با سلام خدمت کاربران در صورتی که با خطای سیستم پرداخت بانکی مواجه شدید از طریق کارت به کارت (6037997535328901 بانک ملی ناصر خنجری ) مقاله خود را دریافت کنید (تا مشکل رفع گردد).
ردیف | عنوان | نوع |
---|---|---|
11 |
GAFL: Global adaptive filtering layer for computer vision
GAFL: لایه فیلتر تطبیقی جهانی برای بینایی کامپیوتر-2022 We devise a universal global adaptive filtering layer, GAFL, capable of ‘‘learning’’ optimal frequency filter for
each image in a dataset together with the weights of the base neural network that performs some computer
vision task. The proposed approach takes the source image in the spatial domain, selects the best frequencies
in the Fourier domain for the benefit of the global task, and prepends the inverse-transform image to the
main neural network for a joint training. Remarkably, such a simple add-on layer, capable of optimizing the
frequency content of an input for a specific task, dramatically improves the performance of the main network
regardless of its design. We observe that the light networks gain a noticeable boost in the performance metrics;
whereas, the training of the heavy ones converges faster when GAFL is prepended to the main architecture.
We showcase the performance of the layer in four classical computer vision tasks: classification, segmentation,
denoising, and erasing, considering popular natural and medical data benchmarks.
keywords: Adaptive neural layer | Efficient training | Fourier filtering |
مقاله انگلیسی |
12 |
Co-segmentation inspired attention module for video-based computer vision tasks
ماژول توجه الهام گرفته از تقسیم بندی مشترک برای وظایف بینایی کامپیوتری مبتنی بر ویدئو-2022 Video-based computer vision tasks can benefit from estimation of the salient regions and interactions between
those regions. Traditionally, this has been done by identifying the object regions in the images by utilizing
pre-trained models to perform object detection, object segmentation and/or object pose estimation. Although
using pre-trained models is a viable approach, it has several limitations in the need for an exhaustive annotation
of object categories, a possible domain gap between datasets and a bias that is typically present in pre-trained
models. In this work, we propose to utilize the common rationale that a sequence of video frames capture a
set of common objects and interactions between them, thus a notion of co-segmentation between the video
frame features may equip the model with the ability to automatically focus on task-specific salient regions
and improve the underlying task’s performance in an end-to-end manner. In this regard, we propose a generic
module called ‘‘Co-Segmentation inspired Attention Module’’ (COSAM) that can be plugged in to any CNN
model to promote the notion of co-segmentation based attention among a sequence of video frame features.
We show the application of COSAM in three video-based tasks namely: (1) Video-based person re-ID, (2) Video
captioning, & (3) Video action classification and demonstrate that COSAM is able to capture the task-specific
salient regions in video frames, thus leading to notable performance improvements along with interpretable
attention maps for a variety of video-based vision tasks, with possible application to other video-based vision
tasks as well.
keywords: توجه | تقسیم بندی مشترک | شناسه شخص | زیرنویس ویدیویی | طبقه بندی ویدیویی | Attention | Co-segmentation | Personre-ID | Video-captioning | Video classification |
مقاله انگلیسی |
13 |
Mixed Quantum-Classical Method For Fraud Detection with Quantum Feature Selection
روش ترکیبی کوانتومی-کلاسیک برای تشخیص تقلب با انتخاب ویژگی کوانتومی-2022 This paper presents a first end-to-end application of a Quantum Support Vector Machine
(QSVM) algorithm for a classification problem in the financial payment industry using the IBM Safer
Payments and IBM Quantum Computers via the Qiskit software stack. Based on real card payment
data, a thorough comparison is performed to assess the complementary impact brought in by the current
state-of-the-art Quantum Machine Learning algorithms with respect to the Classical Approach. A new
method to search for best features is explored using the Quantum Support Vector Machine’s feature map
characteristics. The results are compared using fraud specific key performance indicators: Accuracy, Recall,
and False Positive Rate, extracted from analyses based on human expertise (rule decisions), classical
machine learning algorithms (Random Forest, XGBoost) and quantum-based machine learning algorithms
using QSVM. In addition, a hybrid classical-quantum approach is explored by using an ensemble model
that combines classical and quantum algorithms to better improve the fraud prevention decision. We found,
as expected, that the results highly depend on feature selections and algorithms that are used to select them.
