Sungkyunkwan University · 工学
Professor Minsoo Park's research lab specializes in leveraging artificial intelligence, particularly deep learning and computer vision, to address critical environmental and societal challenges. The lab focuses on early detection and monitoring of wildfires, construction site safety through advanced object detection, and climate-related disaster prediction using big data analytics. Key research directions include developing robust AI models for small and overlapping object detection, addressing data imbalance in imbalanced datasets via generative models like CycleGAN, and creating multilabel classification systems for comprehensive disaster response. The lab emphasizes real-world applicability, integrating transfer learning, data augmentation, and explainable AI to enhance model performance and reliability in complex, dynamic environments.
Figures are computed from collected data and may differ slightly.
Although there has been study on worker detection using computer vision (CV) for the safety of construction sites, it is still challenging to identify employees who are obstructed or have poor vision. To solve these problems, we propose a method of small and overlapping target (worker) detection at a complex construction site named SOC-YOLO. The method is based on YOLOv5 and utilizes distance intersection over union (DIoU) non-maximum suppression (NMS), incorporating weighted triplet attention,
To minimize the damage caused by wildfires, a deep learning-based wildfire-detection technology that extracts features and patterns from surveillance camera images was developed. However, many studies related to wildfire-image classification based on deep learning have highlighted the problem of data imbalance between wildfire-image data and forest-image data. This data imbalance causes model performance degradation. In this study, wildfire images were generated using a cycle-consistent generati
Owing to abnormal climate phenomena worldwide, forests are becoming dry and heat waves have started to occur, increasing the damage caused by wildfires. In addition to causing significant human and material damage, wildfires are also a major cause of critical pollutant emissions, in which fine dust generated by incomplete combustion pollutes the atmosphere, soil, and water. Early detection and monitoring are some of the main ways for minimizing wildfire damage, and a topic of research interest i
Climate change increases the frequency and intensity of heatwaves, causing significant human and material losses every year. Big data, whose volumes are rapidly increasing, are expected to be used for preemptive responses. However, human cognitive abilities are limited, which can lead to ineffective decision making during disaster responses when artificial intelligence-based analysis models are not employed. Existing prediction models have limitations with regard to their validation, and most mo
In this article, we estimate the user substitutability and complementarity of media by using media diary data on the media use of individuals over the course of three days. Fixed-effects panel data models allow us to eliminate possible bias due to individual-specific media use propensity. We observe significant substitution among paper, television, and computer use, while telephone and computer use seem to be complementary in time of use. The magnitudes of substitutability and complementarity be
Given the explosive growth of information technology and the development of computer vision with convolutional neural networks, wildfire field data information systems are adopting automation and intelligence. However, some limitations remain in acquiring insights from data, such as the risk of overfitting caused by insufficient datasets. Moreover, most previous studies have only focused on detecting fires or smoke, whereas detecting persons and other objects of interest is equally crucial for w
By linking the estimated working height and deep learning multi-detection results to established safety regulations, the proposed method shows the potential to automatically monitoring unsafe behaviors in construction site.
Along with the fourth industrial revolution, the use of unmanned aerial vehicles (UAV) has grown very rapidly over the past decade. With this rapid growth, studies using UAVs are underway in various areas. UAVs are more economical and effective when utilizing several nodes rather than operating a single aircraft. In general, UAVs collect and transmit information to the control center (CC), and act on control commands from the CC. Communication between UAVs and the control center is usually achie
We formulate a model that captures the inter-dependence between hardware demand and software supply – indirect network effect – in the DVD industry. The identification of the network effect comes from the difference in software availability across two different formats: VHS and DVD. We find that a 1% increase in the number of DVD titles raises the demand for DVD players by 0.87%. Simultaneously, a 1% increase in video player ownership leads to a 0.14% increase in the variety of video titles. Our
In recent years, there has been significant interest in simple and affordable technologies that produce 3D printed embedded electronics and interconnects, in order to increase the functionality of these structures. This study describes the development of 3D printing processes for conductive patterns using 3D printing of metal particle dispersed polymer. This fabrication method is based on 3D molded interconnect device (MID) technology with laser direct structuring. The pattern on 3D printed stru
Open papers in the app to read, cite, and organize with AI.
