases the performance with less computation required, allowing for real-time detection. Our method is versatile, targeting both bird’s eye view objects (e.g., cars) and frontal view objects (e.g., pedestrians), accommodating various types of 2D object detectors. We showcase the efficacy of our approach through the integration of YOLO3D, utilizing LiDAR point clouds on the KITTI dataset, to achieve real-time efficiency aligned with the demands of autonomous vehicle navigation. Our model selection process, tailored to the specific needs of quadrupedal robots, emphasizes considerations such as model complexity, inference speed, and customization flexibility, achieving an accuracy of up to 99.93%. This research represents a significant advancement in enabling quadrupedal robots to navigate complex and dynamic environments with heightened precision and safety.

Working Memory Capacity for Mid-Air Gestures in Human–Computer Interaction

Mid-air gesture interactions have emerged as one of the predominant modalities used in human-computer interaction (HCI). However, extant research on gestures predominantly concentrates on the design of gestures for singular functionalities, which requires a comprehensive investigation into gesture interaction design from an integrative HCI system perspective. Working memory capacity plays a pivotal role in constraining the number of gestures accommodated within an HCI system. This study adopts the change detection paradigm to conduct an empirical investigation into the working memory capacities associated with mid-air gestures, aiming to elucidate the working memory capacity for such gestures. The experimental design included two independent variables: five distinct categories of mid-air gestures (physical, symbolic, metaphorical, abstract, and mixed) and memory set sizes ranging from two to nine items. The findings revealed that working memory capacities for different types of gestures fluctuate between three and five items, with symbolic gestures presenting the most negligible cognitive load for recall and abstract gestures proving to be the most challenging. The outcomes of this research have significant implications for both the design of individual gestures and the overarching design of gesture interaction systems, serving as a valuable reference for future endeavors in HCI design optimization.

Leveraging AI and Machine Learning to Optimize SAP-Driven Financial Operations: A New Paradigm in Risk Management and Fraud Detection

As the global business landscape continues to evolve, organizations are increasingly relying on advanced technologies to enhance their financial operations. The integration of Artificial Intelligence (AI) and Machine Learning (ML) with SAP-driven financial systems represents a transformative shift in how companies manage risk and detect fraud. This article explores how AI and ML can optimize SAP systems for proactive risk management and real-time fraud detection. It highlights the benefits, challenges, and best practices for implementing these technologies within SAP environments, drawing on industry case studies and expert insights

STMixer: A One-Stage Sparse Action Detector.

Traditional video action detectors typically adopt the two-stage pipeline, where a person detector is first employed to generate actor boxes and then 3D RoIAlign is used to extract actor-specific features for action recognition. This detection paradigm requires multi-stage training and inference, and the feature sampling is only constrained inside the box, failing to effectively leverage richer context information outside. Recently, several query-based action detectors have been proposed to predict action instances in an end-to-end manner. However, they still lack adaptability in feature sampling and decoding, thus suffering from the issues of inferior performance or slower convergence. In this paper, we propose two core designs for a more flexible one-stage sparse action detector. First, we present a query-based adaptive feature sampling module, which endows the detector with the flexibility of mining a group of discriminative features from the entire spatio-temporal domain. Second, we devise a decoupled feature mixing module, which dynamically attends to and mixes video features along the spatial and temporal dimensions respectively for better feature decoding. Based on these designs, we instantiate two detection pipelines, that is, STMixer-K for keyframe action detection and STMixer-T for action tubelet detection. Without bells and whistles, our STMixer detectors obtain the state-of-the-art results on five challenging spatio-temporal action detection benchmarks for keyframe action detection or action tube detection.

Knowledge-aware multimodal pre-training for fake news detection

Amidst the rapid propagation of multimodal fake news across various social media platforms, the identification and filtering of disinformation have emerged as critical areas of academic research. A salient characteristic of fake news lies in its diversity, encompassing text–image inconsistency, content–knowledge inconsistency, and content fabrication. However, existing endeavors are generally tailored to a specific subset of fake news, leading to limited universality. Moreover, these models primarily rely on scarce and exorbitant manually labeled annotations, which is incapable of providing sufficient learning signals to detect a variety of fake news. To address these challenges, we propose a novel knowledge-aware multimodal pre-training paradigm for fake news detection, dubbed KAMP. Our motivation lies in incorporating unsupervised correlations through pre-training tasks as complementary to alleviate the dependency on annotations. KAMP consists of a novel multimodal learning model and various delicate pre-training tasks to simultaneously capture valuable knowledge from single modality, multiple modalities, and background knowledge graphs. Our proposal undergoes comprehensive evaluation across two widely utilized datasets, and experimental results demonstrate the superiority of our proposal.

