Critical Task Research Articles

Enhanced Abandoned Object Detection through Adaptive Dual-Background Modeling and SAO-YOLO Integration

Abandoned object detection is a critical task in the field of public safety. However, existing methods perform poorly when detecting small and occluded objects, leading to high false detection and missed detection rates. To address this issue, this paper proposes an abandoned object detection method that integrates an adaptive dual-background model with SAO-YOLO (Small Abandoned Object YOLO). The goal is to reduce false and missed detection rates for small and occluded objects, thereby improving overall detection accuracy. First, the paper introduces an adaptive dual-background model that adjusts according to scene changes, reducing noise interference in the background model. When combined with an improved PFSM (Pixel-based Finite State Machine) model, this enhances detection accuracy and robustness. Next, a network model called SAO-YOLO is designed. Key improvements within this model include the SAO-FPN (Small Abandoned Object FPN) feature extraction network, which fully extracts features of small objects, and a lightweight decoupled head, SODHead (Small Object Detection Head), which precisely extracts local features and enhances detection accuracy through multi-scale feature fusion. Finally, experimental results show that SAO-YOLO increases mAP@0.5 and mAP@0.5:0.95 by 9.0% and 5.1%, respectively, over the baseline model. It outperforms other advanced detection models. Ultimately, after a series of experiments on the ABODA, PETS2006, and AVSS2007 datasets, the proposed method achieved an average detection precious of 91.1%, surpassing other advanced methods. It significantly outperforms other advanced detection methods. This approach notably reduces false and missed detections, especially for small and occluded objects.

CJE-PCHF: Chinese Joint Entity and Relation Extraction Model Based on Progressive Contrastive Learning and Heterogeneous Feature Fusion

The joint extraction of entities and relations is a critical task in information extraction, and its performance directly affects the performance of downstream tasks. However, existing joint extraction models based on deep learning exhibit weak processing capabilities for the phenomenon of multiple pronunciations of one character and multiple characters of one pronunciation when processing Chinese texts, resulting in a performance loss. To address these issues, this paper introduces part-of-speech (POS) and pinyin features to aid the model in learning semantic features that are more contextually appropriate. We propose a Chinese Joint Entity and Relation Extraction Model based on progressive contrastive learning and heterogeneous feature fusion (CJE-PCHF). During model training, an interactive fusion network based on progressive contrastive learning is employed to learn the dependencies between pinyin, POS, and semantic features. This guides the model in heterogeneous feature fusion, capturing higher-order semantic associations between heterogeneous features. On the commonly used DuIE evaluation dataset for joint extraction, our model achieved a significant improvement, with the F1 score increasing by 5.4% compared to the benchmark model CasRel.

Advanced artificial intelligence framework for T classification of TNM lung cancer in 18FDG-PET/CT imaging

The integration of artificial intelligence (AI) into lung cancer management offers immense potential to revolutionize diagnostic and treatment strategies. The aim is to develop a resilient AI framework capable of two critical tasks: firstly, achieving accurate and automated segmentation of lung tumors and secondly, facilitating the T classification of lung cancer according to the ninth edition of TNM staging 2024 based on PET/CT imaging. This study presents a robust AI framework for the automated segmentation of lung tumors and T classification of lung cancer using PET/CT imaging. The database includes axial DICOM CT and18FDG-PET/CT images. A modified ResNet-50 model was employed for segmentation, achieving high precision and specificity. Reconstructed 3D models of segmented slices enhance tumor boundary visualization, which is essential for treatment planning. The Pulmonary Toolkit facilitated lobe segmentation, providing critical diagnostic insights. Additionally, the segmented images were used as input for the T classification using a CNN ResNet-50 model. Our classification model demonstrated excellent performance, particularly for T1a, T2a, T2b, T3 and T4 tumors, with high precision, F1 scores, and specificity. The T stage is particularly relevant in lung cancer as it determines treatment approaches (surgery, chemotherapy and radiation therapy or supportive care) and prognosis assessment. In fact, for Tis-T2, each increase of one centimeter in tumor size results in a worse prognosis. For locally advanced tumors (T3-T4) and regardless of size, the prognosis is poorer. This AI framework marks a significant advancement in the automation of lung cancer diagnosis and staging, promising improved patient outcomes.

