检索结果-内蒙古大学图书馆

XAI-DSCSA: explainable-AI-based deep semi-supervised convolutional sparse autoencoder for facial expression recognition

引用

SIGNAL image AND VIDEO processing 2025年第5期19卷 1-18页

作者： Mohana, M. Subashini, P. Ghinea, George Avinashilingam Inst Ctr Machine Learning & Intelligence Dept Comp Sci Coimbatore India Brunel Univ Dept Comp Sci London England

Facial expression recognition (FER) continues to be a vibrant research field, driven by the increasing need for its practical applications in areas such as e-learning, healthcare, candidate interview analysis, and more. Most deep learning approaches in supervised FER systems heavily rely on large, labeled datasets. Implementing FER in Convolutional Neural Networks (CNNs) often requires many layers, leading to extended training times and difficulties in finding optimal parameters. This can result in challenges in creating distinct facial expression patterns for classification, leading to poor real-time emotion classification In this paper, we propose a novel approach known as the deep Semi-supervised Convolutional Sparse Autoencoder to address the aforementioned issues and enhance FER performance and prediction accuracy. This approach comprises two parts: (i) Initially, a deep convolutional sparse autoencoder is trained with unlabeled samples of facial expressions. Here, sparsity is introduced in the convolutional block to enforce penalties, focusing on more relevant features for feature representation in the latent space. (ii) A trained encoder with a feature map is connected to a fully connected layer with softmax for final fine-tuning with learned weights and labeled facial expression samples in a semi-supervised approach for emotion classification. This approach was experimented with two benchmark datasets, namely CK + and JAFFE, and achieved significant results of 98.98% and 93.10% accuracy, respectively. The results were analyzed using established state-of-the-art techniques. Additionally, eXplainable AI (XAI) methods like Grad-CAM and image-LIME were employed to interpret the performance and prediction outcomes of the DSCSA model.

关键词： Autoencoder Convolutional sparse autoencoder Facial expression recognition Feature representation Unsupervised learning Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

FastBeltNet: a dual-branch light-weight network for real-time conveyor belt edge detection

引用

JOURNAL OF real-time image processing 2024年第4期21卷 1-11页

作者： Zhao, Xing Zeng, Minhao Dong, Yanglin Rao, Gang Huang, Xianshan Mo, Xutao Anhui Univ Technol Sch Microelect & Data Sci Maanshan 243032 Anhui Peoples R China Shanghai Meishan Iron & Steel Co Ltd Steelmaking Plant Nanjing 210039 Jiangsu Peoples R China Anhui Univ Technol Sch Innovat & Entrepreneurship Maanshan 243032 Anhui Peoples R China

Belt conveyors are widely used in multiple industries, including coal, steel, port, power, metallurgy, and chemical, etc. One major challenge faced by these industries is belt deviation, which can negatively impact production efficiency and safety. Despite previous research on improving belt edge detection accuracy, there is still a need to prioritize system efficiency and light-weight models for practical industrial applications. To meet this need, a new semantic segmentation network called FastBeltNet has been developed specifically for real-time and highly accurate conveyor belt edge line segmentation while maintaining a light-weight design. This network uses a dual-branch structure that combines a shallow spatial branch for extracting high-resolution spatial information with a context branch for deep contextual semantic information. It also incorporates the Ghost blocks, Downsample blocks, and Input Injection blocks to reduce computational load, increase processing frame rate, and enhance feature representation. Experimental results have shown that FastBeltNet has performed comparatively better than some existing methods in different real-world production settings, achieving promising performance metrics. Specifically, FastBeltNet achieves 80.49% mIoU accuracy, 99.89 FPS processing speed, 895 k parameters, 8.23 GFLOPs, and 430.95 MB peak CUDA memory use, effectively balancing accuracy and speed for industrial production.

关键词： Belt deviation Machine vision deep learning Edge detection

来源：评论

学校读者我要写书评

暂无评论

A Mobile LiDAR-Based deep learning Approach for real-time 3D Body Measurement

引用

APPLIED SCIENCES-BASEL 2025年第4期15卷

作者： Jeong, Yongho Noh, Taeuk Lee, Yonghak Lee, Seonjae Choi, Kwangil Jeong, Sujin Kim, Sunghwan Konkuk Univ Dept Appl Stat Seoul 05029 South Korea

In this study, we propose a solution for automatically measuring body circumferences by utilizing the built-in LiDAR sensor in mobile devices. Traditional body measurement methods mainly rely on 2D images or manual measurements. This research, however, utilizes 3D depth information to enhance both accuracy and efficiency. By employing HRNet-based keypoint detection and transfer learning through deep learning, the precise locations of body parts are identified and combined with depth maps to automatically calculate body circumferences. Experimental results demonstrate that the proposed method exhibits a relative error of up to 8% for major body parts such as waist, chest, hip, and buttock circumferences, with waist and buttock measurements recording low error rates below 4%. Although some models showed error rates of 7.8% and 7.4% in hip circumference measurements, this was attributed to the complexity of 3D structures and the challenges in selecting keypoint locations. Additionally, the use of depth map-based keypoint correction and regression analysis significantly improved accuracy compared to conventional 2D-based measurement methods. The real-time processing speed was also excellent, ensuring stable performance across various body types.