Concanavalin: A Loaded Chitosan Nanoparticles Synthesis and Characterization to Inhibit the Proliferation of Lung Cancer

Background Non-small-cell lung cancer is the most ubiquitous malignancy, and the present treatments do not offer the most effective outcomes for therapy. Natural drug therapy, which directly targets tumor cells, is now recognized as a safe technique for cancer treatment. The use of tailored nanoparticles also improves the efficacy of cancer therapy. Purpose To synthesize, characterize, and evaluate the natural drug concanavalin A (ConA) coated with chitosan nanoparticles (ConA-CS NPs) for cytotoxicity against lung cancer cell line (NCI-H358). Methods The current study demonstrates the development of ConA-CS NPs as well as their unique characteristics, anticancer processes, and possible applications in treatment. Using UV–Vis, Fourier transform infrared, X-ray diffraction, field emission scanning electron microscopy, and dynamic light scattering studies, ConA-CS NPs were generated and evaluated. The mitochondrial toxicity test assay was used to examine ConA-CS NP-induced cytotoxicity. Dual (acridine orange/ethidium bromide) staining was performed to examine apoptotic modifications, and mitochondrial membrane potential levels in NCI-H358 cells were examined using the appropriate fluorescence staining assays. Results and Conclusion The use of ConA-CS NPs for enhanced tumor treatment, which combines natural drugs to eradicate the tumor and prevent relapses, is supported by in vitro investigations. The ability of ConA-encapsulated CS NPs to impede human lung carcinoma cell growth has not yet been investigated. Therefore, this investigation may provide a new paradigm for cancer detection and therapy, and the results may surely help to enhance the quality of life for cancer patients.

Efficient object detector via dynamic prior and dynamic feature fusion

Abstract Sparse R-CNN is a new paradigm of object detection, which predicts objects in a sparse way. However, there are some limitations in Sparse R-CNN. One is the presence of weak prior information caused by fixed learnable proposal boxes and features across different images, necessitating excessive iterations for the model to refine its predictions; the other is the inadequate exploitation of multi-scale information, leading to the sub-optimal detection performance. Thus, building upon Sparse R-CNN, we propose an efficient detector that incorporates dynamic prior and dynamic feature fusion, called $D^{2}$-Det. In particular, for the dynamic prior part, a prior information generator module dynamically generates proposal features and boxes as the dynamic prior for different images to alleviate the inference-inefficient iterative refinement process of predictions, and we further propose the class scores decoupling method to reduce the computation overhead. Furthermore, for the dynamic feature fusion part, we develop a novel lightweight multi-scale feature fusion module, which dynamically aggregates features from all layers for each proposal box, enabling adaptive feature fusion and improving detection precision by nearly 2 AP. Experiments show that $D^{2}$-Det can achieve 46.6 AP on COCO 2017 with fewer computations for the backbone ResNet50, surpassing most of the state-of-the-art detectors.

Enhancing chiral molecule detection: A weak measurement approach utilizing two-dimensional information

This study addresses the critical need for high purity chiral molecules in biological systems by overcoming the challenges associated with the quantitative detection of chiral molecules and their enantiomeric mixtures. We developed an innovative detection approach that leverages the two-dimensional information gleaned from natural optical rotation (NOR) and Faraday optical rotation (FOR) under magnetic fields in chiral molecules, combined with an ultrahigh-resolution weak measurement sensor. This novel weak measurement system achieves unparalleled accuracy in detecting spin angles, with a precision of 1.86 × 10−5°. Notably, our method introduces no chemical reactions or interference with the substances under test. It offers enhanced discrimination capabilities through the dual-dimensional analysis of both natural and Faraday optical rotation, alongside a simple and compact sensor design. Conclusively, our study introduces a novel, high-precision, and multi-dimensional optical detection paradigm for chiral molecules. By incorporating Faraday rotation in the presence of a magnetic field, we expand the informational dimensionality accessible to the original weak measurement sensor, facilitating the quantitative analysis of chiral molecules and their enantiomers. This breakthrough not only furnishes a novel instrument for the exploration and development of chiral pharmaceuticals but also propels the advancement of weak measurement sensing technology forward.

