Search results

Increasing reliance on autonomous systems requires confidence in the accuracies produced from computer vision classification algorithms. Computer vision (CV) for video classification provides phenomenal abilities, but it often suffers from “flickering” of results. Flickering occurs when the CV algorithm switches between declared classes over successive frames. Such behavior causes a loss of trust and confidence in their operations.

This “flickering” behavior often results from CV algorithms treating successive observations as independent, which ignores the dependence inherent in most videos. Bayesian neural networks are a potential remedy to this issue using Bayesian priors. This research compares a traditional video classification neural network to its Bayesian equivalent based on performance and capabilities. Additionally, this work introduces the concept of smoothing to reduce the opportunities for “flickering.”

The augmentation of Bayesian layers to CNNs matched with an exponentially decaying weighted average for classifications demonstrates promising benefits in reducing flickering. In the best case the proposed Bayesian CNN model reduces flickering by 67% while maintaining both overall accuracy and class level accuracy.

The training of the Bayesian CNN is more computationally demanding and the requirement to classify frames multiple times reduces resulting framerate. However, for some high surety mission applications this is a tradeoff the decision analyst may be willing to make.

Our research expands on previous efforts by first using a variable number of frames to produce the moving average as well as by using an exponentially decaying moving average in conjunction with Bayesian augmentation.

The growth of lending from nontraditional lenders may pose challenges for official US Department of Agriculture (USDA) farm sector debt estimates, but it is difficult to find data to assess official estimates. The purpose of this study is to examine whether debt provided by nontraditional lenders is accurately accounted for in official estimates.

We compare traditional and nontraditional lending data from farm equipment lien collateral values and the USDA Agricultural Resource Management Survey (ARMS). After analyzing trends in equipment lending implied by farm equipment lien data and ARMS, we estimate whether changes in farm equipment lien values predict changes in equipment debt reported in ARMS and whether lender type influences that relationship.

We find that credit provided by nontraditional lenders is likely underreported in ARMS. Our econometric model shows that equipment debt volumes for nontraditional lenders are consistently lower than traditional loan volumes in ARMS across a variety of model specifications. We also find that an increase in lien values for nontraditional lenders is less likely to predict an increase in ARMS equipment debt volumes than an increase for traditional lenders.

Official farm sector debt estimates may not fully account for nontraditional lenders.

This study demonstrates how the growth of nontraditional lending poses challenges for estimating US farm sector debt. We evaluate farm sector debt estimates and advance knowledge of the role of nontraditional lenders in farm equipment credit provision. The farm equipment lien dataset provides a rich source of novel data for research on local and national equipment debt and investment.

Agricultural banks likely respond differently to economic downturns compared to nonagricultural banks. Limited previous research has examined the performance of agricultural banks under economic crisis and in the presence of banking regulations. This study aims to explore agricultural banks' responses to economic and regulation shocks relative to nonagricultural banks.

This study uses bank-quarter level data from 2002 to 2022 for virtually all commercial banks in the U.S. In this research, the Z-score measures the bank’s default risk, the return on assets measures bank profitability and changes in amount of farm loans indicate the wider impact on the agricultural sector. Effects of the financial crisis, Basel III reforms to banking regulation and the coronavirus (COVID-19) pandemic on these banking measures are assessed using distinct empirical frameworks. The empirical estimations use various subsamples based on bank types, bank sizes and time periods.

Economic downturns are associated with fluctuations in returns and the risk of default of commercial banks. Agricultural banks appeared to be more resilient to economic downturns than nonagricultural banks. However, Basel III regulated agricultural banks were more likely to fail amidst the pandemic-related economic shocks than the regulated non-agricultural banks.

This study examines the resiliency of agricultural banks during economic downturns and under postfinancial crisis regulation. This is one of the first empirical works to analyze the effectiveness of Basel III regulation across bank types and sizes considering the COVID-19 pandemic. The key finding suggests that banking regulation should consider not only size heterogeneity but also the heterogeneity in lending portfolios.

This paper explores the relationship between droughts and U.S. agricultural sector profitability in select U.S. regions. We also examine the extent to which the farm safety net, including direct government payments and the federal crop insurance program, may compensate for the impacts of drought on farm sector profitability.

Fixed effect regressions estimate the relationship between drought and profitability for the farm sector. The analysis uses a panel of annual, state-level net farm income and its subcomponents (from the USDA, Economic Research Service) and state-level annual drought measures constructed from county crop damage days (as reported in the Spatial Hazard Events and Losses Database for the United States, or SHELDUS). SHELDUS includes Severe (D2) or more intense drought shocks for the states east of the Rocky Mountains, and these states provide the focus for our analysis.

Market net farm income negatively correlates with the drought measures. This relation is partially driven by the increase in production expenses during drought episodes. Further, sales from inventories tend to increase during drought periods. A significant share of damages to sector performance are offset by federal crop insurance program indemnities. Finally, our results show that drought impacts and the effects of the farm safety net are distributed differently across geographic regions.

To our knowledge, this study is the first attempt to examine drought impacts on the components of agricultural profitability, including the farm safety net and production expenses, at the sector level.

The purpose of this paper is to examine the effect of falling commodity prices on farm debt usage of corn and soybean farms, and how this debt usage differs based on the financial leverage of the farm.

Using panel data on farms surveyed at least twice in the Agricultural Resource Management Survey (ARMS) from 1996 to 2015, this paper uses a difference-in-differences approach to measure the effect of low commodity price shocks on financially vulnerable farms. To account for the correlation in the error structure between the three dependent variables (real estate debt, non-real estate debt, and interest payments) we use a seemingly unrelated regression approach.

