Engineering topics
Campbell, James F.
Publications and source records attributed to Campbell, James F..
Identifying common stored product insects using automated deep learning methods
Monitoring stored product insect pests is a common practice for post-harvest management of stored grain and grain-based commodities, which helps ensure product quality from harvest to final consumer. Current methods of sampling and monitoring can be time-consuming, labor-intensive, expensive and require expertise in insect identification. Therefore, this study aims to develop an image-based automated identification system for common stored product insect species using deep-learning methods. Top-down images of the common stored product adult insect species of Rhyzopertha dominica, Cryptolestes ferrugineus, Tribolium castaneum, Sitophilus oryzae, and Oryzaephilus surinamensis were acquired and analyzed. Deep learning-based, state-of-the-art Convolutional Neural Networks (CNN) models (ResNet-50, MobileNet-v2, DarkNet-53, and EfficientNet-b0) were fine-tuned with a transfer learning approach to classify the insect species. All models were able to correctly identify the insect species with at least 96% accuracy and with few misclassifications. One issue with trained CNNs is that they do not explain the reasoning for the classification and are often called a “black box”. Therefore, visualization methods called Gradient-weighted Class Activation Mapping (Grad-CAM) were implemented to explore the black box network. The Grad-CAM uses heat maps to highlight the major image features that the network focused on to make insect species predictions. The Grad-CAM verifies the network's prediction and also helps improve network performance. This study contributes to the overall goal of developing a camera-based system for monitoring stored grain insects. As a result, the developed system would empower warehouse, flour mills, and other food facilities with a tool to quickly and accurately identify insect species in stored product environments and could be implemented as part of a close to real-time monitoring system.
Patterns in the sky: Natural visualization of aircraft flow fields
The objective of the current publication is to present the collection of flight photographs to illustrate the types of flow patterns that were visualized and to present qualitative correlations with computational and wind tunnel results. Initially in section 2, the condensation process is discussed, including a review of relative humidity, vapor pressure, and factors which determine the presence of visible condensate. Next, outputs from computer code calculations are postprocessed by using water-vapor relationships to determine if computed values of relative humidity in the local flow field correlate with the qualitative features of the in-flight condensation patterns. The photographs are then presented in section 3 by flow type and subsequently in section 4 by aircraft type to demonstrate the variety of condensed flow fields that was visualized for a wide range of aircraft and flight maneuvers.
Interactive three-dimensional boundary-layer method for transonic flow over swept wings
Explore the source record for details and available documents.
Observation of airplane flowfields by natural condensation effects
Explore the source record for details and available documents.
An interactive three-dimensional boundary-layer method for transonic flow over swept wings
A three-dimensional laminar/turbulent boundary-layer method is developed for transonic flow over swept wings. The governing equations and curvature terms are written for a nonorthogonal body-oriented curvilinear coordinate system. The viscous method is coupled to the full-potential inviscid code, FLO-30, through a displacement-surface interaction. Typically, for transonic Mach numbers and moderate sweep angles, between 5 and 12 viscous updates are required for convergence. The method is applied to a variety of wing planforms, and the results are also compared to those obtained with the three-dimensional integral inviscid-viscous interaction method TAWFIVE (Streett and Melson, 1983).
Observation of airplane flow fields by natural condensation effects
In-flight condensation patterns can illustrate a variety of airplane flow fields, such as attached and separated flows, vortex flows, and expansion and shock waves. These patterns are a unique source of flow visualization that has not been utilized previously. Condensation patterns at full-scale Reynolds number can provide useful information for researchers experimenting in subscale tunnels. It is also shown that computed values of relative humidity in the local flow field provide an inexpensive way to analyze the qualitative features of the condensation pattern, although a more complete theoretical modeling is necessary to obtain details of the condensation process. Furthermore, the analysis revealed that relative humidity is more sensitive to changes in local static temperature than to changes in pressure.
Vortical flow management techniques
The aerodynamic performance and controllability of advanced, highly maneuverable supersonic aircraft can be enhanced by means of 'vortex management', which refers to the purposeful manipulation and reordering of stable and concentrated vortical structures due to flow separations from highly swept leading edges and slender forebodies at moderate-to-high angles-of-attack. Attention is presently given to a variety of results obtained in the course of experiments on generic research models at NASA Langley, clarifying their underlying aerodynamics and evaluating their performance-improvement potential. The vortex-management concepts discussed encompass aerodynamic compartmentation of highly swept leading edges, vortex lift augmentation and modulation, and forebody vortex manipulation.
Supersonic Aerodynamic Characteristics and Shock Standoff Distances for Large-Angle Cones with and Without Cylindrical Afterbodies
Supersonic aerodynamic characteristics and shock standoff distances for large angle cones with and without cylindrical afterbodies