Stander Symposium Projects

Follow

• 

  Comparative Study of Region Localization Methods with Image Enhancement for Computer Vision

  Quinn Graehling

  Region localization is one of the main tasks within computer vision and pattern recognition. Early forms of region localization relied on basic pixel intensity thresholding while later versions used machine learning methods to locate and segment objects of interest within an image. Today the region localization fields are dominated by adaptive progressive thresholding methods, region growing segmentation and neural networks designed for semantic segmentation. With the creation of new image enhancement methods, such as the Retinex method, and with the increase in demand for quick image segmentation for use in artificial autonomy, the need for methods that can quickly and accurately segment images has grown exponentially. This presentation aims to analyze modern image segmentation methods and determine which method performs the quickest and with the highest accuracy. This presentation will also look at the difference in results between segmentation of raw images and segmentation of images with contrast enhancement via Retinex image enhancement.

• 

  Comparing Magnetron Sputtered ScN Films Grown on Sapphire (1 0 -1 0) and (1 -1 0 2) Substrates

  Tobin C. Muratore

  ScN films with high quality crystal structure and desirable carrier concentration have previously been grown on sapphire (0001) substrates. This study seeks to determine whether similarly high quality films can be grown on (101-0) and (1-102) orientated sapphire. The depositions for this study were carried out via reactive, unbalanced, DC magnetron sputtering. Previous growth of ScN on sapphire (0001) showed film quality is sensitive to sputtering conditions. This study seeks to determine what, if any, sputtering conditions can produce films of comparable quality to those grown on sapphire (0001) substrates on the lower symmetry surfaces. The impact of these growth conditions on the crystal quality and electrical properties were evaluated using x-ray diffraction and Hall-effect measurements. X-ray diffraction results indicate that growth on sapphire (1-102) is sensitive to temperature, with optimal growth occurring in a 40°C window. For (10-10) sapphire, similar crystal quality occurs over temperatures from 500-900°C. XRD also shows no conditions tested for either substrate displayed the single crystal growth present on (0001) sapphire.*This research was supported by the Air Force Office of Scientific Research through prject FA9550-RYCOR490

• 

  Comparing Patterns of Insect Communities in Remnant, Old Constructed and New Constructed Prairies

  Sienna N. Mcpeek, Emily L. Ott

  The purpose of our study was to determine if insect communities differ between remnant and new/old constructed prairies in the Greater Miami Valley. We hypothesize that insect communities within may vary in constructed prairies compared to remnants, and older constructed prairies may have similar insect patterns when compared to remnant prairies. We predict that certain insects may have similar abundance in remnant prairies and older constructed prairies will have greater diversity across orders in comparison to newer constructed prairies. We took sweep net samples each month in June, July, and August, doing four sets of 25 sweeps per prairie. We placed the caught insects in Ziploc bags to later be frozen. In total, there were 156 bags sorted through at the end of the summer. Samples were sorted to order using a dissecting microscope, and numbers of individuals in each order were counted. This research illustrates the contrast of the different types of prairies through their varying insect abundance and diversity.

• 

  Comparison of Machine Learning Frameworks for Mobile Devices

  Vamsi Charan Adari

  The purpose of this project is to compare different deep learning frameworks used to detect, recognizeobjects and images using mobile devices. In particular, mobile recognizing frameworks recognize objectsbased on features extracted, color patterns and object segments. When a similar object from traineddataset is identified matching to the recognized object, framework checks the accuracy rate among allthe other related objects and translated it as a command and shows the name of approximate detectedobject. First, the camera of the mobile device captures image of the object and send the image toframework to extract features, segment the object based on shape, color and size. Second, the relatedapproximate object from trained dataset is identified and sent as a return command to the user mobile.Third, the approximate name of the object is displayed on the user’s mobile phone helping the user torecognize. To demonstrate the accuracy rate and functioning of the mobile frameworks, we developedthree mobile applications to demonstrate the effectiveness of the new of mobile interaction. All thethree applications take image as an input and shows an approximate output on the mobile to the user.This comparison in mobile frameworks will facilitate the better usability of different mobile recognitionframeworks in mobile devices to recognize better and unknown objects moving forward.

