检索结果-内蒙古大学图书馆

Deep learning-based software engineering: progress,challenges, and opportunities

Science China(Information Sciences) 2025年第1期68卷 57-144页

作者： Xiangping CHEN Xing HU Yuan HUANG He JIANG Weixing JI Yanjie JIANG Yanyan JIANG Bo LIU Hui LIU Xiaochen LI Xiaoli LIAN Guozhu MENG Xin PENG Hailong SUN Lin SHI Bo WANG Chong WANG Jiayi WANG Tiantian WANG Jifeng XUAN Xin XIA Yibiao YANG Yixin YANG Li ZHANG Yuming ZHOU Lu ZHANG School of Journalism and Communication Sun Yat-sen University School of Software Technology Zhejiang University School of Software Engineering Sun Yat-sen University School of Software Dalian University of Technology School of Computer Science and Technology Beijing Institute of Technology Key Laboratory of High Confidence Software Technologies (Peking University) Ministry of EducationSchool of Computer Science Peking University State Key Laboratory for Novel Software Technology Nanjing University School of Computer Science and Engineering Beihang University Institute of Information Engineering Chinese Academy of Sciences School of Computer Science Fudan University State Key Laboratory of Complex & Critical Software Environment (CCSE) School of Software Beihang University School of Computer and Information Technology Beijing Jiaotong University School of Computer Science and Technology Harbin Institute of Technology School of Computer Science Wuhan University Huawei Technologies

Researchers have recently achieved significant advances in deep learning techniques, which in turn has substantially advanced other research disciplines, such as natural language processing, image processing, speech recognition, and software engineering. Various deep learning techniques have been successfully employed to facilitate software engineering tasks, including code generation, software refactoring, and fault localization. Many studies have also been presented in top conferences and journals, demonstrating the applications of deep learning techniques in resolving various software engineering tasks. However,although several surveys have provided overall pictures of the application of deep learning techniques in software engineering,they focus more on learning techniques, that is, what kind of deep learning techniques are employed and how deep models are trained or fine-tuned for software engineering tasks. We still lack surveys explaining the advances of subareas in software engineering driven by deep learning techniques, as well as challenges and opportunities in each subarea. To this end, in this study, we present the first task-oriented survey on deep learning-based software engineering. It covers twelve major software engineering subareas significantly impacted by deep learning techniques. Such subareas spread out through the whole lifecycle of software development and maintenance, including requirements engineering, software development, testing, maintenance, and developer collaboration. As we believe that deep learning may provide an opportunity to revolutionize the whole discipline of software engineering, providing one survey covering as many subareas as possible in software engineering can help future research push forward the frontier of deep learning-based software engineering more systematically. For each of the selected subareas,we highlight the major advances achieved by applying deep learning techniques with pointers to the available datasets i

关键词： deep learning software engineering software benchmark software artifact representation survey

来源：评论

学校读者我要写书评

暂无评论

CoSAF: Towards a Secure Meta-Computing IIoT Infrastructure through Collaborative Source Address Filtering

引用

IEEE Internet of Things Journal 2025年第10期12卷 13709-13722页

作者： Yang, Shu Zhang, Zequn Cui, Laizhong Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China

The rapid growth of the Industrial Internet of Things (IIoT) requires a secure Meta-Computing environment to support applications like industrial monitoring and remote control. However, this environment faces major security challenges, especially the risk of source address forgery, which can enable DDoS and botnet attacks, disrupting operations and compromising equipment. Current Internet infrastructure forwards packets based only on destination addresses, lacking the capability for source address verification. Although edge-based solutions like firewalls and systems such as SAVA (Source Address Validation Architecture) and SAVI (Source Address Validation Improvement) are deployed, they fall short of comprehensive source address validation (SAV), allowing malicious traffic to propagate through core networks. To enhance security, a collaborative approach based on Meta-Computing principles is needed, allowing routers to verify source addresses cooperatively. Given the impracticality of fully upgrading routers, incremental deployment is essential. We show that optimizing incremental SAV deployment is NP-hard. To address this, we propose COSAF (Collaborative Optimized Source Address Filtering), a heuristic algorithm that uses a sink-tree structure to effectively filter attack flows and optimize resource allocation. COSAF also takes SAV table capacity into account to improve resource utilization. Extensive simulations demonstrate that COSAF outperforms traditional methods. © 2025 IEEE.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