关键词： LiDAR HRNet deep learning keypoint

来源：评论

学校读者我要写书评

暂无评论

DMAeEDNet: Dense Multiplicative Attention Enhanced Encoder Decoder Network for Ultrasound-Based Automated Breast Lesion Segmentation

引用

IEEE ACCESS 2024年 12卷 60541-60555页

作者： Saini, Manali Afrin, Humayra Sotoudehnia, Setayesh Fatemi, Mostafa Alizad, Azra Mayo Clin Coll Med & Sci Dept Radiol Rochester MN 55905 USA Mayo Clin Coll Med & Sci Dept Physiol & Biomed Engn Rochester MN 55905 USA

Automated and precise segmentation of breast lesions can facilitate early diagnosis of breast cancer. Recent research studies employ deep learning for automatic segmentation of breast lesions using ultrasound imaging. Numerous studies introduce somewhat complex modifications to the well adapted segmentation network, U-Net for improved segmentation, however, at the expense of increased computational time. Towards this aspect, this study presents a low complex deep learning network, i.e., dense multiplicative attention enhanced encoder decoder network, for effective breast lesion segmentation in the ultrasound images. For the first time in this context, two dense multiplicative attention components are utilized in the encoding layer and the output layer of an encoder-decoder network with depthwise separable convolutions, to selectively enhance the relevant features. A rigorous performance evaluation using two public datasets demonstrates that the proposed network achieves dice coefficients of 0.83 and 0.86 respectively with an average segmentation latency of $19 ms$ . Further, a noise robustness study using an in-clinic recorded dataset without pre-processing indicates that the proposed network achieves dice coefficient of 0.72. Exhaustive comparison with some commonly used networks indicate its adeptness with low time and computational complexity demonstrating feasibility in real time.

关键词： Lesions Breast image segmentation Ultrasonic imaging Decoding Feature extraction real-time systems Ultrasound breast lesion segmentation deep learning U-Net convolution neural network

来源：评论

学校读者我要写书评

暂无评论

deep-learning real-time phase retrieval of imperfect diffraction patterns from X-ray free-electron lasers

引用

NPJ COMPUTATIONAL MATERIALS 2025年第1期11卷 1-10页

作者： Lee, Sung Yun Cho, Do Hyung Jung, Chulho Sung, Daeho Nam, Daewoong Kim, Sangsoo Song, Changyong POSTECH Dept Phys Pohang South Korea POSTECH Photon Sci Ctr Pohang South Korea Max Planck POSTECH Korea Res Initiat Ctr Ultrafast Sci Quantum Matter Pohang South Korea POSTECH XFEL Beamline Div Pohang Accelerator Lab Pohang South Korea

Machine learning is attracting surging interest across nearly all scientific areas by enabling the analysis of large datasets and the extraction of scientific information from incomplete data. Data-driven science is rapidly growing, especially in X-ray methodologies, where advanced light sources and detection technologies produce vast amounts of data that exceed meticulous human inspection capabilities. Despite the increasing demands, the full application of machine learning has been hindered by the need for data-specific optimizations. In this study, we introduce a new deep-learning-based phase retrieval method for imperfect diffraction data. This method provides robust phase retrieval for simulated data and performs well on partially damaged and noisy single-pulse diffraction data from X-ray free-electron lasers. Moreover, the method significantly reduces data processing time, facilitating real-time image reconstructions that are crucial for high-repetition-rate data acquisition. This approach offers a reliable solution to the phase problem to be widely adopted across various research areas confronting the inverse problem.