• 

  Comparison of the Environmental Impact in Production of Lithium-Ion and Lead-Acid Batteries

  Toni Josipovic, Christopher Hartnagel, Gavin Swink

  Batteries are humanity's tools for personal exergy storage in our device enabled world. Exergy allows work to be done. We gather exergy from energy-carrying substances in the natural world. The two most popular types of natural resources for battery cell chemistries are lithium and lead metal. While energy is conserved, the exergetic portion can be destroyed when it undergoes energy conversion. Although both store exergy, each has unique drawbacks and advantages. The chemistry chosen has environmental consequences, knowing the full range of impacts may assist in efforts to decouple energy use from environmental damage. Properties that differ include raw material input, manufacturing technique and performance characteristics. Raw materials such as lithium cobalt oxide and lithium titanate are expensive, rare earth metals that are used in the production of lithium-ion batteries. Lead-acid batteries are typically made up of lead, barium sulfate, lead sulfate, and sulfuric acid. Though less expensive than lithium-ion, the efficiency compared to its counterpart is much lower. Also, the lead-based ionic compound byproduct is very toxic. To quantify our comparison, an economic input-output hybrid life cycle assessment (EIO-LCA) will be performed. Economic activity, conventional air pollutants, greenhouse gas emissions, and energy consumption for the manufacturing stage of both types of battery cell chemistries will be discussed. Knowledge of this EIO-LCA will inform the public on how battery selection is coupled with environmental damage. The production and use of different battery types would fall under UN Sustainable Development Goals relating to industry, innovation, and infrastructure and responsible consumption and production.

• 

  Constitutive and Predictive Modeling of cDLP Additively Manufactured Hyperelastic Polymers for Soft Robotics

  Kevin Lawson

  Recent developments in polymer chemistries along with timely advancements in the field of additive manufacturing (AM) have expanded the possibilities for soft material application. Specifically, novel UV-curable, ultra-stretchable (hyperelastic) liquid resins have been generated for compatibility with continuous digital light processing (cDLP) AM, a subset of vat photopolymerization (VP). However, the use of these materials in relevant applications is impeded by a lack of thorough mechanical testing and subsequent material modeling to clarify behavior. This project aims to address this issue through (1) building a framework of multimodal experimental test data, (2) fashioning descriptive constitutive (material) models, and (3) scripting representative finite element simulations, all towards the implementation of hyperelastic materials in soft robotics.

• 

  Content Randomization for Early Math Education

  Thomas Sherk, Manasi Jayesh Thakkar

  This is an interactive problem-solving counting game built using Unity 3D. It has scattered arrangements of animals that are harder to count which helps kids to develop different mental images of quantities. It also helps them to learn about number sequences as they select the side which has more number of animals. There are 4 levels in this game and difficulty keeps increasing at every level. The number of animals ranges from 1 to 5 in level 1, level 2 and level 3. In level 1, the animals on each side are of the same type and same size. Example, 5 lions on one side and 3 monkeys on the other side. Level 2 is similar to level 1 except for the fact that the area occupied by the animals on one side is the same as the area occupied by the animals on the other side. For example, 2 elephants on one side and 4 lions on the other side occupy the same area. In level 3, animals on each side are of different type and the area occupied by the animals on both sides is the same. For example, 2 elephants and 1 lion on one side and 1 elephant, 1 giraffe and 2 monkeys on the other side. Level 4 is similar to level 3 but now the number of animals varies from 1 to 9. The animals are always displayed in a random manner, making sure that the same pattern never repeats.

• 

  Co-Op Dayton- Creating an Economy that Works for All.

  Connor M. Dushane

  Co-Op Dayton is a non-profit founded in 2015 that incubates cooperative businesses to create and retain quality jobs in our region, building a local economy that works for all. The cooperative business model puts the power of decision making in the hands of the employees themselves giving them greater control over the future of their business. Co-Op Dayton is best known for its flagship project the Gem City Market. Over the past year, I have been working with them to develop new cooperative businesses in the manufacturing, construction, and agricultural industries. Figuring out how Co-Op Dayton’s work impacts the economic growth of the city of Dayton, and comparing it to the very successful Mondragon Corporation in Spain, creates an opportunity to see how investing resources in worker or community-owned companies could spur further economic growth in Dayton and beyond. This poster will dive into how cooperatives can grow businesses by analyzing different Dayton small businesses that either are or are in the process of transitioning to the cooperative model and how their business can grow with the benefits of the coop model, as well as analyzing the current economical climate of the city to understand how shifting to worker-owned could spur regrowth of certain industries. The goal of this poster is to provide a greater understanding of how cooperatives can change the economy for the better and meet the goals of positive social change while working to overcome some of today’s economic challenges.