LoRaDIP: Low-Rank Adaptation With Deep Image Prior for Generative Low-Light Image Enhancement

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第4期6卷 909-920页

作者： Zhao, Zunjin Shi, Daming Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China

This article presents LoRaDIP, a novel low-light image enhancement (LLIE) model based on deep image priors (DIPs). While DIP-based enhancement models are known for their zero-shot learning, their expensive computational cost remains a challenge. In addressing this issue, our proposed LoRaDIP introduces a low-rank adaptation technique, significantly reducing computational expenses without compromising performance. The contributions of this work are threefold. First, we eliminate the need for estimating initial illumination and reflectance, opting instead to directly estimate the illumination map from the observed image in a generative fashion. The illumination is parameterized by a DIP network. Second, considering the overparameterization of DIP networks, we introduce a low-rank adaptation technique to decrease the number of trainable parameters, thereby reducing computational demands. Third, differing from the existing DIP-based models that rely on a preset fixed number of iterations to halt the optimization process of Retinex decomposition, we propose an automatic stopping criterion based on stable rank, preventing unnecessary iterations. LoRaDIP not only inherits the advantage of requiring only the single input image but also exhibits reduced computational costs while maintaining or even surpassing the performance of state-of-the-art models. © 2020 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Object Detection Model for Remote Sensing Images Based on YOLOv9

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2025年第3期52卷 840-847页

作者： Hou, Donghao Zhang, Yujun School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China

In the field of object detection for remote sensing images, especially in applications such as environmental monitoring and urban planning, significant progress has been made. This paper addresses the common challenges faced by traditional object detection methods in remote sensing images, such as the large number of targets and complex backgrounds, by proposing a novel network based on YOLOv9. The network innovatively introduces the C3_CD_CGA module, an enhanced module based on Cascaded Group Attention, designed to reduce computational redundancy and increase attention diversity, and enhances the processing capability of multi-scale information through the CD module. The C3 module employs deep asymmetric convolution to mitigate information loss and increase the receptive field. Additionally, the network integrates DSConv with the RepNCSPELAN4 module to adaptively focus on and precisely capture the features of elongated and curved local structures, such as vehicles. The introduction of the CARAFE module further improves the spatial resolution of the feature maps, significantly enhancing performance across various visual tasks. Experimental results show that the improved YOLOv9 achieves a mean average precision (mAP) of 88% on the SIMD dataset, which is an improvement of 1.6% compared to the baseline YOLOv9 model and 1.5% higher than the state-of-the-art YOLO-SE model. This model not only achieves more effective multi-target recognition in complex backgrounds but also strikes a good balance between accuracy and efficiency. © (2025), (International Association of Engineers). All rights reserved.

关键词： Urban planning

来源：评论

学校读者我要写书评

暂无评论

A Scalable Unsupervised and Back Propagation Free Learning With SACSOM: A Novel Approach to SOM-Based Architectures

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第4期6卷 955-967页

作者： Hirani, Gaurav R. Wang, Kevin I-Kai Abdulla, Waleed H. The University of Auckland Department of Electrical Computer and Software Engineering Auckland1010 New Zealand

The field of computer vision is predominantly driven by supervised models, which, despite their efficacy, are computationally expensive and often intractable for many applications. Recently, research has expedited alternative avenues such as self-organizing maps (SOM)-based architectures, which offer significant advantages such as tractability, the absence of back-propagation, and feed-forward unsupervised learning. However, these SOM-based approaches frequently suffer from lower accuracy and limited generalization capabilities. To address these shortcomings, we propose a novel model called split and concur SOM (SACSOM). SACSOM overcomes the limitations of closely related SOM-based algorithms by utilizing multiple parallel branches, each equipped with its own SOM modules that process data independently with varying patch sizes. Furthermore, by creating groups of classes and using respective training samples to train independent subbranches in each branch, our approach accommodates datasets with a large number of classes. SACSOM employs a simple yet effective labeling technique requiring minimal labeled samples. The outputs from each branch, filtered by a threshold, contribute to the final prediction. Experimental validation on MNIST-digit, Fashion-MNIST, CIFAR-10, and CIFAR-100 demonstrates that SACSOM achieves competitive accuracy with significantly reduced computation time. Furthermore, it exhibits superior performance and generalization capabilities, even in high-noise scenarios. The weights of the single-layered SACSOM provide meaningful insights into the patch-based learning pattern, enhancing its tractability and making it ideal from the perspective of explainable AI. This study addresses the limitations of current clustering techniques, such as K-means and traditional SOMs, by proposing a lightweight, manageable, and fast architecture that does not require a GPU, making it suitable for low-powered devices. © 2024 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Multi-label, Classification-based Prediction of Breast Cancer Metastasis Directions