The QSVM provides a complementary exploration of the feature space which led to an improved accuracy
of the mixed quantum-classical method for fraud detection, on a drastically reduced data set to fit current
state of Quantum Hardware.
INDEX TERMS: Fraud Detection | Quantum | Feature Selection | QSVM | Quantum Kernel Alignment |
مقاله انگلیسی |
14 |
VisuaLizations As Intermediate Representations (VLAIR): An approach for applying deep learning-based computer vision to non-image-based data
تجسم ها به عنوان بازنمایی های میانی (VLAIR): رویکردی برای به کارگیری بینایی کامپیوتری مبتنی بر یادگیری عمیق برای داده های غیر مبتنی بر تصویر-2022 Deep learning algorithms increasingly support automated systems in areas such as human activity
recognition and purchase recommendation. We identify a current trend in which data is transformed
first into abstract visualizations and then processed by a computer vision deep learning pipeline. We
call this VisuaLization As Intermediate Representation (VLAIR) and believe that it can be instrumental
to support accurate recognition in a number of fields while also enhancing humans’ ability to
interpret deep learning models for debugging purposes or for personal use. In this paper we describe
the potential advantages of this approach and explore various visualization mappings and deep
learning architectures. We evaluate several VLAIR alternatives for a specific problem (human activity
recognition in an apartment) and show that VLAIR attains classification accuracy above classical
machine learning algorithms and several other non-image-based deep learning algorithms with several
data representations.
keywords: تجسم اطلاعات | شبکه های عصبی کانولوشنال | تشخیص فعالیت های انسانی | خانه های هوشمند | بازنمایی داده ها | نمایندگی های میانی | تفسیر پذیری | یادگیری ماشین | یادگیری عمیق | Information visualization | Convolutional neural networks | Human activity recognition | Smart homes | Data representation | Intermediate representations | Interpretability | Machine learning | Deep learning |
مقاله انگلیسی |
15 |
Animal biometric assessment using non-invasive computer vision and machine learning are good predictors of dairy cows age and welfare: The future of automated veterinary support systems
ارزیابی بیومتریک حیوانات با استفاده از بینایی کامپیوتری غیرتهاجمی و یادگیری ماشینی پیشبینیکننده خوبی برای سن و رفاه گاوهای شیری هستند: آینده سیستمهای پشتیبانی خودکار دامپزشکی-2022 Digitally extracted biometrics from visible videos of farm animals could be used to automatically assess animal
welfare, contributing to the future of automated veterinary support systems. This study proposed using non-
invasive video acquisition and biometric analysis of dairy cows in a robotic dairy farm (RDF) located at the
Dookie campus, The University of Melbourne, Australia. Data extracted from dairy cows were used to develop
two machine learning models: a biometrics regression model (Model 1) targeting (i) somatic cell count, (ii)
weight, (iii) rumination, and (iv) feed intake and a classification model (Model 2) mapping features from dairy
cow’s face to predict animal age. Results showed that Model 1 achieved a high correlation coefficient (R = 0.96),
slope (b = 0.96), and performance, and Model 2 had high accuracy (98%), low error (2%), and high performance
without signs of under or overfitting. Models developed in this study can be used in parallel with other models to
assess milk productivity, quality traits, and welfare for RDF and conventional dairy farms. keywords: هوش مصنوعی | فیزیولوژی گاو | ماستیت | بیومتریک حیوانات | سنجش از راه دور برد کوتاه | Artificial intelligence | Cows physiology | Mastitis | Animal biometrics | Short range remote sensing |
مقاله انگلیسی |
16 |
Practical Quantum K-Means Clustering: Performance Analysis and Applications in Energy Grid Classification
خوشهبندی کاربردی کوانتومی K-Means: تحلیل عملکرد و کاربردها در طبقهبندی شبکه انرژی-2022 In this work, we aim to solve a practical use-case of unsupervised clustering that has applications in predictive maintenance in the energy operations sector using quantum computers. Using only cloud
access to quantum computers, we complete thorough performance analysis of what some current quantum
computing systems are capable of for practical applications involving nontrivial mid-to-high-dimensional
