Best Feature for CNN Classification of Human Activity Using IOT Network

Amroun, Hamdi; Temkit, MʼHamed; Ammi, Mehdi

doi:10.1109/ithings-greencom-cpscom-smartdata.2017.145

Cited by 7 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this approach suffers from limitations such as inadequate representation of various image variations, timeconsuming manual design for large datasets, and an inability to adapt to diverse tasks or image conditions. The fixed nature of predefined features restricts the capture of complex relationships within the data, keeping recognition accuracy at 70% [15]. Manual feature design takes a lot of time for large datasets and is not scalable.…”

Section: Methodsmentioning

confidence: 99%

Decoding Innocence: Advancing Forensic Facial Discrimination through Comparative Analysis of Conventional CNN and Advanced Architectures

Ahmed,

Abdullah

2024

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

The field of Digital Image Forensics (DIF) faces a critical issue in accurately identifying children in digital images, notably in cases involving the proliferation of child sexual abuse content. Existing techniques face hurdles due to model architecture limitations, dataset suitability concerns, and classification imbalance, impeding their ability to recognize children to deter pornographic images. Addressing this challenge, this study introduces Implicit Feature Extraction (IFE), a specialized approach for distinguishing child and adult images in object detection. Leveraging Convolutional Neural Networks (CNNs), the IFE method automates the extraction of discriminative facial features, surpassing the constraints of Explicit Feature Extraction (EFE) methods, which achieve an accuracy of around 70%. The research focuses on three core objectives introducing IFE for detailed face feature detection in DIF's child and adult image identification, implementing IFE with CNNs to enhance image classification, and conducting a thorough evaluation of the proposed technique's performance using key metrics like accuracy and balanced classification results and comparing the result with a basic CNN model’s performance. This research's significance lies in its notable contributions to digital image forensics, particularly in combatting child exploitation. The fusion of IFE with CNNs showcases 92% accuracy in distinguishing child and adult images, promising advancements with practical implications in child protection and forensic investigations. The comprehensive evaluation using the UTKFace dataset underscores the proposed technique's efficacy, marking a substantial improvement in child image identification within digital image forensics.

show abstract

Section: Methodsmentioning

confidence: 99%

Decoding Innocence: Advancing Forensic Facial Discrimination through Comparative Analysis of Conventional CNN and Advanced Architectures

Ahmed,

Abdullah

2024

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

show abstract

“…They processed the time series through short-time Fourier transform (STFT) spectrograms, then designed a deep learning-based human activity recognition architecture, and finally achieved accurate real-time classification. Amroun et al [ 14 ] collected four types of activity data, including standing, sitting, lying down and walking, to extract the best feature descriptors of activities, and identified human activities through the CNN model, with a recognition accuracy rate of over 98%. Reference [ 15 ] designed a LSTM network, then performed experimental evaluation on three standard benchmark (Opportunity, PAMAP2, Skoda) datasets, and finally achieved better recognition results.…”

Section: Related Workmentioning

confidence: 99%

A Hierarchical Ensemble Deep Learning Activity Recognition Approach with Wearable Sensors Based on Focal Loss

Zhao

Chen²,

Bai³

et al. 2022

IJERPH

View full text Add to dashboard Cite

Abnormal activity in daily life is a relatively common symptom of chronic diseases, such as dementia. There will probably be a variety of repetitive activities in dementia patients’ daily life, such as repeated handling of objects and repeated packing of clothes. It is particularly important to recognize the daily activities of the elderly, which can be further used to predict and monitor chronic diseases. In this paper, we propose a hierarchical ensemble deep learning activity recognition approach with wearable sensors based on focal loss. Seven basic everyday life activities including cooking, keyboarding, reading, brushing teeth, washing one’s face, washing dishes and writing are considered in order to show its performance. Based on hold-out cross-validation results on a dataset collected from elderly volunteers, the average accuracy, precision, recall and F1-score of our approach are 98.69%, 98.05%, 98.01% and 97.99%, respectively, in identifying the activities of daily life for the elderly.

show abstract

“…In order to set the scene for researchers in the years to come, an overview of the approaches that have been utilized in the areas of Internet of Things-big data and Internet of Things-data mining is provided here in the context of three categories. According to research by Amroun et al [98], using a convolutional neural network is the best way to describe human movement. Walking, sitting, standing, and laying down are the four different types of actions for which we opted to assign directions.…”

Section: Feature Extractionmentioning

confidence: 99%

Big Data Privacy and Security Using Abundant Data Recovery Techniques and Data Obliviousness Methodologies

Funde

Swain

2022

IEEE Access

View full text Add to dashboard Cite

The concept of big data security is introduced in this article along with many features. It illustrates the need for security in healthcare systems as the volume of data increases continuously over the period of time. The necessity of big data security as well as several big data analytics phases highlighted. It covers many big data privacy-preserving strategies. Many digital storage solutions being used in today's world are designed to work only with fixed format of the data. This paper introduces some methods for maintaining metadata obliviousness. The oblivious RAM technology mentioned in the research article address security concerns and it can be handled with the daily increase in data in several industries. Security needs are introduced at many phases of big data creation, such as information extraction, storage systems, and analytics of the information. Additionally, it presents several data recovery methods for recovering original data in the event of a data crash. This paper covers several data categorization methods for sorting data into normal and sensitive categories as well as methods for anomaly detection. It discusses the advantages and disadvantages of various security measures.INDEX TERMS Security, privacy, obliviousness, data recovery. I. INTRODUCTIONThe core concepts and phases of big data are introduced in this section. At several stages of big data production, such as data generation, data storage, and data analytics, security requirements are introduced. The importance of security in big data, cloud, and Internet of Things (IoT) infrastructure is discussed in this section. It discusses about several security measures that must be defended against various kinds of attacks. It discusses on the security of healthcare systems. It gives a general overview of the medical industry and discusses how patient data must be safeguarded if it is stored in a dispersed setting. A. BIG DATABig data is data that has large volume, heterogeneity, speed, and volatility all at the same time. The term "high volume" denotes to the significant quantity of data that is produced on a daily basis by a variety of companies and organizations. The rapid proliferation of new data types, many of which are going to be incorporated in current datasets that have been acquired from a variety of companies and devices, is referred to as heterogeneity. The term "high speed" refers to the rapid rate at which data is collected or captured from a variety of social networking sites into a database [1]. Huge amounts of data present additional difficulties to the security systems that are already in place due to the variety, volume, and unpredictability of the data. Most data storage models used today are designed to work with data that has been properly arranged. The currently available encryption methods are inefficient for use with massive data as once data has been encrypted, producing keys, encrypting it, and then decrypting it takes a significant amount of time. For instance, the data collected from patients, various sensors, soc...

show abstract

Best Feature for CNN Classification of Human Activity Using IOT Network

Cited by 7 publications

References 20 publications

Decoding Innocence: Advancing Forensic Facial Discrimination through Comparative Analysis of Conventional CNN and Advanced Architectures

Decoding Innocence: Advancing Forensic Facial Discrimination through Comparative Analysis of Conventional CNN and Advanced Architectures

A Hierarchical Ensemble Deep Learning Activity Recognition Approach with Wearable Sensors Based on Focal Loss

Big Data Privacy and Security Using Abundant Data Recovery Techniques and Data Obliviousness Methodologies

Contact Info

Product

Resources

About