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2025年第1期52卷 1-10页

作者： Wang, Tingting Fan, Qi Tan, Liang Zhang, Beier School of Computer and Software Engineering Anhui Institute of Information Technology China School of Computer Science and Technology Huaibei Normal University China School of Computer and Software Engineering Anhui Institute of Information Technology China School of Computer Science and Technology Huaibei Normal University China

Predicting the metastatic direction of primary breast cancer (BC), thus assisting physicians in precise treatment, strict follow-up, and effectively improving the prognosis. The clinical data of 293,946 patients with primary BC diagnosed between 2010 and 2015 were collected from the Surveillance, Epidemiology, and End Results database. Multiple interpolations and Multi-label Synthetic Minority Over-sampling Technique methods were used for data analysis, and machine learning model was established for multi-label classification. Finally, Surgical information, lymph node status, distant metastasis, tumor size, chemotherapy, histological type, and radiotherapy had significant influence as inputs. Compared with the k-nearest neighbor model, average accuracies of the decision tree and random forest (RF) models increased from 88.84% to 93.59% and 94.14%, respectively. Their average precision, recall rate, F1 score, area under the receiver operating characteristic curve and weighted-F1 increased from 87.24% to 95.85% and 94.74%, 87.73% to 90.40% and 91.76%, 87.07% to 92.16% and 93.45%, 97.11% to 99.53% and 99.95%, 82.13% to 89.44% and 90.48%, respectively. In conclusion, the RF model, which showed the best performance, can be used in multi-label prediction of BC metastasis directions, and can assist physicians in diagnosing and treating patients with primary BC. © (2025), (International Association of Engineers). All rights reserved.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Enhancing the adversarial robustness in medical image classification: exploring adversarial machine learning with vision transformers-based models

引用

Neural Computing and Applications 2025年第12期37卷 7971-7989页

作者： Kanca Gulsoy, Elif Ayas, Selen Baykal Kablan, Elif Ekinci, Murat Software Engineering Karadeniz Technical University Trabzon Turkey Computer Engineering Karadeniz Technical University Trabzon Turkey

Despite their remarkable achievements in computer-aided medical image-related tasks, including detection, segmentation, and classification, deep learning techniques remain vulnerable to imperceptible adversarial attacks, which could lead to potential misdiagnosis in clinical applications. Therefore, adversarial attacks and their defense in deep medical diagnosis systems have been witnessed remarkable progress in recent years. Although the importance of transformers in various medical applications has grown immensely, a critical concern for the reliability and security of their susceptibility to adversarial attacks has not yet been sufficiently investigated. Furthermore, many studies in the field of ViT-based adversarial machine learning mainly focus on pure ViT architecture. To this end, this paper provides a comprehensive evaluation, comparison, and analysis of state-of-the-art ViT-based models such as ViT, DeiT, Swin transformer, and PVTv2 for their robustness against FGSM and PGD adversarial machine learning attacks and investigates the impact of the k-step PGD adversarial training defense mechanism in the domain of various medical imaging tasks. The findings indicate that ViT-based models are vulnerable to adversarial attacks even at a small perturbation degree. Furthermore, the significant drop in accuracy from around 90.0% underlines the vulnerability of these models to adversarial attacks and highlights the urgent need for robust defenses. We also conclude that the ViT-based models show significant robustness against adversarial attacks using adversarial training, i.e., the defense strategy can achieve improved classification accuracy, close to the clean image accuracy. By analyzing quantitative results, we believe that this study aims to fill the gap in research on the robustness of ViT-based models to adversarial machine learning attacks in medical image analysis, highlighting future research directions. © The Author(s), under exclusive licence to Springer-

