Search

A door security system utilizing face recognition technology based on TensorFlow Lite has been developed to enhance access security and convenience. This research aims to design a system capable of accurately recognizing faces in real time, integrating it with door lock devices, and ensuring user data security. Employing the waterfall method, the system was implemented using an ESP32-CAM microcontroller and deep learning algorithms. Testing results demonstrated a face recognition accuracy of 91% in identifying and processing commands from 200 trials with ten facial variations. Successful integration with door lock devices was achieved through serial communication. The system also features activity log recording for monitoring purposes. This solution offers greater practicality and security than RFID systems as it eliminates the need for physical cards. This research contributes to developing more sophisticated and user-friendly home security systems, with the potential for further enhancements in recognition capabilities under various lighting conditions and integration with other biometric technologies

In the pursuit of organizational goals, addressing the fundamental needs of employees, such as ensuring the availability of drinking water, plays a pivotal role in fostering a productive, healthy, and value-aligned work environment. This imperative necessitates the implementation of intelligent service management solutions that are both efficient and intelligent. In this research, we propose a visionary architectural transformation that seamlessly integrates intelligent agent-based smart systems within the domain of Smart Organizations. Our novel architectural approach draws inspiration from Activity Theory, strategically orchestrating interactions between employees, company objectives, and tools. Additionally, we employ a meticulously crafted 4-layer organizational system structure to delineate roles, establish rules, foster communities, and optimize task allocation. This forward- looking architecture is further fortified through the integration of cutting-edge technologies, encompassing the Internet of Things (IoT) with precision dispensing scales for real-time predictive capabilities, human-computer interaction (HCI) mobile applications and web services to enhance user-system engagement, and pervasive artificial intelligence (AI) implementations across every layer of the system. The architectural framework is consolidated within a robust Big Data platform, enabling the collection and comprehensive analysis of data on a grand scale. This multifaceted approach seeks to usher in profound and all-encompassing changes in the management of drinking water supply services, encompassing organizational paradigms, technological advancements, and methodological enhancements to enhance efficiency and effectiveness. The focus of this transformation is primarily directed toward companies involved in supply activities, with particular emphasis on drinking water provisioning. Through the development of predictive system prototypes and meticulous performance analysis, our study conclusively demonstrates that the Smart System-based prediction architecture significantly enhances operational performance and elevates the quality of drinking water supply services.

Background Children with nephrotic syndrome need high-dose corticosteroids to achieve remission. Studies have estimated a 35-43% risk of obesity in these patients after corticosteroid treatment. Objective To determine the prevalence of obesity in children who received corticosteroids for nephrotic syndrome, and to compare the risk of obesity in children with steroid-sensitive nephrotic syndrome (SSNS) and steroid-resistant nephrotic syndrome (SRNS). Methods We performed a retrospective cohort study in 50 children with SSNS or SRNS who received corticosteroid treatment. Obesity was defined to be a BMI-for-age Z-score above +2.0 SD, according to the WHO Growth Reference 2007. Central obesity was defined to be a waist-to-height ratio > 0.50. Results The overall prevalence of obesity was 22%, with 29% and 14% in the SSNS and SRNS groups, respectively. The overall prevalence of central obesity was 50%, with 54% and 46% in the SSNS and SRNS groups, respectively. The cumulative steroid doses in this study were not significantly different between the SSNS and SRNS groups. There were also no significant differences between groups for risk of obesity (RR 2.53; 95%CI 0.58 to 10.99) or central obesity (RR 1.39; 95%CI 0.45 to 4.25). Conclusion In children with nephrotic syndrome who received corticosteroids, the prevalence of obesity is 22% and of central obesity is 50%. In a comparison of SSNS and SRNS groups, cumulative steroid dose as well as risks of obesity and central obesity do not significantly differ between groups.