• 

  Crosstalk between IMD and JNK pathways in Amyloid-beta 42 mediated neurodegeneration in Alzheimer’s Disease

  Prajakta D. Deshpande, Samuel J. Shroder

  Alzheimer's Disease (AD), affects 5.8 million people, is an incurable neurodegenerative disorder that leads to memory loss and reduced cognitive abilities. In AD, the miscleavage of the amyloid precursor protein (APP) results in the accumulation of Amyloid-beta 42 plaques outside neurons that triggers neuronal cell death. The mechanism for cell death is widely unknown and is currently being explored. In this study, Drosophila was used as a model system to mimic AD neuropathology by misexpressing human Aβ42 plaques in the developing eye of the Drosophila using the Gal4/UAS system. Misexpressing Aβ42 plaques triggers neuronal cell death and results in a glazed and reduced eye phenotype. It is known the immune deficiency (IMD) pathway plays a role in cell survival and the c-Jun N-terminal kinase (JNK) plays a role in cell death. Induction of the IMD pathway leads to the activation of NF-kB transcription factor Relish. Gain of function of Relish in the background of GMR> Aβ42, results in suppression of Aβ42 mediated neurodegeneration. Down regulation of Relish results in the smaller eye phenotype with necrotic spots. Activation of Relish is linked to proteasomal degradation of TGF-beta activated kinase 1 (TAK1), upstream component for JNK activation. So, it would be interesting to understand the crosstalk between Immune deficiency (IMD) pathway, and the c-Jun N-terminal kinase (JNK) pathway in Aβ42 mediated neurodegeneration.

• 

  DALES: A Large-scale Aerial LiDAR Data Set for Semantic Segmentation

  Nina Marie Varney

  We present DALES, a new large scale aerial LiDAR data set with over a half-billion hand-labeled points, spanning 10 $km^2$ of area and eight object categories. Large annotated point cloud data sets have become the standard for evaluating deep learning methods. However, most focus on data collected from a mobile or terrestrial scanner with few focusing on aerial data. Point cloud data collected from an Aerial Laser Scanner (ALS) presents a new set of challenges and applications in areas such as 3D urban modeling and large scale surveillance. DALES is the largest publicly available ALS data set with over 400 times the number of points and six times the per meter resolution than other currently available annotated aerial point cloud data sets. We describe the nature of our data and the annotation workflow as well as provide a benchmark of current state-of-the-art algorithms and their performance on our data set. This data set gives a critical number of expert verified hand-labeled points for the evaluation of new 3D deep learning algorithms, helping to expand the focus of current algorithms to aerial data.

• 

  Dayton and Sustainable Development Goal 16

  Jared Dominic Marsh

  During my time researching human rights cities and how to establish Dayton as a human rights city, I learned the importance of collaboration and practicality. This is evident in the SDGs with the broad goals but paired with practical targets. When you take the practical targets and work with local partners to align the targets with their needs I think you have an improved chance of being successful. The people and planet are the most important factors, in my opinion, so we must start using language and mechanisms that are tangible for all. I have learned that without institutional and systemic change these unjust systems are unlikely to change. This is why we need a top down approach. Without the will of local individuals initiatives are likely to fail. Bottom up approaches allow initiatives to be locally relevant and effective.

• 

  Dayton as a Human Rights City in the Future

  Maranda Elise Lewis

  Ideas for Advocacy Campaign: Get a motivated group consisting of victims (if willing) and non-victims; educate the community through group presentations; band together to report, be on the lookout, and assist with resources for victims; create a non-profit aimed at reducing, supporting, and dispensing resources for those in need.

