Master of Science (MS), Ohio University, 2024, Computer Science (Engineering and Technology)

Larger, commercial, towered airports are highly equipped to provide pilots with real-time weather related data before flying the aircraft. In the case of small airports, there is a weather data gap for the aircraft flying at a lower altitude. Accurate cloud information (cloud type and its height from the ground) is crucial data for pilots flying at low altitudes because it affects both visibility and safety. A ceilometer is a device that estimates cloud height from the ground, but it is often costly and lacks portability. This thesis proposes an innovative, cost-effective approach using computer vision and deep learning to address these limitations. One of the primary challenges for these methods is the need for extensive datasets for training and evaluation, as real-world data collection of cloud height and type is time-consuming and resource-intensive. To overcome this, we generated synthetic cloud data using a stereo camera setup with ground truth height information in a virtual environment. In this thesis, cloud information involves cloud-base height estimation and classification of the type of cloud. We proposed methods that can provide better efficiency in predicting the cloud-base height than state-of-the-art methods when applied to the real-world dataset in the future. We have incorporated synthetic data to evaluate the performance of our method. These synthetic data, created by simulating VDB clouds, enable the testing and validation of cloud detection models and calibrating height predictions. We rendered the 3D scene and created ground truth bounding box and cloud-type datasets, such as Altocumulus, Altostratus, Cirrocumulus, Cumulonimbus, Cumulus, Cirrostratus, Cirrus, Stratocumulus, and Stratus. We trained the YOLO-v8 model with the cloud detection dataset and employed unseen synthetic data to assess its robustness and accuracy. Once vetted, we generated disparity images from the stereo pairs. We trained several CNN-based regression models using this di (open full item for complete abstract)

Committee: Chad Mourning (Advisor); Zhewei Wang (Committee Member); Jundong Liu (Committee Member); Bhaven Naik (Committee Member) Subjects: Computer Science

Doctor of Philosophy, The Ohio State University, 2006, Computer and Information Science

Although many advances have been achieved within the visualization community in the last decade, many challenging problems are still open in volume rendering. In this dissertation, we mainly study three challenging topics in advanced volume rendering on shadows, flows, and high-dimensional rendering. Shadows are essential to realistic and informative scenes. In volume rendering, the shadow calculation is difficult because the light intensity is attenuated as the light traverses the volume. We investigate a new shadow algorithm that properly determines the light attenuation and generates more accurate volumetric shadows with low storage requirements by using 2D shadow buffers. We have extended our shadow algorithm to deal with extended light sources and generate volumetric soft shadows with an analytic method and using a convolution technique. This shadow and soft shadow algorithm also has been applied to mixed scenes of volumetric and polygonal objects. Multiple light scattering is also modeled in our volumetric lighting model. Interval volume algorithm is a region-of-interest extraction algorithm for steady and time-varying three-dimensional structured and unstructured grids. We present several new rendering operations to provide effective visualizations of the 3D scalar field. This technique has been extended to four dimensions to extract time-varying interval volumes. The time-varying interval volumes are rendered directly, from 4-simplices to image space. We propose a high-dimensional rendering algorithm and solve this technical challenge. In this way, we can visualize the integrated interval volumes across time steps and see how interval volumes change over time in a single view. Three-dimensional flow visualization is a challenging topic. We propose an implicit flow field method to visualize 3D flow fields. An implicit flow field is first extracted using an advection operator on the flow, with a set of flow-related attributes stored. Two techniques are then em (open full item for complete abstract)

Committee: Roger Crawfis (Advisor) Subjects: Computer Science