Mississippi State University Research Articles

Expanding Ground Vehicle Autonomy into Unstructured, Off-Road Environments: Dataset Challenges

As with the broad field of deep learning, autonomy is a research topic that has experienced a heavy explosion in attention from both the scientific and commercial industries due to its potential for the advancement of humanity in many cross-cutting disciplines. Recent advancements in computer vision-based autonomy has highlighted the potential for the realization of increasingly sophisticated autonomous ground vehicles for both commercial and non-traditional applications, such as grocery delivery. Part of the success of these technologies has been a boon in the abundance of training data that is available for training the autonomous behaviors associated with their autonomy software. These data abundance advantage is quickly diminished when an application moves from structured environments, i.e., well-defined city road networks, highways, street signage, etc., into unstructured environments, i.e., cross-country, off-road, non-traditional terrains. Herein, we aim to present insights, from a dataset perspective, into how the scientific community can begin to expand autonomy into unstructured environments, while highlighting some of the key challenges that are presented with such a dynamic and ever-changing environment. Finally, a foundation is laid for the creation of a robust off-road dataset being developed by the Engineer Research and Development Center and Mississippi State University’s Center for Advanced Vehicular Systems.

Peanut cultivar response to residual soil test potassium in North Mississippi

AbstractThe average U.S. peanut (Arachis hypogaea L.) yield has increased by approximately 25% with the adoption of peanut cultivar ‘Georgia‐06G’. Since this adoption, many new high yielding runner cultivars with similar yield potential have been released. However, current nutrient recommendations are based on soil tests that were developed prior to the release of Georgia‐06G. Particularly for potassium, current soil test potassium (STK) critical values were established on soil textures with relatively low cation exchange capacity (CEC) but were not validated on soil textures with high CEC. This study aimed to evaluate the growth and yield response of five recently released peanut cultivars to four STK levels ranging from very low to medium based on Mississippi State University Extension soil testing recommendations. The STK classification levels were also based on two soil series categorized with high CEC—Leeper (∼38.4 meq 100 g−1) and Marietta (∼15.9 meq 100 g−1) soil series. Cultivars Georgia‐06G, ‘Georgia‐16HO’, ‘Georgia‐18RU’, FloRun ‘331’, and ‘AU‐NPL‐17’ were evaluated in this study. No STK × variety interaction occurred, indicating similar K requirements across all varieties evaluated. However, a positive pod yield response occurred in both soil types when the average STK increased from 128 to 167 lbs ac−1 for all cultivars and site years. Critical STK values on both soils were greater than many current Extension recommendations, and the critical STK value of Leeper is greater than the Marietta soil series, likely due to the higher CEC value. These results demonstrate the need to adjust peanut STK sufficiency levels based on soil CEC. Further evaluation of modern peanut cultivar productivity response to STK sufficiency levels is needed for soils with moderate CEC.

Waste Separation and Perceptions of Environmental Health in a College Floral Design Course

Industries have found themselves under a microscope because consumers are basing more of their purchasing decisions on a company’s sustainability practices. The floral industry is perceived as being environmentally friendly by consumers. However, based on waste production, the life cycle of fresh merchandise, and the carbon footprint of flowers shipped across the world, this purchasing assumption is not entirely true. To align with consumer perceptions and become more sustainable, the industry must adapt to include more sustainable practices. New and experienced florists alike must determine how they can make slight changes in their businesses to become more environmentally savvy. The purpose of this study was to investigate whether the introduction of more sustainable waste practices into a college floral design course influenced the students’ perceptions of environmental health. Two groups of students enrolled in the basic floral design course at Mississippi State University completed a survey about environmental health at the start and at the end of the semester. However, one group sorted their laboratory wastes into compostable material and landfill material. The results of this study indicated that students who separated their floral waste maintained the same level of environmental concern throughout the testing period, whereas the environmental health scores of those who did not separate their waste declined slightly. The results also indicated that the participants who did not separate their floral waste had lower mean scores at the beginning and at the end of the study. Finally, the results of this study indicated that students who separated floral waste were ranked as having high environmental concern compared to those who did not separate their floral waste in the laboratory.