Following a commodity price shock, financially vulnerable farms (debt-to-asset ratio greater than 40 percent) were found to increase their non-real estate debt when compared with non-financially vulnerable farms. Off-farm business income was found to help farms reduce real estate debt and interest payments in the face of these shocks.

Data consist of corn and soybean farms surveyed more than once in the ARMS from 1996 to 2015 and are not representative of all US farms, but have similar characteristics to US commercial farms.

The results indicate that financially vulnerable commercial crop farms respond to lower prices by taking on non-real estate debt, increasing financial stress. Well-targeted federal programs could prevent further financial stress for this group.

This is the first paper to use unbalanced panel data from ARMS to examine how farm debt use responds to commodity prices. This paper can inform policymakers about the financial risks to farms resulting from the current low-price environment.

Rates of less perceptible social–behavioral–emotional disorders thought to be based in neurobiological brain differences have burgeoned, though much of disability studies remains focused on the need to challenge compulsory able-bodiedness. This chapter examines instead diverse families living with adult sons’ and daughters’ invisible disabilities, asking how mothers may challenge compulsory able-mindedness.

This chapter is based on 15 in-depth interviews conducted in 2017 and 2018 with mothers originally interviewed between 2003 and 2008.

The accounts foreground tensions for those at the boundaries of “normality” in a culture that valorizes citizen’s independence, productivity, and heroic overcoming of any inability. Mothers of “precariously normal” adult sons and daughters invited to reflect on their earlier accounts reveal both the power of such dominant narratives and the possibilities to disrupt and challenge this public storytelling.

Findings of this study point to the alternative narratives and identities sought by disability studies and bring invisible social–behavioral–emotional disabilities into discussions that have largely centered on visible physical disabilities. These findings also underscore the complex similarities and differences in families’ experiences of disability across class and race divides, while suggesting the need for institutional change and greater, less punitive, public resources.

The present chapter reviews part of the literature that focuses on dark tourism and dark consumption. The main theories were placed under the critical lens of scrutiny. With strongholds and weaknesses, dark tourism seems to be enframed in an ‘economic-based paradigm’, which prioritises the managerial perspective over other methods. Like Dark Tourist, the Netflix documentary assessed in this chapter, this academic perspective accepts that the tourist's experience is the only valid source of information to understand the phenomenon. Rather, we hold the thesis that far from being a local trend, dark tourism evinces a morbid drive which not only emerges recently but involves other facets and spheres of society. We coin the term Thana-capitalism to denote a passage from risk society to a new stage, where the Other's death is situated as the main commodity to exchange. The risk society as it was imagined by Beck, set finally the pace to thana-capitalism. Dark Tourist proffers an interesting platform to gain further understanding of this slippery matter. In sharp contrast to Seaton, Sharpley or Stone, we argue that dark tourists are unable to create empathy with the victims. Instead, they visit these types of marginal destinations in order to re-elaborate a political attachment with their institutions. They consume the Other's pain not only to feel unique and special (a word that sounds all the time in the documentary) but also to affirm their privileged role as part of the selected peoples.

With three credited scriptwriters and five credited directors, the 1967 release of Casino Royale saw a gang of multifaceted James Bond 007s facing off against an army of beautiful, hypersexualised, personality-less female spies, headed by the real James Bond’s neurotic, insecure, American nephew Jimmy. Perhaps this wasn’t Fleming’s intended storyline for Bond’s first outing at Casino Royale, but the resulting parodic outing absorbed and commented upon some of the inherent gendered archetypes of Fleming’s work. What the 1967 Casino Royale accomplishes is a narrative which contrasts varieties of masculinity which are segmented forms of the masculinity defined by Fleming’s Bond. This chapter compares the masculinity of Bond developed in Fleming’s novel, before examining the representations of masculinity inherent within the four key male characters: Sir James Bond (David Niven), Evelyn Tremble (Peter Sellers), Cooper (Terence Cooper) and Dr Noah/Jimmy Bond (Woody Allen). By showing the depictions of masculine elements each of these characters embodies, along with the metanarrative elements of each performer’s persona, this chapter aims to identify how the 1967 Casino Royale both faithfully depicts the masculine elements of Bond while at the same time satirizing Bond’s particular brand of masculinity. This examination ultimately argues that this segmentation of Bondian masculinity is the core point of cohesion in a deeply incoherent, parodic film adaptation of Fleming’s novel.

The purpose of this paper is to present evaluation findings from a proof-of-concept virtual reality adaptive skills intervention called Virtuoso, designed for adults with autism spectrum disorders.

A user-centric usage test was conducted to investigate the acceptability, feasibility, ease-of-use and relevance of Virtuoso to the unique needs of participants, as well as the nature of participants’ user experiences. Findings are presented from the perspectives of expert testers and participant testers with autism.

This paper offers findings that suggest Virtuoso is feasible and relevant to the unique needs of the target population, and that user experience was largely positive. Anecdotal evidence of skills transfer is also discussed.

The research was conducted in limited settings and with a small number of participants. Multiple VR hardware systems were used, and some experienced instability. This could be accounted for in future research by deploying across multiple settings and with a larger number of participants. Some evidence of cybersickness was observed. Future research must carefully consider the trade-offs between VR-based training and cybersickness for this vulnerable population.

This paper reports on cutting-edge design and development in areas that are under-represented and poorly understood in the literature on virtual reality for individuals with autism.