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced facial action unit detection with adaptable patch sizes on representative landmarks

引用

Neural Computing and Applications 2025年第5期37卷 3777-3791页

作者： Cakir, Duygu Yilmaz, Gorkem Arica, Nafiz Department of Software Engineering Bahcesehir University Istanbul Turkey Department of Computer Engineering Bahcesehir University Istanbul Turkey Department of Computer Engineering Piri Reis University Istanbul Turkey

The human face displays expressions through the contraction of various facial muscles. The Facial Action Coding System (FACS) is a widely accepted taxonomy that describes all visible changes in the face in terms of action units (AUs). In this study, AUs are examined by finding the most active landmarks of the face and then examining the most representative patch sizes of each landmark for the AU detection task. Sparse learning is employed to learn the most active landmarks for each AU, and then the active landmark patches are fed to ViT and Perceiver mechanisms independently. Experiments indicate that using active landmark patches with their most representative size improves the results when compared to using all the landmarks, especially when it is used on more challenging datasets as a support for the attention mechanism of the classifier. The results demonstrate that the proposed method improves the performance of the employed models and are further supported by experiments conducted across different datasets. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

A Dangerous Driving Behavior Detection Method Based on Improved YOLOv8s

引用

engineering Letters 2025年第3期33卷 721-731页

作者： Zhou, Tong Zhang, Xiaoxia Chen, Huilong School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

Detecting dangerous driving behavior is a critical research area focused on identifying and preventing actions that could lead to traffic accidents, such as smoking, drinking, yawning, and drowsiness, through technical methods. Advanced computer vision and machine learning technologies enable efficient detection models to monitor and analyze driver behavior, improving road safety. Due to challenges posed by complex environments, this paper introduces an enhanced detection algorithm, YOLOv8s-CDS, to improve the identification of dangerous driving behaviors. First, the ConvNeXt V2 module is integrated with the C2f module to form C2fNeb2, optimizing feature extraction for behaviors like smoking or phone use. Second, the DASI (Dimension-Aware Selective Integration) module enhances detection accuracy through multi-scale fusion and dimension perception. Additionally, the SCConv module replaces the Conv module in the Bottleneck, forming C2fSCConv, which reduces spatial redundancy and improves detection efficiency.A comprehensive experimental analysis of dangerous driving image datasets demonstrates that the mean average precision (mAP) of the YOLOv8s-CDS algorithm is 91.20%, which is 2.4% higher than that of the YOLOv8s algorithm. Compared to other object detection algorithms, such as Faster R-CNN, YOLOv5s, YOLOv7s, YOLOX and YOLOv8, YOLOv8s-CDS demonstrates greater practicality in detecting dangerous driving behaviors, contributing to a reduction in traffic accidents and enhancing the safety of life and property. © 2025, International Association of Engineers. All rights reserved.

关键词： Highway accidents

来源：评论

学校读者我要写书评

暂无评论

The Improved Unet Semantic Segmentation Network for Remote Sensing Images

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2025年第4期52卷 1187-1195页

作者： Zhu, Hang Zhao, Ji School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

With the development of artificial intelligence, deep learning has been increasingly used to achieve automatic detection of geographic information, replacing manual interpretation and improving efficiency. However, remote sensing images themselves have the issue of slight inter-class variance and significant intra-class variance, making it challenging to extract valuable information. Additionally, the increasing resolution and size of remote sensing images in recent years have introduced more complexity in the types of information, further increasing the difficulty of extracting valuable data. This paper proposes an improved Unet semantic segmentation network (referred to as RAUnet). First, in the encoder, continuous convolutional blocks are enhanced to extract features. At the same time, the EMAM multi-scale attention module is employed for cross-channel learning, capturing information from different feature channels of the target and using the surrounding feature information to assist in distinguishing target information. To capture multi-directional long-range dependencies, the Lo2 module is used for long-range modeling, which captures not only local contextual information but also long-range dependencies. In the decoder, a Dysample upsampling module is used to restore feature details, and in the skip connection layer, features are added for feature fusion. Experimental results show that compared to mainstream models, the proposed method achieves superior segmentation results on the Potsdam and Vihingen datasets. © (2025), (International Association of Engineers). All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：