Evaluating E-Resources Subscriptions: Creating Cost Savings with CloudSource+

ABSTRACT The ongoing evaluation of electronic resource subscriptions is crucial. This article will provide an overview of the new process implemented at Mississippi State University Libraries for evaluating e-resource subscriptions. Quantitative and qualitative data was gathered from usage statistics and via a survey to inform data-guided decisions regarding subscription renewals. Additionally, discovery platform CloudSource+ from SirsiDynix provided analysis reports showing the amount of open access content available in each database. The author hopes to provide other librarians with a guide for making data-informed decisions on subscription renewals, to maintain the e-resources their faculty and students need for teaching and research.

Evaluating E-Resources Subscriptions: Creating Cost Savings with CloudSource+

ABSTRACT The ongoing evaluation of electronic resource subscriptions is crucial. This article will provide an overview of the new process implemented at Mississippi State University Libraries for evaluating e-resource subscriptions. Quantitative and qualitative data was gathered from usage statistics and via a survey to inform data-guided decisions regarding subscription renewals. Additionally, discovery platform CloudSource+ from SirsiDynix provided analysis reports showing the amount of open access content available in each database. The author hopes to provide other librarians with a guide for making data-informed decisions on subscription renewals, to maintain the e-resources their faculty and students need for teaching and research.

Mitigating Missing Rate and Early Cyberattack Discrimination Using Optimal Statistical Approach with Machine Learning Techniques in a Smart Grid

In the Industry 4.0 era of smart grids, the real-world problem of blackouts and cascading failures due to cyberattacks is a significant concern and highly challenging because the existing Intrusion Detection System (IDS) falls behind in handling missing rates, response times, and detection accuracy. Addressing this problem with an early attack detection mechanism with a reduced missing rate and decreased response time is critical. The development of an Intelligent IDS is vital to the mission-critical infrastructure of a smart grid to prevent physical sabotage and processing downtime. This paper aims to develop a robust Anomaly-based IDS using a statistical approach with a machine learning classifier to discriminate cyberattacks from natural faults and man-made events to avoid blackouts and cascading failures. The novel mechanism of a statistical approach with a machine learning (SAML) classifier based on Neighborhood Component Analysis, ExtraTrees, and AdaBoost for feature extraction, bagging, and boosting, respectively, is proposed with optimal hyperparameter tuning for the early discrimination of cyberattacks from natural faults and man-made events. The proposed model is tested using the publicly available Industrial Control Systems Cyber Attack Power System (Triple Class) dataset with a three-bus/two-line transmission system from Mississippi State University and Oak Ridge National Laboratory. Furthermore, the proposed model is evaluated for scalability and generalization using the publicly accessible IEEE 14-bus and 57-bus system datasets of False Data Injection (FDI) attacks. The test results achieved higher detection accuracy, lower missing rates, decreased false alarm rates, and reduced response time compared to the existing approaches.

Application of UAV Multispectral Imaging to Monitor Soybean Growth with Yield Prediction through Machine Learning

The application of remote sensing, which is non-destructive and cost-efficient, has been widely used in crop monitoring and management. This study used a built-in multispectral imager on a small unmanned aerial vehicle (UAV) to capture multispectral images in five different spectral bands (blue, green, red, red edge, and near-infrared), instead of satellite-captured data, to monitor soybean growth in a field. The field experiment was conducted in a soybean field at the Mississippi State University Experiment Station near Pontotoc, MS, USA. The experiment consisted of five cover crops (Cereal Rye, Vetch, Wheat, Mustard plus Cereal Rye, and native vegetation) planted in the winter and three fertilizer treatments (Fertilizer, Poultry Liter, and None) applied before planting the soybean. During the soybean growing season in 2022, eight UAV imaging flyovers were conducted, spread across the growth season. UAV image-derived vegetation indices (VIs) coupled with machine learning (ML) models were computed for characterizing soybean growth at different stages across the season. The aim of this study focuses on monitoring soybean growth to predict yield, using 14 VIs including CC (Canopy Cover), NDVI (Normalized Difference Vegetation Index), GNDVI (Green Normalized Difference Vegetation Index), EVI2 (Enhanced Vegetation Index 2), and others. Different machine learning algorithms including Linear Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) are used for this purpose. The stage of the initial pod development was shown as having the best predictability for earliest soybean yield prediction. CC, NDVI, and NAVI (Normalized area vegetation index) were shown as the best VIs for yield prediction. The RMSE was found to be about 134.5 to 511.11 kg ha−1 in the different yield models, whereas it was 605.26 to 685.96 kg ha−1 in the cross-validated models. Due to the limited number of training and testing samples in the K-fold cross-validation, the models’ results changed to some extent. Nevertheless, the results of this study will be useful for the application of UAV remote sensing to provide information for soybean production and management. This study demonstrates that VIs coupled with ML models can be used in multistage soybean yield prediction at a farm scale, even with a limited number of training samples.