• 

  Dayton & Sustainable Development Goal 4: Turning the City of Learners into a Human Rights City

  Tiffany Erin Hendricks

  Lessons: How to apply international protocols, treaties, and issues to a local environment/context; current human rights issues in Dayton and their impact on Daytonians; compilation of data across a variety of platforms and with a wide variety of contributors.

  Reflections: Human rights can only be accomplished truly when the entire community, from every level, is on board and involved. Every voice needs to be heard and every person given an invitation to and seat at the table.

• 

  Deep Learning and Natural Language Processing to Detect Misinformation

  Ian M. Cannon, David Fan

  Misinformation is the act of spreading incorrect or false information about a given topic. With the upcoming presidential campaign, COVID-19, and other major events ongoing it is especially important to identify sources of misinformation. Our group proposes a novel method of classifying articles by using context and content indicators to debunk news articles containing misinformation. Most models take in articles and determine whether it is fake or not. We propose a method to predict content indicators to that highlight the credibility of an article and then decide if it is real or fake. This gives some agency to the reviewer by allowing one to see what were the major factors in deciding if an article is fake or not.

• 

  Designing Fictional Spaces: Questionable Architecture that Supports Sustainable Design

  Noël J. Michel

  This thesis presents the modeling of spaces described in short stories that are difficult to visualize. The three stories are Kafka's "The Burrow", Borges's "The Library of Babel" and Barthelme's "The Balloon." Three dimensional models were created based on the details provided by the authors in each story. Several 2-D images are then generated from these models to match specific scenes. This consideration of these works of fiction provokes the asking of several questions about the science, mathematics and engineering that underpins the stories. In all cases, questions about sustainability arise.

• 

  Design Modeling of Spatial Shape-Change Linkages

  Chengwei Shi

  The goal of this research is to develop mechanical designs of spatial shape-changing linkages. Mechanical systems often benefit from the capacity to vary between specific shapes in a controlled manner, such as a morphing aircraft wing that can adapt to different in-flight requirements. Spatial shape-changing linkages consist of a chain of three-dimensional bodies connected with ball joints. When the chain segments are repositioned, they match a set of arbitrary spatial curves. These chains are composed of two segments types: a twisted rigid segment and a helical segment with constant curvature and torsion but varying length. The research project involves creating the mechanical designs of the segments and motion control schemes that move the chain from the origin position to the target position. Animations are created in SolidWorks that demonstrate various motion schemes and illustrate the chain’s approximation to the target spatial curves.

• 

  Design of a Trike for Paraplegics Powered By Functional Electrical Stimulation of Leg Muscles

  Andy Bazler, Nicholas Andrew Lanese

  The goal of this project is to design a performance tricycle for paraplegics whose leg muscles are stimulated to pedal via Functional Electrical Stimulation (FES). FES stimulates muscle contraction with small electrical currents and has proven useful in building muscle in patients while relieving soreness and promoting cardiovascular health. An FES-stimulated cyclist produces approximately 25 Watts of power, nearly 20 times less than a typical rider. At these reduced power levels, the challenges of pedaling are amplified. For example, as the pedal follows the traditional circular path, there are portions referred to as dead zones, where neither FES-stimulated leg actively propels the bike forward. One possibility for reducing or eliminating dead zones is to redesign the circular path of the pedaling motion. Bicycles have recently been marketed that feature pedaling mechanisms that employ alternate pedaling motions. In addition to addressing dead zones, these bikes also optimize the muscle capacity of the rider to deliver torque to the wheels. These new bikes achieve alternate pedaling paths through the introduction of more complicated mechanisms including four-bar and ratchet-and-pawl linkages. Such alternates are being considered for the redesign of the performance tricycle piloted by FES-stimulated riders. To investigate possible changes to the tricycle, quasi-static models have been developed for traditional and alternate cycling mechanisms. This allows for a comparison of torque generation between the mechanisms which facilitates selecting the optimal design. Such a tricycle is viewed as beneficial due to the health advantages, improved mobility, and independence created for the end user.