datasets. We first benchmark how well distance estimation can be performed using two different metrics
based on the swap-test, using angle and amplitude data embedding. Next, for the clustering performance
analysis, we generate sets of synthetic data with varying cluster variance and compare simulation to physical
hardware results using the two metrics. From the results of this performance analysis, we propose a general,
competitive, and parallelized version of quantum k-means clustering to avoid some pitfalls discovered due
to noisy hardware and apply the approach to a real energy grid clustering scenario. Using real-world German
electricity grid data, we show that the new approach improves the balanced accuracy of the standard quantum
k-means clustering by 67.8% with respect to the labeling of the classical algorithm.
INDEX TERMS: Cloud quantum computing | quantum clustering | quantum computing | quantum distance estimation. |
مقاله انگلیسی |
17 |
Quantum–Classical Image Processing for Scene Classification
پردازش تصویر کوانتومی کلاسیک برای طبقه بندی صحنه-2022 Deep-learning-based convolutional neural network (CNN) models are prominent in processing and analyzing
sensor signal data, such as images for classification. Data augmentation is a powerful technique used in training
such models to avoid overfitting and to improve accuracy. This letter proposes a data augmentation technique using
a quantum circuit for image data. The proposed quantum circuit is suitable to implement on real hardware provided by
the IBM Quantum Experience platform. In comparison with other classical data augmentation techniques, the proposed
technique increased the prediction accuracy of the CNN from 68.65 to 76.03%. However, CNN models for image
classification use many parameters during the training process. Quantum computers can efficiently handle large-scale
data inputs using qubits for information processing. Hence, we also propose a hybrid quantum–classical convolutional
neural network model (HQCNN) for scene classification. The proposed model uses a combination of CNN layers and
quantum layers to process images. The proposed HQCNN reduces parameters used for training due to the use of quantum
layers in the model. Our experimental results show that the proposed HQCNN can classify the scenes in the UC Merced
land-use dataset with an accuracy of 85.28% compared to the other models.
Index Terms—Sensor signal processing | hybrid model | quantum–classical computing | scene classification | sensor signal processing. |
مقاله انگلیسی |
18 |
Quantum Machine Learning Applications in the Biomedical Domain: A Systematic Review
کاربردهای یادگیری ماشین کوانتومی در حوزه زیست پزشکی: مرور سیستماتیک-2022 Quantum technologies have become powerful tools for a wide range of application disciplines,
which tend to range from chemistry to agriculture, natural language processing, and healthcare due to
exponentially growing computational power and advancement in machine learning algorithms. Furthermore,
the processing of classical data and machine learning algorithms in the quantum domain has given rise to
an emerging field like quantum machine learning. Recently, quantum machine learning has become quite a
challenging field in the case of healthcare applications. As a result, quantum machine learning has become
a common and effective technique for data processing and classification across a wide range of domains.
Consequently, quantum machine learning is the most commonly used application of quantum computing.
The main objective of this work is to present a brief overview of current state-of-the-art published articles
between 2013 and 2021 to identify, analyze, and classify the different QML algorithms and applications in the
biomedical field. Furthermore, the approach adheres to the requirements for conducting systematic literature
review techniques such as research questions and quality metrics of the articles. Initially, we discovered
3149 articles, excluded the 2847 papers, and read the 121 full papers. Therefore, this research compiled
30 articles that comply with the quantum machine learning models and quantum circuits using biomedical
data. Eventually, this article provides a broad overview of quantum machine learning limitations and future
prospects.
