Weapon detection in surveillance settings is a critical task for ensuring public safety and security. This system presents an analysis of weapon detection using the state-of-the-art object detection model, You Only Look Once version 4(Yolov4). Yolov4 is a deep learning architecture known for its speed and accuracy in real-time object detection. The goal of this thesis is to investigate the effectiveness of Yolov4 in detecting weapons from images. The system involves training Yolov4 on a large dataset containing weapon categories and evaluating its performance on test images. The dataset used in this system comprises annotated surveillance images capturing various real-world scenarios. The training is carried out on a high-performance computing infrastructure, utilizing GPU acceleration for efficient model convergence. The evaluation phase involves analyzing the performance metrics of the trained model on the test images. The experimental results demonstrate the effectiveness of Yolov4 in accurately detecting weapons from image dataset. The demonstrated performance of Yolov4 highlights its potential for enhancing security measures in various domains, including airports, public spaces, and critical infrastructure facilities. The system was tested and was found to achieve high accuracy in detecting weapons with low false positive rates.
Keywords: Weapon Detection, Deep Learning, Annotated Surveillance Images, Object Detection,

This paper focuses on the changes in the number of scientific research achievements in Tai'an City. In the knowledge-based economy, technological innovation is of great significance to regional development, and an in-depth study of the trends in the number of scientific research achievements in Tai'an City has far-reaching implications. From 2013 to 2023, the number of scientific research achievements in Tai'an City has generally shown a steady upward trend. Among different fields, the industrial and healthcare fields accounted for a larger proportion of the achievements. The proportion in the industrial field has gradually decreased, while that in the healthcare field has steadily increased. Overall, Tai'an City's scientific research capacity has been continuously enhanced, with certain fields demonstrating advantages and potential. To promote further development, it is recommended that the government increase its support for science and technology policies and build a comprehensive system. However, this study has limitations in data collection, with some data being missing, inaccurate, or lacking comparability. These issues need to be further improved in future research.
Keywords: Tai'an City, Number of Scientific Research Achievements, Trend of Change, Technological Innovation, Economic Development.

Sustainable transportation is a critical component of urban development, economic growth, and environmental protection in developing countries like Nepal. With rapid urbanization, increasing vehicle ownership, and deteriorating air quality, Nepal faces significant challenges in establishing an efficient and eco-friendly transport system. This paper examines the current state of transportation in Nepal, identifies key challenges, and explores sustainable solutions such as mass transit systems, non-motorized transport (NMT), electric mobility, and policy reforms. By analyzing successful case studies from other developing nations, this research provides recommendations for Nepal to transition toward a sustainable transportation future. The study draws on academic literature, government reports, and international best practices to propose actionable strategies for policymakers.
Keywords: Sustainable transport, Nepal, electric vehicles, public transit, non-motorized transport, urban mobility, policy reforms.