-
-
วิทยาลัยนวัตกรรมการผลิตขั้นสูง
The objective of this research is to utilize waste slag in industrial applications and help mitigate flooding, water accumulation, and ponding issues. Currently, slag from the steel smelting or refining process is commonly used as a component in construction materials, such as road surfaces. However, slag has properties that make it difficult for water to permeate, leading to poor drainage and increased flooding problems. This study focuses on improving the properties of pavement materials to enhance their strength and water permeability. This can be achieved through physical structural modifications or the addition of chemical agents such as HPMC, which increases void spaces to facilitate water absorption and drainage according to required standards. The utilization of waste slag not only helps reduce production costs and improve material performance but also minimizes environmental impacts and promotes the sustainable use of resources.
คณะเทคโนโลยีการเกษตร
The design of a 50-rai public park in the Lat Krabang district of Bangkok aims to provide a recreational space for urban residents in Lat Krabang and nearby areas. The focus is on user groups such as students, university students, and working individuals, incorporating the concept of Universal Design to ensure that everyone in society can use the space equally. However, there is still an emphasis on creating active recreational areas to meet the sports and exercise needs of students, university students, and working individuals. The design of the Lat Krabang area, which is a low-lying region resembling a basin, includes features for water retention, water management, and water treatment for use within the park. The area will focus on exercise, sports, running, walking, relaxation, and educational garden spaces.
คณะแพทยศาสตร์
This study explores the application of deep convolutional neural networks (CNNs) for accurate pill identification, addressing the limitations of traditional human-based methods. Using a dataset of 1,250 images across 10 household remedy drugs, various CNN architectures, including YOLO models, were tested under different conditions. Results showed that natural lighting was optimal for imprinted pills, while a lightbox improved detection for plain pills. The YOLOv5-tiny model demonstrated the best detection accuracy, and efficientNet_b0 achieved the highest classification performance. While the model showed strong results, its generalization is limited by sample size and drug variability. Nonetheless, this approach holds promise for enhancing medication safety and reducing errors in outpatient care.