INDEX TERMS: Quantum computing | quantum machine learning | biomedical and healthcare. |
مقاله انگلیسی |
19 |
Multi-Ontology Mapping Generative Adversarial Network in Internet of Things for Ontology Alignment
نگاشت چند هستی شناسی شبکه متخاصم مولد در اینترنت اشیا برای تراز هستی شناسی-2022 On the Semantic web, ontologies are thought to be the remedy to data heterogeneity, and
correlating ontologies is a highly effective technique. Although the use of representation
learning approaches to a variety of applications has showed significant promise, they have had
little effect on the issue of ontology matching and classification. In order to establish
alignments between two ontologies, this research presents the Multi-Ontology Mapping
Generative Adversarial Network in Internet of Things (MOMGANI). For the instance of
ontology mapping, we suggest using a two-system representation learning network consisting
of a Generator and Discriminator. The Generator applies a probabilistic softmax classifier to
the different Name, Label, Comments, Properties, Instance descriptions, concept
characteristics, and the neighbourhood concepts for each of the ontologys properties. In order
to support the assertions that the Generator has generated, the Discriminator network employs
a novel Bidirectional Long Short-Term Memory (Bi-LSTM network) with an Ontology
Attention mechanism enhanced by the concept’s descriptions. As a result, both systems are in
a feedback mechanism where they can learn from one another. The system will produce a set
of triples that list all the associated concepts from various ontologies as its final product.
Domain experts will review these triples outside of the band to ensure that only true concepts
and triples are chosen for the alignment. In comparison to using the ontologies separately, the
aligned ontology enables extended querying and inference across related ontologies and
domains. Considering metrics like recall, precision, and F-measure, the experimental
evaluation was performed utilizing the datasets for classes alignment, property alignment, and
instances alignment. The proposed architecture provides a recall, precision, and F-measure of
0.92, 0.99, and 0.83 respectively which reveals that this model outperforms the traditional
methods.
Keywords: Generative adversarial network | Ontology alignment | IoT and OntoGenerator and OntoLSTM |
مقاله انگلیسی |
20 |
Secure firmware Over-The-Air updates for IoT: Survey, challenges, and discussions
به روز رسانی ایمن میان افزار خارج از هوا برای اینترنت اشیا: بررسی، چالش ها و بحث ها-2022 The Internet of Things (IoT) market has shown strong growth in recent years, where many
manufacturers of IoT devices and IoT-related service providers are competing. Time to market
has become essential to be competitive. The faster a competitor develops and integrates his
product, the more likely he is to dominate the market. This competition could lead to critical
security issues due to the lack of testing or the short development time. Moreover, lots of
IoT devices present some vulnerabilities that can be exploited by attackers. They are also
constantly subject to Zero-days, which require quick intervention to maintain the security of
the environments in which they are deployed in. For these purposes, the quick update of the
firmware image of these IoT devices is an effective way to counter most of these attacks. This
document starts by defining the firmware update mechanisms for IoT, and in particular the ones
done Over-The-Air. Then presents a state-of-the-art of the currently proposed solutions, with the
particularity of surveying from the literature, the standardization bodies and from some well known industrial solutions. It also proposes a new classification of the different types of System
on Chip (SoC) present in the marketed IoT devices together with an analysis of the different
challenges and threats related to the OTA update. The objective is to open up the horizon for
future research directions.
keywords: دستگاه های اینترنت اشیا | نرم افزار | به روز رسانی | خارج از هوا | امنیت | حریم خصوصی | بررسی | مرور | چالش ها | تهدیدات | زنجیره اعتماد | IoT devices | Firmware | Updates | Over-The-Air | Security | Privacy | State of the art | Survey | Challenges | Threats | Trust chain |
مقاله انگلیسی |