The Critical Role of Product Managers and Their Responsibilities in Software Startups: A Systematic Literature Review

In the high-paced, uncertain, and resource-constrained environment of software startups, achieving product–market fit and acquiring initial customers are critical yet challenging tasks. These challenges are often exacerbated by the absence of strategic planning, inefficient requirements selection processes, inadequate resource allocation, and insufficient market research, leading to an alarming 63% failure rate among software startups. The Product Manager (PM), responsible for product strategy, planning, and results monitoring, plays a pivotal role in navigating these complexities; however, the PM role is often undervalued and lacks a universally accepted definition. This study aims to enhance the understanding of the PM's role within software startups by conducting a comprehensive Systematic Literature Review (SLR). Employing a rigorous selection methodology, we reviewed 134 studies and identified 662 distinct PM tasks across 122 activities. Notably, only 7 studies (5.73%) specifically addressed the software startup context, revealing a substantial gap in the literature. Our findings indicate that while product planning and requirements prioritization are dominant topics, many critical activities remain underrepresented, underscoring the need for targeted research in these areas. By addressing the unique challenges faced by software startups, we pave the way for more targeted and effective PM strategies. Future research should consider adopting mixed-method approaches to deepen the understanding of PM practices and investigate underexplored areas such as go-to-market strategies, roadmap creation, and requirements selection within the startup context.

A semantic segmentation method integrated convolutional nonlinear spiking neural model with Transformer

Semantic segmentation is a critical task in computer vision, with significant applications in areas like autonomous driving and medical imaging. Transformer-based methods have gained considerable attention recently because of their strength in capturing global information. However, these methods often sacrifice detailed information due to the lack of mechanisms for local interactions. Similarly, convolutional neural network (CNN) methods struggle to capture global context due to the inherent limitations of convolutional kernels. To overcome these challenges, this paper introduces a novel Transformer-based semantic segmentation method called NSNPFormer, which leverages the nonlinear spiking neural P (NSNP) system—a computational model inspired by the spiking mechanisms of biological neurons. The NSNPFormer employs an encoding–decoding structure with two convolutional NSNP components and a residual connection channel. The convolutional NSNP components facilitate nonlinear local feature extraction and block-level feature fusion. Meanwhile, the residual connection channel helps prevent the loss of feature information during the decoding process. Evaluations on the ADE20K and Pascal Context datasets show that NSNPFormer achieves mIoU scores of 53.7 and 58.06, respectively, highlighting its effectiveness in semantic segmentation tasks.

Intelligent Method of Identifying the Nonlinear Dynamic Model for Helicopter Turboshaft Engines

This research focused on the helicopter turboshaft engine dynamic model, identifying task solving in unsteady and transient modes (engine starting and acceleration) based on sensor data. It is known that about 85% of helicopter turboshaft engines operate in steady-state modes, while only around 15% operate in unsteady and transient modes. Therefore, developing dynamic multi-mode models that account for engine behavior during these modes is a critical scientific and practical task. The dynamic model for starting and acceleration modes has been further developed using on-board parameters recorded by sensors (gas-generator rotor r.p.m., free turbine rotor speed, gas temperature in front of the compressor turbine, fuel consumption) to achieve a 99.88% accuracy in identifying the dynamics of these parameters. An improved Elman recurrent neural network with dynamic stack memory was introduced, enhancing the robustness and increasing the performance by 2.7 times compared to traditional Elman networks. A theorem was proposed and proven, demonstrating that the total execution time for N Push and Pop operations in the dynamic stack memory does not exceed a certain value O(N). The training algorithm for the Elman network was improved using time delay considerations and Butterworth filter preprocessing, reducing the loss function from 2.5 to 0.12% over 120 epochs. The gradient diagram showed a decrease over time, indicating the model’s approach to the minimum loss function, with optimal settings ensuring the stable training.

