With the popularity of mobile devices, lightweight deep learning models have important value in various application scenarios. However, how to effectively fuse the feature information from different ...dimensions while ensuring the model’s lightness and high accuracy is a problem that has not been fully solved. In this paper, we propose a novel feature fusion module, called φunit, which can fuse the features extracted by different dimensional networks according to the order of feature information with a small computational cost, avoiding the problems of information fragmentation caused by simple feature stacking in traditional information fusion. Based on φunit, this paper further builds an extremely lightweight model φNet, which can achieve performance close to the highest accuracy on several public datasets under the condition of very limited parameter scale. The core idea of φunit is to use deconvolution to reduce the discrepancy among the features to be fused, and to lower the possibility of feature information fragmentation after fusion by fusing the features from different dimensions sequentially. φNet is a lightweight network composed of multiple φunits and bottleneck modules, with a parameter scale of only 1.24 M, much smaller than traditional lightweight models. This paper conducts experiments on public datasets, and φNet achieves an accuracy of 71.64% on the food101 dataset, and an accuracy of 75.31% on the random 50-category food101 dataset, both higher than or close to the highest accuracy. This paper provides a new idea and method for feature fusion of lightweight models, and also provides an efficient model selection for deep learning applications on mobile devices.
Mobile forensics, particularly in the Android ecosystem, is a rapidly evolving field that demands continuous advancements to address the growing complexity and diversity of mobile devices. This ...article emphasizes the importance of developing techniques for digitally analyzing Android smartphones, which dominate the smartphone market. The primary objective of this research is to contribute to the development of effective forensic investigation strategies tailored specifically for Android mobile devices, providing insights into the tools and methods used for this purpose. The objective of this study is to improve the precision and effectiveness of forensic examinations pertaining to Android mobile phones. It discusses the fundamental functionality of mobile devices as a source of digital evidence and provides an overview of tools and methodologies for collecting and analyzing such evidence. The importance of comprehending the hardware and software architecture of Android handsets in order to choose the right forensic tools is also highlighted in the article. Furthermore, it proposes future enhancements for Andriller, a popular digital forensic tool, to improve its effectiveness in Android forensic investigations. These enhancements include advancements in data extraction techniques, compatibility with new Android versions, support for additional data types, integration with advanced analysis methods, and addressing identified limitations. Additionally, the paper stresses the need for robust methodologies for conducting cloud forensics on Android devices, particularly in the context of data stored in cloud storage services. The proposed work aims to enhance the capabilities of Andriller and improve the efficiency of digital forensic investigations on Android devices.
Criminalistica mobilă ONOFREI-RIZA, Deniss Bogdan
Revista română de informatică și automatică = Romanian journal of information technology and automatic control,
12/2023, Letnik:
29, Številka:
1
Journal Article
Recenzirano
Odprti dostop
The "Informational Revolution" as well as the fantastic development of the "IT world" come and mark the XXI century, as a witness to the emergence and continued evolution of mobile devices. Compact ...equipment that combines traditional cell phone features with personal computer functionality, today's mobile devices are the result of a growing number of hardware and operating systems (Android, iOS, Windows Phone, BlackBerry). Today's smart mobile devices are being used less and less for calls and more and more for socializing and information! Thus, the intelligent mobile device has become a complex storage of sensitive data that helps identify the "behavior" of its owner. This has generated the forensic evolution of mobile devices, a branch of "digital forensics" dealing with the recovery of data stored by a mobile device.
Mobile devices are a ubiquitous part of American life, yet how families use this technology has not been studied. We aimed to describe naturalistic patterns of mobile device use by caregivers and ...children to generate hypotheses about its effects on caregiver-child interaction.
Using nonparticipant observational methods, we observed 55 caregivers eating with 1 or more young children in fast food restaurants in a single metropolitan area. Observers wrote detailed field notes, continuously describing all aspects of mobile device use and child and caregiver behavior during the meal. Field notes were then subjected to qualitative analysis using grounded theory methods to identify common themes of device use.
Forty caregivers used devices during their meal. The dominant theme salient to mobile device use and caregiver-child interaction was the degree of absorption in devices caregivers exhibited. Absorption was conceptualized as the extent to which primary engagement was with the device, rather than the child, and was determined by frequency, duration, and modality of device use; child response to caregiver use, which ranged from entertaining themselves to escalating bids for attention, and how caregivers managed this behavior; and separate versus shared use of devices. Highly absorbed caregivers often responded harshly to child misbehavior.