关键词： Pulse repetition rate

来源：评论

学校读者我要写书评

暂无评论

New Statistic Detector for Structural image Similarity

引用

IEEE TRANSACTIONS ON SIGNAL processing 2025年 73卷 1168-1183页

作者： Diaw, Moustapha Retraint, Florent Morain-Nicolier, Frederic Delahaies, Agnes Landre, Jerome Univ Reims CReST EA 3804 IUT Troyes F-51097 Reims France Univ Technol Troyes LIST3N UTT F-10300 Troyes France Jonkoping Univ Comp Sci & Informat Dept S-55318 Jonkoping Sweden

Social networks like LinkedIn, Facebook, and Instagram contribute significantly to the rise of image prevalence in daily life, with numerous images posted in everyday. Detecting image similarity is crucial for many applications. While deep learning methods like Learned Perceptual image Patch Similarity (LPIPS) are popular, they often overlook image structure. An alternative method involves using pre-trained models ($e.g.$, LeNet-$5$ and VGG-$16$) to extract features and employing classifiers. However, deep learning methods demand substantial computational resources and they also suffer from uncontrolled false alarms. This paper proposes a novel Generalized Likelihood Ratio Test (GLRT) detector based on a hypothesis testing framework to identify the similarity of structural image pairs. The proposed approach minimizes the need for extensive computational resources, and false alarms can be regulated by employing a threshold. The detector is applied to Local Dissimilarity Maps (LDM), with gray-level values modeled by a statistical distribution. Experimental results on simulated and real data confirm its effectiveness for structural similarity detection. Additionally, a Simple Likelihood Ratio Test (SLRT) is tested on simulated data. Comparisons with deep learning and classical measures like Structural Similarity Index (SSIM) and Feature Similarity Index (FSIM) show the proposed detector performs comparably or better in terms of Area Under the Curve (AUC) with less computing time, especially for structural similarity.

关键词： Feature extraction Detectors Convolutional neural networks Transformers image quality Noise Computational modeling Indexes image classification Training Classification local dissimilarity map weibull distribution GLRT similarity measures pre-trained CNN

来源：评论

学校读者我要写书评

暂无评论

deep transfer learning-based computer vision for real-time harvest period classification and impurity detection of Porphyra haitnensis

引用

AQUACULTURE INTERNATIONAL 2024年第4期32卷 5171-5198页

作者： Gao, Zhenchang Huang, Jinxian Chen, Jiashun Shao, Tianya Ni, Hui Cai, Honghao Jimei Univ Sch Sci Dept Phys Xiamen Fujian Peoples R China Jimei Univ Coll Food & Biol Engn Xiamen Fujian Peoples R China Fujian Prov Key Lab Food Microbiol & Enzyme Engn Xiamen Fujian Peoples R China

Seaweed has attracted great attention as a healthy and nutritious food. Traditional seaweed processing plants mainly rely on manual visual inspection to identify and remove inferior seaweed. Accurate and rapid harvest period classification and impurity detection are key to improving productivity and processing speed in seaweed processing plants. Although many studies on seaweed have been conducted in laboratory environments, currently, the plants lack effective tools to obtain real-time and reliable information on seaweed quality. To address this challenge, the deep transfer learning-based computer vision was applied to identify inferior seaweeds, including those from the third harvest, fourth harvest, and impure seaweeds in this work. Specifically, YOLOv8 and YOLOv5 were utilized as base transfer learning models. By loading various pre-trained weight files, this study was able to automatically classify Porphyra haitnensis into four categories based on the harvest period and simultaneously detect four types of common impurities in it. Among the tested models, YOLOv8n-cls achieved the best trade-off in classifying the harvest period, with a Top-1 accuracy of 93.5%. This represented a significant improvement of 16% compared to the performance without transfer learning. The detection speed for a single image was 8.2 ms, and the model size was only 2.82 Mb. On the other hand, YOLOv8n exhibited outstanding performance in impurity detection, with a mean average precision of 99.14%, a single image detection speed of 4.3 ms, and a model size of 5.95 Mb. The results demonstrated the potential of YOLOv8 with transfer learning to objectively assist or even replace decision-making by assembly line workers. This study will not only enhance the quality control, production efficiency, and economic benefits of the seaweed processing industry but also drive the automation equipment and systems of seaweed-related enterprises towards greater intelligence and efficiency.

关键词： Transfer learning Computer vision YOLOv8 Seaweed Harvest period Impurity detection

来源：评论

学校读者我要写书评

暂无评论

SP-CrackNet: serial-parallel network with boundary contrastive learning for real-time crack detection

引用

SIGNAL image AND VIDEO processing 2024年第4期18卷 3265-3274页

作者： Xie, Ying Yin, Hui Chong, Aixin Yang, Ying Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing 100044 Peoples R China Beijing Jiaotong Univ Beijing Key Lab Traff Data Anal & Min Beijing 100044 Peoples R China Beijing Jiaotong Univ Frontiers Sci Ctr Smart High Speed Railway Syst Beijing 100044 Peoples R China Beijing Jiaotong Univ Key Lab Beijing Railway Engn Beijing 100044 Peoples R China