RPEPL: Tibetan Sentiment Analysis Based on Relative Position Encoding and Prompt Learning

Sentiment analysis is a critical task for natural language processing. Much research has been done for high-resource languages such as English and Chinese. However, Tibetan is an extremely low-resource language with less reference information. According to the practical demands, this paper proposes a Tibetan sentiment analysis method based on relative position encoding and prompt learning, the method is abbreviated as RPEPL. First, Word information is introduced to syllable sequences by converting the directed acyclic lattice into a squashed structure. Second, a relative position encoding is used to encode the position information of syllables and words. Then the association relations and semantic information of tokens are identified by leveraging the multi-attention. Finally, the sentiment category of Tibetan sentence is obtained through a prompt learning framework. Experimental results demonstrate that RPEPL significantly outperforms the baseline methods on the TUSA dataset and TNEC dataset. Additionally, Traditional recurrent neural networks cannot perform large-scale parallel computation and convolutional neural networks have difficulty modeling long-distance dependencies in Tibetan text both of which are resolved using RPEPL. Furthermore, the use of a multi-attention not only enriches the association relations between syllables and words, but also enhances the understanding of sentence semantic and syntactic structure information, and improves the performance of Tibetan sentiment analysis.

VarLifter: Recovering Variables and Types from Bytecode of Solidity Smart Contracts

Since funds or tokens in smart contracts are maintained through specific state variables, contract audit, an effective means for security assurance, particularly focuses on these variables and their related operations. However, the absence of publicly accessible source code for numerous contracts, with only bytecode exposed, hinders audit efforts. Recovering variables and their types from Solidity bytecode is thus a critical task in smart contract analysis and audit, yet this is a challenging task because the bytecode loses variable and type information, only with low-level data operated by stack manipulations and untyped memory/storage accesses. The state-of-the-art smart contract decompilers miss identifying many variables and incorrectly infer the types for many identified variables. To this end, we propose VarLifter, a lifter dedicated to the precise and efficient recovery of typed variables. VarLifter interprets every read or written field of a data region as at least one potential variable, and after discarding falsely identified variables, it progressively refines the variable types based on the variable behaviors in the form of operation sequences. We evaluate VarLifter on 34,832 real-world Solidity smart contracts. VarLifter attains a precision of 97.48% and a recall of 91.84% for typed variable recovery. Moreover, VarLifter finishes analyzing 77% of smart contracts in around 10 seconds per contract. If VarLifter is used to replace the variable recovery modules of the two state-of-the-art Solidity bytecode decompilers, 52.4%, and 74.6% more typed variables will be correctly recovered, respectively. The applications of VarLifter to contract decompilation, contract audit, and contract bytecode fuzzing illustrate that the recovered variable information improves many contract analysis tasks.

Pavement Distress Detection Based on the Staged Deep Learning Classifier Approach

Pavement crack detection is a critical task that is essential to ensure the safety of roads and highways. Accurate detection and classification of pavement cracks can help transportation agencies to identify potential maintenance needs and plan appropriate interventions. In this project, we propose a staged deep learning classifier approach for pavement crack detection to provide accurate and comprehensive information about the pavement condition. In this proposed pavement crack detection model, there are four main phases: pre-processing, feature extraction, feature selection, and crack detection. Firstly, raw images are pre-processed through median filtering and histogram equalization for noise removal and contrast enhancement. Next, features such as the improved local binary pattern, gray-level co-occurrence matrix, and gray-level run length matrix are extracted from the pre-processed images. The optimal features are selected using a hybrid optimization model, the tuna customized honey badger optimization algorithm, which combines the honey badger algorithm and tuna swarm optimization. This new hybrid optimization model will enhance the search strategy for solving optimization problems. The loss function of an augmented convolutional neural network (A-CNN) is modified to compute the root mean square error instead of the entropy-based loss function. The final detected outcomes (presence or absence of cracks) are acquired from the A-CNN.

A Hybrid Approach for Robust Object Detection: Integrating Template Matching and Faster R-CNN

Object detection is a critical task in computer vision, with applications ranging from autonomous vehicles to medical imaging. Traditional methods like template matching offer precise localization but struggle with variations in object appearance, while deep learning approaches such as Faster R-CNN excel in handling diverse and complex datasets but often require extensive computational resources and large amounts of labeled data. This paper proposes a hybrid approach that integrates template matching with Faster R-CNN to leverage the strengths of both techniques. By combining the accuracy of template matching with the robustness and generalization capabilities of Faster R-CNN, our method achieves superior performance in challenging scenarios, including objects with occlusions, varying scales, and complex backgrounds. Extensive experiments demonstrate that the hybrid model not only enhances detection accuracy but also reduces computational load, making it a practical solution for real-world applications.