• 

  Destructive Interparental Conflict: The Moderating Effect of Awareness of Emotions on Parent and Partner Attachment

  Amanda R. Barry, Fiona O'Malley, Jasmine Marsh

  Childhood exposure to destructive interparental conflict (DIC) may have negative effects on adult romantic relationships, in part via impairment in the parent-child bond. Good emotion regulation, including emotional awareness, usually fosters wellbeing, but recent studies suggest that, following DIC, nonawareness (NA) may be partially protective. The present study (N= 807 undergraduate students) tested maternal and paternal attachment anxiety (MAA and PAA) as mediators of the link between DIC and adult romantic attachment anxiety (ARAA) with NA as a moderator. Higher DIC frequency was linked with ARAA via higher MAA at low (CI = 1.37, .62) and moderate (CI = .92, .36) NA and via PAA at low (CI = 1.52, .74) and moderate (CI = 1.01, .48) NA; at high NA paths were not significant. NA also moderated the DIC intensity/ARAA link via MAA (CI = .01, .12). While significant at all levels of NA, this path was stronger at low (b = 1.12) versus high levels (b = .51). Exposure to DIC can negatively impact ARAA via impaired child-parent attachment; this relation may be stronger for children more aware of their emotions, as increased awareness may underlie a sensitization to DIC and its impact. Clinically, it may be helpful to aid children with high emotional awareness to manage emotional reactivity, and as young adults to process the implications of their parents’ DIC for their own lives.

• 

  Developing an Analytical Solution to expand Quantum Computation

  Ryan Massie

  Quantum Computation is the study of the information processing tasks that can be accomplished using quantum mechanical systems. It could open a whole new world of quicker processing and computations and offer more secure networks. By constructing systems that are able to use spontaneous parametric down conversions to reach high levels of entanglement, a property of quantum mechanics where particles within a system cannot be separated, we are able to construct these quantum mechanical systems that can then be used for quantum computation. Our work began by being able to recreate the quantum states of past work analytically by reproducing calculations. Then using that to construct our own versions that through seeding, no external stimulation, can produce the desired entangled results.

• 

  Developing, Implementing and Evaluating a Nutrition Intervention for Adolescents at Wright Brothers Middle School.

  Kathleen Curran, Madalynn Eads, Caroline Grannan, Jessica Moore, Julia Muccio, Addy Nichols, Dana Roscoe, Abby Sibley, Haley Stewart, Taylor Stewart, Hannah Waters

  In the US, overall dietary habits of adolescents promote obesity and chronic disease. For example, data from national sources indicated that less than 10% of adolescents consume the recommended servings of fruits and vegetables per day with the majority exceeding the recommendations for saturated fat, sugar and calories. In the adolescent population, various individual, social, environmental and political factors influence dietary behavior. Therefore, the purpose of this project was to examine factors contributing to dietary behavior in a small cohort of adolescents attending an after-school program at Wright Brothers Middle School, a Dayton Public School. Guided by the Social-Ecological Model, University of Dayton students enrolled in Community Nutrition, completed a community needs assessment to identify dietary-related issues and contributing factors in middle school students. The information gained from the needs assessment was used to inform the development and implementation of a culturally-appropriate intervention to address identified nutrition issues in the target adolescent population. The intervention was implemented during the after-school program at Wright Brothers Middle School. Using various data collection instruments and methods, the intervention process and impact was evaluated. In the future, the evaluation will be used to revise the intervention, if needed. The intervention protocol and materials will be provided to the coordinators and teachers of the after-school program for future use.

• 

  Development of Safety Performance Functions for Two-Lane Rural Highways in the State of Ohio

  Abdulrahman Faden

  The Highway Safety Manual (HSM), which is the guidance document for state departments of transportation (DOTs), was published in 2010 and one of its sections, called Part C of HSM, it involves the development of crash prediction methods. The goal of the predictive method is to develop and calibrate safety performance functions (SPFs). SPFs are mostly regression models that correlate the expected number of crashes quantitatively with traffic exposure and geometric characteristics of the road. However, HSM's default prediction models are not suitable for all states or jurisdictions because each state and jurisdiction have different characteristics, such as terrain, driver behaviors, weather conditions, etc. Hence, the principal objective of this study is to develop a prediction method for producing Ohio-specific SPF models to use for rural two-lane highways in the state of Ohio. This study aims to create SPFs or jurisdiction-specific SPFs for two-lane rural highway segments as the first study for this type of roadway facility in the state of Ohio. Almost 28,700 miles of highway geometric data were obtained from the Highway Safety Information System (HSIS) to create these new models using negative binomial regression. The most critical variables to be used for analyzing and creating the best models for the state of Ohio are average annual daily traffic (AADT), segment length, lane width, shoulder width, posted speed limit, presence of curves and grades.