We documented a range of patterns of mobile device use, characterized by varying degrees of absorption. These themes may be used as a foundation for coding schemes in quantitative studies exploring device use and child outcomes.
Health information like heart rate (HR) and electrocardiogram (ECG) patterns are available to the public on smartwatches; however, there may be a disconnect between these health measures and how ...users subjectively experience feelings of stress. This study examines the health detection features of two leading smartwatches in the industry, the Apple Watch Series 6 and Fitbit Sense, to determine if these devices may be used to accurately measure stress. Participants engaged in a multi-tasking program (MATB-II) that varied in cognitive workload demand while wearing smartwatches measuring cardiac data. Subjective workload responses resulted in significant differences between low and high workload conditions, indicating an increase in stress. However, both smartwatches were unable to detect significant differences in stress responses between low and high workload conditions. Overall, these results indicate that smartwatch HR and ECGs may not reflect internal feelings of stress and are sensitive to variability in measurement.
Many children are spending more time with screen media than has been recommended by the American Academy of Pediatrics. There is evidence that parent television use is associated with higher levels ...of child television time, but we know little about what predicts children's media use with other technology. Using a nationally representative sample of more than 2300 parents of children ages 0–8, children's time spent with four digital media devices – television, computers, smartphones, and tablet computers – was examined. Results from linear regression analyses indicate across all four platforms that parents' own screen time was strongly associated with child screen time. Further analyses indicate that child screen time use appears to be the result of an interaction between child and parent factors and is highly influenced by parental attitudes. Results suggest that policymakers should consider the family environment as a whole when developing policy to influence children's screen media use at home.
•Parent screen time is the strongest predictor of child screen time.•Parent screen time and parent attitudes influence child screen time.•Child screen time varies as a function of child age and by device.
The aim of this paper is to analyse the acceptance of online games, based on the unified theory of acceptance and use of technology 2 (UTAUT 2). The data analysed correspond to a sample of online ...players through mobile devices in Spain. A structural equation approach based on Partial Least Squares was used to assess the acceptance model. The result of the analysis indicates that UTAUT 2 explains 71% of the use of online games in mobile devices. The main conclusion of the study highlights the importance of habit in the use of online games. Specifically, the intention to play online is explained, in order of importance, by the variables habit, hedonic motivation and social identity. In addition, the use of an online game is determined by habit and intention to play. We proposed a simplified UTAUT2 model adapted to the online game scope.
This study investigated how caregivers’ mobile device use influenced the quality of their interactions with their children. The associations between quality of interactions and the type of activity ...(eg, typing/swiping, looking at screen), setting, caregiver-child proximity, and child behaviors were also examined.
Researchers anonymously and systematically observed and coded the behavior of 98 caregiver-child dyads in public settings (eg, parks, food courts) during real-time, naturally occurring interactions using time sampling.
Caregivers who used a mobile device for the entire observation engaged in less joint attention and were less responsive than caregivers who used the device some of the time. When looking at patterns within caregivers who used the device intermittently, the probability that they would engage in joint attention, initiate interactions with their child, talk, and display positive emotions was lower when they used a mobile device than when they did not. Child talking and positive affect were unrelated to caregiver device use. Activity type with the device, caregiver-child proximity and setting also related to interaction quality.
Caregiver device use was negatively associated with adult behaviors that are key components of high-quality caregiver-child interactions. Additionally, setting, activity type, and caregiver-child proximity are factors that should be considered because they relate to the quality of caregiver-child interactions in the context of mobile screen technologies.
•A new dataset of iris images acquired by mobile devices can support researchers.•MICHE-I will assist with developing continuous authentication to counter spoofing.•The dataset includes images from ...different mobile devices, sessions and conditions.
We introduce and describe here MICHE-I, a new iris biometric dataset captured under uncontrolled settings using mobile devices. The key features of the MICHE-I dataset are a wide and diverse population of subjects, the use of different mobile devices for iris acquisition, realistic simulation of the acquisition process (including noise), several data capture sessions separated in time, and image annotation using metadata. The aim of MICHE-I dataset is to make up the starting core of a wider dataset that we plan to collect, with the further aim to address interoperability, both in the sense of matching samples acquired with different devices and of assessing the robustness of algorithms to the use of devices with different characteristics. We discuss throughout the merits of MICHE-I with regard to biometric dimensions of interest including uncontrolled settings, demographics, interoperability, and real-world applications. We also consider the potential for MICHE-I to assist with developing continuous authentication aimed to counter adversarial spoofing and impersonation, when the bar for uncontrolled settings raises even higher for proper and effective defensive measures.