Flight Arrival Scheduling via Large Language Model

The flight arrival scheduling problem is one of the critical tasks in air traffic operations, aiming to ensure that the flight arrive in the correct sequence safely. Existing methods primarily focus on the terminal area and often overlook the presence of training flight at the airport. Due to the limited generalization of traditional methods and varying control practices at different airports, training flight at airports still rely on manual control for arrival sorting. To effectively address these issues, we propose a novel method for slot allocation that leverages the strong reasoning capabilities and generalization potential of large language models (LLMs). Our method conceptualizes the dynamic scheduling problem for training flight as a language modeling problem, a perspective not previously explored. Specifically, we represent the allocator’s inputs and outputs as language tokens, utilizing LLMs to generate conflict-free results based on a language description of requested landing information and assigned training flight information. Additionally, we employ a reset strategy to create a small dataset for scenario-specific samples, enabling LLMs to quickly learn allocation schemes from the dataset. We demonstrated the capability of LLMs in addressing time conflicts by evaluating metrics such as answer accuracy, conflict rate, and total delay time (without the wrong answer). These findings underscore the feasibility of employing LLMs in the field of air traffic control.

A comprehensive framework for estimating the remaining useful life of Li-ion batteries under limited data conditions with no temporal identifier

The escalating applications of Lithium-ion (Li-ion) batteries in renewable energy and electric vehicles underscore the need for enhanced prognostics and health management systems to reduce the risk of sudden failures. Remaining useful life (RUL) determination is one of the most critical tasks in the field of battery prognostics nowadays. Even though statistical and machine learning (ML) methods have proven effective in research setups, many challenges prevent applying these prediction methods to real-life scenarios. These challenges include (1) scarcity of run-to-failure datasets with similar experimental conditions, (2) low data granularity when presented in capacity vs. discharge cycle pairs, and (3) lack of “temporal identifiers” in real-life scenarios. A temporal identifier is any label that provides knowledge about the current degradation state of a working battery. The research question developed for this study was, ‘Can the remaining useful life of a Li-ion battery having limited data without a temporal identifier be predicted?’ The specific aims were to estimate the temporal identifier of limited data and to predict the remaining useful life (RUL). An innovative framework incorporating reliability analysis and deep learning addresses these specific aims. Experimental data is used to test the framework's capabilities, limiting the training dataset to only three batteries and the testing dataset to a small sample (< 10 data points) of another battery. This new approach enabled the RUL prediction to achieve errors as low as ∼5 cycles and root mean square error of 6.24 cycles, outperforming other benchmark studies on Li-ion battery RUL prediction that use more battery degradation data without temporal identifier.

Impact of traffic simulation parameters on the estimation of noise exposure in an urban environment

Since road traffic is the most impactful noise source in the European cities, the evaluation of citizens' exposure is crucial. Although the noise map accuracy is affected by uncertainties in traffic volume, to perform extensive traffic monitoring is not practical and expensive, even in small cities. Traffic simulation software uses routing algorithms to suggest the fastest path to distribute vehicles from an origin through a network to a desired destination which interacts within the urban environment. Therefore, it could represent an important tool for noise mapping in the critical task to fill the data gaps in traffic volumes, also for implementing action plans using a suitable traffic management approach. However, there is a lack of evidence on how the choice of traffic simulation parameters influences noise estimation. In this work, we design a simulation framework to evaluate the impact of routing algorithms on the estimation of population noise exposure in an urban area. Using an open-source pipeline based on public databases and open-source software SUMO and NoiseModelling, we evaluated the implication on the traffic distribution and resulting noise exposure for a set of traffic model key parameters.

Heart Disease Prediction Using Machine Learning Algorithms

The cases of heart disease increase rapidly day by day and it is essential to predict that type of disease. The prediction is a critical task it must be performed accurately and precisely. This research paper focuses on which patient suffers from cardiovascular disease based on different medical attributes like age, cholesterol level, chest pain types, fasting blood sugar, etc. Taking the medical history of the patient we have created a cardiovascular disease prediction system to predict if the patient is suffering from disease or not. This prediction system is designed by using some of the machine learning algorithms or classifiers such as Decision Tree, Random Forest, Support Vector Machine, Naïve Bayes, K-Nearest Neighbor, Logistic Regression, and Gradient Boosting. Before running the model, the data is preprocessed to get the better accuracy. To train the model the dataset is divided into 20% testing and 80% training and achieves accuracy as follows: DT: 83.90%, RF: 88.78%, SVM: 83.90%, NB: 82.43%, K-NN: 95.60%, LR: 83.90% and GB: 86.82%. The prediction accuracy of the K-Nearest Neighbor is higher than the other algorithms. The given prediction system reduces the cost and enhances medical care. This prediction system helps us to predict patients with heart disease and is implemented in Colab .pynb format.