Semi-implicit-Type Order-Adaptive CAT2 Schemes for Systems of Balance Laws with Relaxed Source Term

AbstractIn this paper, we present two semi-implicit-type second-order compact approximate Taylor (CAT2) numerical schemes and blend them with a local a posteriori multi-dimensional optimal order detection (MOOD) paradigm to solve hyperbolic systems of balance laws with relaxed source terms. The resulting scheme presents the high accuracy when applied to smooth solutions, essentially non-oscillatory behavior for irregular ones, and offers a nearly fail-safe property in terms of ensuring the positivity. The numerical results obtained from a variety of test cases, including smooth and non-smooth well-prepared and unprepared initial conditions, assessing the appropriate behavior of the semi-implicit-type second order CATMOOD schemes. These results have been compared in the accuracy and the efficiency with a second-order semi-implicit Runge-Kutta (RK) method.

An Improved EEG Signal Feature Selection Paradigm for Migraine Detection

An Improved EEG Signal Feature Selection Paradigm for Migraine Detection

Pediatric Tuberculosis Screening through Parent Involvement

Introduction: Tuberculosis (TB) remains a significant public health concern, particularly in developing countries. Pediatric TB poses a unique challenge due to the difficulty in diagnosing the disease in young children. Early detection and treatment of TB are crucial to prevent transmission and improve outcomes. Parent involvement in pediatric TB screening can be vital in increasing case detection and reducing morbidity and mortality. Objectives: This study aims to investigate the effectiveness of pediatric TB screening through parent involvement in a community setting in Indonesia. The study will assess this approach's feasibility, acceptability, and impact on identifying TB cases among children. The findings will contribute to developing effective strategies for pediatric TB screening in resource-limited settings. Methods: The research used a pre-and post-test design with a control group design involving 177 mothers with children aged 0-14 years in the Seberang Health Center Working Area, Padang City. Three interventions were applied: self-screening education, structured training, and self-screening simulation. Results: The intervention group showed notable improvements in knowledge, understanding, confidence, and skills across all stages, with statistically significant results (p < 0.05). In contrast, the control group experienced either minimal improvements or decreases in these areas. These findings suggest that the comprehensive approach of education, training, and simulation-based interventions is highly effective in enhancing parents' capacity for early detection of childhood tuberculosis, offering a promising solution for preventing and managing the disease. Conclusions: The results showed significant improvements in mothers' knowledge, understanding, and skills, with each approach contributing differently. The implications are substantial, providing practical guidance for the implementation of TB self-screening in communities and changing the paradigm of TB detection in children nationally and globally.

Semantic-driven dual consistency learning for weakly supervised video anomaly detection

Video anomaly detection presents a significant challenge in computer vision, with the aim of distinguishing various anomaly events from numerous normal ones. Weakly supervised video anomaly detection has recently emerged as a promising solution, enabling the detection of anomaly snippets with only video-level annotations. However, knowledge about anomaly annotation remains underutilized, resulting in a gap between visual space and semantic understanding of anomalies, thus failing to capture the clear boundary between anomalies and normalities. Therefore, we propose a weakly supervised paradigm of cross-modal detection and consistency learning, leveraging dual consistency to provide discriminative representations for anomalies at both the semantic-to-target and target-to-snippet levels. Specifically, we introduce a cross-modal detection network, which detects the targets in each frame according to given semantic rules, to derive semantic-consistent visual embeddings. To depict the clear boundary between anomalies and normalities, a cross-domain alignment module is proposed to enhance the discriminative representation of abnormal targets by learning the contextual consistency between the target and snippet embeddings. Our architecture integrates the detection of semantic-consistent targets based on variable semantic rules, ensuring transferable deployment across scenarios and enabling comprehensive identification, localization, and recognition of abnormal events through a “when-where-which” pipeline. The evaluation of our approach is conducted on four widely used public benchmarks: ShanghaiTech, UCSD Ped2, CUHK Avenue, and UBnormal through extensive qualitative and quantitative analyzes. The results demonstrate the remarkable performance of our approach in dealing with the VAD task.

Enhancing early detection of pancreatic cancer by integrating AI with advanced imaging techniques

