This study aims to develop a pothole detection system as an auxiliary tool for ensuring a safe driving environment, enhancing driver safety, and preventing accidents. Potholes are depressions on the road surface, and the number of traffic accidents caused by potholes is increasing yearly. A large-scale pothole image dataset is required for effective pothole detection, but building it requires significant time and money. Recently, pothole image data has been significantly lacking, and the need to develop a dataset for efficient pothole detection research is emerging. The study’s details are outlined as follows: a pothole image dataset is developed for effective pothole detection. 300 images with potholes are collected using a map application’s load view feature. The dataset is then expanded to 2,700 images by diversifying pothole types and applying image augmentation through a generated model on the collected data. Following this, a system is proposed to enhance the accuracy of pothole detection and identify potholes. This study addresses the challenges of manpower and time required for data collection. The development of large-scale datasets can contribute significantly to the enhancement of pothole detection and related research. Future research aims to develop a real-time pothole detection system to prevent accidents caused by potholes while driving.

This study delves into ammunition management, focusing on the 81mm mortar high-explosive shell. Leveraging the Ammunition Stockpile Reliability Program (ASRP) and employing robust statistical methods, the research identifies factors influencing shelf-life prediction through outlier detection, multicollinearity assessment, and linear regression analysis. Meticulous analysis of ASRP data reveals key factors like ‘Standard Deviation of Mean Ammunition Velocity’ and ‘Stabilizer,’ crucial for determining functional grade and predicted shelf-life. The findings contribute to academic discourse and hold practical implications for the Republic of Korea Armed Forces (ROK Armed Forces). Emphasizing the importance of rigorous testing protocols, the study bridges theoretical insights with practical applications, paving the way for informed ammunition management practices. The developed predictive models and methodologies can inform future studies, enhancing defense capabilities and budget efficiency.