BRalu Profile: the procurement strategy

Learning outcomes After completion of the case study, students will be able to understand BRalu Profile’s product range, customer base and historical developments; analyze how BRalu Profile’s procurement strategy evolved and its impact on business profitability; calculate and compare procurement costs and evaluate their role in decision-making for different suppliers; examine how market conditions (includes domestic and international dynamics) and pricing strategies influence procurement choices; assess the pros and cons of different procurement options and make informed recommendations based on supply chain principles; and identify potential procurement risks (e.g. currency exchange rates and supplier reliability) and propose strategies to mitigate them. Case overview/synopsis This case study explored the challenges faced by BRalu Profile, a prominent aluminum profile products company based in Ahmedabad, India. It focused on the critical task of supplier evaluation, selection and the complexities of maintaining relationships with existing suppliers. This case study delved into the intricate dynamics of procurement decisions within the supply chain and their direct impact on the firm’s overall performance. It also emphasized the supply chain’s susceptibility to disruptions and their consequences on company operations and supplier selection criteria. The protagonist, Dhaval Choladiya, had to navigate the complexities of cost-benefit analysis to identify the most suitable supplier, maximizing the firm’s net benefit while considering nonpricing parameters. This case study revealed the critical importance of maintaining strong supplier relationships in a competitive market and offered insights into the complexities of sourcing. Complexity academic level This case study is suitable for an undergraduate or graduate-level course on strategic sourcing or supply chain management or a risk management module in operations, strategy or finance course (e.g. how to deal with input cost fluctuations). Supplementary materials Teaching notes are available for educators only. Subject code CSS 11: Strategy.

Nested transformer decoder using dense skip-connections for change detection in high-resolution remote sensing image

ABSTRACT Change detection is a critical task in remote sensing image analysis, with applications in land management and urban development. Convolutional neural networks (CNNs) have shown success in this area, but existing CNN-based methods often struggle with irrelevant changes caused by interfering factors, leading to unsatisfactory results. To address this, some studies have replaced CNN layers with transformers to enhance the receptive field and utilize global information better. However, this can make models overly reliant on global information, potentially overlooking shallow, fine-grained features, especially at the edges of changed targets. To overcome these limitations, we propose a Nested Transformer Decoder (NTD) that combines the strengths of transformers and CNNs by introducing dense skip connections. While having a large receptive field, it can also better retain shallow and fine-grained features by adding skip-connection. These innovations enable our model to more accurately identify changes and preserve the edges of the changed areas, including small changes that are difficult to detect using existing methods. Additionally, we introduce a Residual Fusion Module (RFM) to reduce false detections caused by irrelevant factors like light and shadow. The RFM employs residual information and spatiotemporal attention to suppress differences in multi-scale features, leading to better detection outcomes. The experimental results demonstrate that our proposed model outperforms the current state-of-the-art change detection methods in various metrics, and represents a novel and effective solution to the challenge of change detection in remote sensing images. Experimental results show that our model surpasses state-of-the-art change detection methods across various metrics. Tested on three public change detection datasets, our method improves the Intersection over Union (IoU) for the change class by 2.83% to 16.39% and enhances the F1-score by 1.60% to 9.63%. These results demonstrate that the proposed approach provides an effective solution to the challenges of change detection in remote sensing images.