• 

  Development Work: FACT Zambia

  Isabel Gerardino, Casey Coyne Willson

  The evaluation of global aid programs is crucial to understand the meaning behind their work in the development of communities to conclude the improvements that can be done as well as the potential inefficient results that can be avoided. The Republic of Zambia, a country rich in natural resources, is also rich in attracting international NGOs that are interested in combating challenges concerning the lack of governmental transparency and advocacy for collaboration with the citizens. One of the organizations that proposes some strategies to create and sustain civil society’s collaborations with policy-makers and citizens is Counterpart International. Counterpart International has developed a program called Fostering Accountability and Transparency in Zambia (FACT Zambia) to address the lack of capacity and transparency among Civil Service Organizations (CSOs). Counterpart is doing this to aid in efforts that already exist in Zambia in order to facilitate more participatory involvement in development.

• 

  Dimensioning Mechanical Press Architectures for Improved Dwell using Advanced Algebraic Techniques

  Tianze Xu

  A mechanical press is a machine that shapes parts by driving a ram into metal and deforming the material into a desirable shape. As this is an incredibly common process for forming metal parts, from pop cans to car fenders, presses see significant use in industry. This research project seeks to develop a numerical algebraic method for determining mechanical press dimensions from a desired dwell displacement pattern. This dwell pattern occurs when the ram lingers near the bottom of the stroke while the rest of the press stays in motion. Longer dwell produces improved part forming at no additional cost. This study focuses on knuckle presses architectures to test the proposed method on a variety of systems and to produce the most feasible design. Numerical algebraic methods are particularly relevant here due to their capacity to accurately describe mechanical press architectures while allowing solutions via current numerical methods that guarantee the determination of all solutions to a set of algebraic equations. As such, there are a significant number of companies designing and building mechanical presses to meet a variety of end used needs. A particularly common need is dwell, the capacity of the press to hold the position on one of its parts while the rest of the machine stays in motion. Dimensioning a new architecture for a mechanical press that produces significantly improved dwell allows for manufacturing parts at a higher rate with lower operating costs.

• 

  Dimensioning Mechanical Presses Driven by a Geared Five-Bar for Desired Dwell using Advanced Algebraic Techniques

  Xingyu Zhu

  A mechanical press uses a linkage that oscillates a ram in order to form or cut sheet metal. This research develops design theories that use a unique mechanical linkage to obtain alternative ram oscillation patterns, such as a prolonged dwell. A geared five-bar press with sliding output is proposed to produce these alternative motions. In one alternative motion, an extended dwell involves a ram that remains near the bottom of the stroke while the crank continues to rotate. A prolonged dwell is ideal for coining operations. Non-linear loop closure equations are generated using isotropic coordinates. After specifying a desired motion pattern, an algorithm that uses the closure equations with numerical algebraic geometry obtains all possible sets of appropriate dimensions for the links. Lastly, a process to determine the best possible set is formulated.

• 

  DIMLabyrinths: Printable 3D Cube Mazes Designed in MATLAB

  Adam Steven Wicks

  DIMLabyrinths are 3-dimensional marble mazes designed for 3D printing. The maze body is acube with an evenly-spaced grid of round holes from top-to-bottom, left-to-right, and front-to-back. The holes are one of two sizes inside the cube, either too small for the marble to passthrough or just big enough to allow the marble passage. As such, the solver can see the marble atall times as it moves through the maze embedded in the cube. The design of the maze itself isgenerated using an algorithm developed in MATLAB. The maze is guaranteed to visit everylocation in the cube on a path that connects the top-front-left corner to the bottom-back-rightcorner. This unique geometry is well-suited for manufacturing via 3D printing. DIMLabyrinthfiles suitable for rapid prototyping are available for free download on the DIMLab My MiniFactory site. The result is a unique puzzle, partially designed in MATLAB, that can be 3Dprinted at home for free.