Crack detectors based on deep learning have made tremendous progress compared to inefficient traditional inspections. However, missing thin local structures, blurred boundary segmentation and slow inference speed restrict the performance improvement of existing crack detectors. To this end, we propose a serial-parallel network (SP-CrackNet) for pixel-level real-time crack detection. Specifically, a serial-parallel feature extractor with global bottleneck blocks (GB Block) is proposed. Based on the serial-parallel structure, SPFE can enlarge the receiving field and capture the local information of thin cracks effectively, and the GB Block designed by us can ensure the continuity of the slender cracks is sensed with a lower computational cost. Moreover, a boundary contrastive learning scheme (BCL scheme) is designed to enhance the learning ability of SP-CrackNet for crack boundary features, thereby improving the segmentation accuracy of crack boundary regions. Extensive experiments on CFD and deepCrack datasets show that SP-CrackNet outperforms the comparative methods while achieving real-time inference speed.

关键词： real-time crack detection Serial-parallel structure Boundary contrastive learning Global bottleneck block

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review on image Encryption Techniques using Memristor based Chaotic System for Multimedia Application

引用

IETE JOURNAL OF RESEARCH 2024年第11期70卷 8160-8183页

作者： Kumar, B. Sakthi Revathi, R. Sri Eshwar Coll Engn Dept ECE Coimbatore India KLEF Univ Dept ECE Guntur India

image encryption is one of the most widely used techniques for securing images in trusted and unrestricted public media. However, the drawback is weak security in chaotic encryption algorithms, small key space and hardware implementation complexities. Various appropriate encryption algorithms have been carried out for secure image transmission;it becomes a critical issue to protect image integrity, confidentiality and authenticity. To overcome these issues, this article examines several existing memristor-based image encryption algorithms with various chaotic maps. As a result, this research aims to deliver comprehensive literature on image encryption techniques through memristor to help the researchers. Finally, this survey provides all vital literature on the current image encryption techniques with their benefits, drawbacks, developments, and future directions. We also give a general guideline about cryptography. We conclude that all methods are helpful for real-time image encryption. A comparison has been made between many image processing approaches (conventional and memristor) based on metrics such as the number of pixels change rate (NPCR) with the value of 0.9986, histogram, entropy (7.9993), unified average varying intensity (UACI) achieves the value as 0.4996, energy consumption as 0.32pJ and time complexity as 0.9s. The experimental results of various approaches, key sensitivity and statistical analysis revealed that the survey of memristor-based image encryption schemes provides an effective way to secure image transmission in real-time application.

关键词： deep learning image decryption image encryption image processing Memristor Security

来源：评论

学校读者我要写书评

暂无评论

Design of a real-time crime monitoring system using deep learning techniques

引用

INTELLIGENT SYSTEMS WITH APPLICATIONS 2024年 21卷

作者： Mukto, Md. Muktadir Hasan, Mahamudul Al Mahmud, Md. Maiyaz Haque, Ikramul Ahmed, Md. Ahsan Jabid, Taskeed Ali, Md. Sawkat Rashid, Mohammad Rifat Ahmmad Islam, Mohammad Manzurul Islam, Maheen East West Univ Dept Comp Sci & Engn Dhaka 1212 Bangladesh

Criminal Activities and Crime Monitoring System has long been a research topic. In this paper, we propose an effective Crime Monitoring System (CMS) that can detect a crime in real-time using a camera surveillance system and notify the appropriate law enforcement officer. The CMS was proposed to counterbalance human weaknesses such as inattention, slow reaction, and slacking, for example, in detecting crimes. The proposed CMS detects crime scenes by combining the mechanisms and functionalities of closed-circuit television (CCTV) cameras with various deep-learning methods and image-processing techniques. The CMS operates in three stages: weapon detection, violence detection, and face recognition. We used transfer learning models to detect weapons, and violence and used a face recognition algorithm to recognize faces. More specifically, to detect weapons YOLOv5, and MobileNetv2 to detect violence and used face recognition algorithms to recognize faces. The image dataset is used in CMS, while the video dataset is used to train the MobileNet-based violence detection model. In this case, frame-by-frame images extracted from video files were used to train the model. All of the models performed admirably. The weapon detection model detected four different weapon classes with greater than 80% accuracy. The violence detection model is also 95% accurate. The face recognition model had a 97% accuracy rate in detecting faces. The CMS's combined model was tested in a variety of real-world scenarios, and its performance was found to be outstanding. It was able to detect crime incidents and generate timely alarms, demonstrating its effectiveness in providing security and safety.

关键词： Crime monitoring system (CMS) Surveillance Weapon detection Violence detection Face recognition deep learning image processing

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：