Pancreatic cancer remains one of the most lethal malignancies, with a five-year survival rate of less than 10%, primarily due to late-stage diagnosis and rapid disease progression. Early detection is critical for improving patient outcomes, yet current diagnostic methods lack the sensitivity and specificity needed for effective screening. This review explores the integration of advanced imaging techniques with artificial intelligence (AI) to enhance the early detection of pancreatic cancer. Emphasizing a biological approach, we examine the underlying molecular and cellular mechanisms that contribute to the pathogenesis of pancreatic cancer and how they manifest in imaging data. Key imaging modalities, including high-resolution magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET), are evaluated for their efficacy in visualizing pancreatic abnormalities. AI algorithms, particularly machine learning and deep learning, are discussed in the context of their ability to analyze complex imaging datasets, identify subtle biomarkers, and predict disease onset with high accuracy. We delve into the biological markers that AI algorithms can detect, such as changes in the tumor microenvironment, alterations in tissue architecture, and specific molecular signatures of pancreatic ductal adenocarcinoma (PDAC). Furthermore, the integration of AI with molecular imaging techniques, such as positron emission tomography-magnetic resonance imaging (PET-MRI) and optical coherence tomography (OCT), is explored to provide a multi-faceted approach to early diagnosis. The review also highlights the potential of combining AI-driven imaging with liquid biopsies and genomics to create a comprehensive diagnostic framework. By leveraging the power of AI to interpret complex biological data, we propose a novel paradigm for the early detection of pancreatic cancer, aiming to improve screening protocols, enable timely therapeutic interventions, and ultimately enhance patient survival rates. In supposition, the integration of AI with advanced imaging techniques holds significant promise for revolutionizing the early detection of pancreatic cancer. Continued research and clinical validation are essential to translate these technological advancements into routine clinical practice, offering hope for better prognostic outcomes in patients with this devastating disease.

Deep CNN-based Classification of Brain MRI Images for Alzheimer’s Disease Diagnosis

As the leading cause of dementia worldwide, Alzheimer's disease afflicts millions, with progressively impaired abilities to carry out daily activities or communicate and even recognize faces. Although the cause behind lupus is not fully understood, it probably reflects lifestyle choices and environmental factors as well as genetic propensity. The largest obstacles in the diagnosis of these diseases are their often subtle early manifestations and absence of sensitive detection paradigms. Deep-learning algorithms first came to the forefront of medical imaging just a few years ago and were celebrated as sophisticated diagnostic aids, able to spot subtle signs in scans usually hidden from human eyes. We are benefitting from the use of these state-of-the-art algorithms to improve Alzheimer's detection, with one of the largest MRI datasets available today (more than 86,000 images) being used to train our model. In view of this vast data set, it was appreciably combined one to be accurate-centric diagnostic tool. The performance of our novel deep learning model is strong and provided state-of-the-art validation accuracy (99.63%), surpassing existing models These figures highlight the great promise of our model as a verifiable method for detecting early-stage Alzheimer's disease - a significant concern in controlling and managing disease progression. Our research truly is a major step forward in the field of Alzheimer's disease diagnosis by employing cutting-edge deep learning techniques. Early diagnosis allows for better treatment and lower disease burden that can prevent morbidity, mortality and even change many patient outcomes. This is a considerable improvement toward diagnosing Alzheimer's disease with the help of artificial intelligence and presents an expectation for more exact and timely finding.

Enhancing 3D object detection by using neural network with self-adaptive thresholding

Robust 3D object detection remains a pivotal concern in the domain of autonomous field robotics. Despite notable enhancements in detection accuracy across standard datasets, real-world urban environments, characterized by their unstructured and dynamic nature, frequently precipitate an elevated incidence of false positives, thereby undermining the reliability of existing detection paradigms. In this context, our study introduces an advanced post-processing algorithm that modulates detection thresholds dynamically relative to the distance from the ego object. Traditional perception systems typically utilize a uniform threshold, which often leads to decreased efficacy in detecting distant objects. In contrast, our proposed methodology employs a Neural Network with a self-adaptive thresholding mechanism that significantly attenuates false negatives while concurrently diminishing false positives, particularly in complex urban settings. Empirical results substantiate that our algorithm not only augments the performance of 3D object detection models in diverse urban and adverse weather scenarios but also establishes a new benchmark for adaptive thresholding techniques in field robotics.

Enhancing 3D object detection by using neural network with self-adaptive thresholding

Robust 3D object detection remains a pivotal concern in the domain of autonomous field robotics. Despite notable enhancements in detection accuracy across standard datasets, real-world urban environments, characterized by their unstructured and dynamic nature, frequently precipitate an elevated incidence of false positives, thereby undermining the reliability of existing detection paradigms. In this context, our study introduces an advanced post-processing algorithm that modulates detection thresholds dynamically relative to the distance from the ego object. Traditional perception systems typically utilize a uniform threshold, which often leads to decreased efficacy in detecting distant objects. In contrast, our proposed methodology employs a Neural Network with a self-adaptive thresholding mechanism that significantly attenuates false negatives while concurrently diminishing false positives, particularly in complex urban settings. Empirical results substantiate that our algorithm not only augments the performance of 3D object detection models in diverse urban and adverse weather scenarios but also establishes a new benchmark for adaptive thresholding techniques in field robotics.