An AI-Based Approach for Developing a Recommendation System for Underground Mining Methods Pre-Selection

Selecting the most appropriate mining method to recover mineral resources is a critical decision-making task in mining project development. This study introduces an artificial intelligence-based mining methods recommendation system (AI-MMRS) for the pre-selection of underground mining methods. The study integrates and evaluates the capability of two approaches for mining methods selection (MMS): the memory-based collaborative filtering (CF) approach aided by the UBC-MMS system to predict the top-3 relevant mining methods and supervised machine learning (ML) classification algorithms to enhance the effectiveness and novelty of the AI-MMRS, addressing the limitations of the CF approach. The results reveal that the memory-based CF approach achieves an accuracy ranging from 81.8% to 87.9%. Among the classification algorithms, artificial neural network (ANN) and k-nearest neighbors (KNN) classifiers perform the best, with accuracy levels of 66.7% and 63.6%, respectively. These findings demonstrate the effectiveness and viability of both approaches in MMS, acknowledging their limitations and the need for continuous training and optimization. The proposed AI-MMRS for the pre-selection stage supplemented by the direct involvement of mining professionals in later stages of MMS, has the potential to significantly aid in the MMS decision-making, providing data-driven and experience-based recommendations following the ongoing evolution of mining practices.

Deep indoor illumination estimation based on spherical gaussian representation with scene prior knowledge

High dynamic range illumination estimation from a single low dynamic range image is a critical task in the fields of computer vision, graphics and augmented reality. However, directly learning the full HDR environment map or parametric lighting information from a single image is extremely difficult and inaccurate. As a result, we propose a two-stage network approach for illumination estimation that integrates spherical gaussian (SG) representation with scene prior knowledge. In the first stage, a convolutional neural network is utilized to generate material and geometric information about the scene, which serves as prior knowledge for lighting prediction. In the second stage, we model indoor environment illumination using 128 SG functions with fixed center direction and bandwidth, allowing only the amplitude to vary. Subsequently, a Transformer-based lighting parameter regressor is employed to capture the complex relationship between the input images with scene prior information and its SG illumination. Additionally, we introduce a hybrid loss function, which combines a masked loss for high-frequency illumination with a rendering loss for improving the visual quality. By training and evaluating the lighting model on the created SG illumination dataset, the proposed method achieves competitive results in both quantitative metrics and visual quality, outperforming state-of-the-art methods.

Gas Classification and Concentration Prediction in Open Environments Using Class Anchor Clustering-Initialized Temporal Convolutional Network

Identifying target gases and predicting their concentrations in open environments are critical tasks, particularly due to fluctuating environmental conditions and the presence of unknown gases. This study aims to address these challenges by developing a model that improves both gas classification and concentration prediction. We introduce two key components: Quantile-Dynamic Associated Features (Q-DAF) and the Class Anchor Clustering-Initialized Temporal Convolutional Network (CAC-InitTCN). Q-DAF dynamically enhances gas signal feature extraction, optimizing the ability of CAC-InitTCN to handle complex and changing gas environments. CAC-InitTCN is designed to efficiently process time-series data, allowing it to manage gas concentration fluctuations while mitigating the interference of unknown gases. CAC-InitTCN was evaluated on two distinct datasets, achieving classification accuracy rates of 93.62% and 98.05%, respectively. In gas concentration prediction tasks, the model demonstrated RMSE values of 0.2102 and 0.0392, and R2 scores of 0.9085 and 0.9784, after data normalization. These results demonstrate the capability of CAC-InitTCN in handling real-time gas fluctuations and recognizing unknown gases, making it a promising approach for open-environment gas detection.

Causal-relationship representation enhanced joint extraction model for elements and relationships

The joint extraction of elements such as entities, relations, events, and their arguments, along with their specific interrelationships, is a critical task in natural language processing. Existing research often employs implicit handling of interactions between tasks through techniques like shared encodings or parameter sharing, lacking explicit modeling of the specific relationships between tasks. This limitation hinders the full utilization of inter-task correlation information and impacts effective collaboration between tasks. To address this, we propose a model for joint extraction of elements and relations based on causal relationship representation enhancement (CRE). This model aims to capture specific relationships between tasks in multiple stages, facilitating finer adjustments and optimization of subtasks and thereby enhancing the overall model performance. Specifically, CRE comprises three key modules: feature adaptation, feature interaction, and feature fusion. The feature adaptation module selects and adjusts features from shared encodings based on the requirements of specific tasks to better adapt to semantic differences between different tasks. The feature interaction module employs causal reasoning to comprehensively capture the causal relationships between tasks while mitigating negative transfer brought about by interference from semantic information. The feature fusion module further integrates features to obtain optimized task-specific representations. Ultimately, CRE exhibits a significant improvement in average performance across multiple information extraction tasks.