Classification of intrusion cyber‐attacks in smart power grids using deep ensemble learning with metaheuristic‐based optimization

AbstractThe most advanced power grid design, known as a ‘smart power grid’, integrates information and communication technology (ICT) with a conventional grid system to enable remote management of electricity distribution. The intelligent cyber‐physical architecture enables bidirectional, real‐time data sharing between electricity suppliers and consumers through smart meters and advanced metering infrastructure (AMI). Data protection issues, such as data tampering, firmware exploitation, and the leakage of sensitive information arise due to the smart power grid's substantial reliance on ICT. To maintain reliable and efficient power distribution, these issues must be identified and resolved quickly. Intrusion detection is essential for providing secure services and alerting system administrators in the case of adversary attacks. This paper proposes an intrusion classification scheme that identifies several types of cyber attacks on modern smart power grids. Grey‐Wolf metaheuristic optimization‐based feature selection is used to learn non‐linear, overlapping, and complex electrical grid properties. An extended deep‐stacked ensemble technique is advanced by putting predictions from weak learners (CNNs) into a meta‐learner (MLP). The outcomes of this approach are explained and confirmed using explainable AI (XAI). The publicly available dataset from Mississippi State University and Oak Ridge National Laboratory (MSU‐ORNL) is used to conduct experiments. The experimental results show that the proposed method achieved a peak accuracy of 96.6% while scrutinizing the original MSU‐ORNL data feature set and a maximum accuracy of 99% when analysing the selected feature set. Therefore, the proposed intrusion classification scheme may protect smart power grid systems against cyber security attacks.

The people behind the papers - Amber Cook, Taylor Bishop, Hannah Crowe and Andrew Lawton.

Most predictions of how brain folding is determined during development are derived from modelling based on adult brains. In this new study, Andrew Lawton and colleagues test multiple predictions of folding by comparing two mouse strains with naturally distinct amounts of folding. To find out more about the people behind the paper, we caught up with first author Amber Cook, second authors Taylor Bishop and Hannah Crowe, and corresponding author Andrew Lawton, Assistant Professor at Mississippi State University.

Short-Term Contribution of Conservation Practice Implementation to Water Quality Impairments in Small Streams

Voluntary conservation practice adoption is a key strategy to reduce the transport of non-point source pollutants from agricultural lands to downstream ecosystems. This study assessed the short-term (1 year) efficacy of conservation practices to reduce non-point source sediment, nutrient, and fecal indicator bacteria (FIB) transport from working agricultural lands on the Mississippi State University campus, Mississippi State, MS, USA. Water quality was monitored at three treatment sites downstream of the critical resource areas, two of which had paired reference locations. All five sites were monitored for one year pre- and post-conservation practice implementation. Downstream treatment sites generally had higher nutrient and sediment concentrations than upstream reference sites. The results confirmed that the total nitrogen (TN) concentration was reduced post implementation at only the treatment site with the smallest catchment area (p < 0.01). Water quality impairments from FIB were observed across all sites, while treatment locations with livestock presence were found to have significantly elevated staphylococci and E. coli levels following the conservation practice implementation during the winter period. The results of this study showed minimal improvements to TN transport, and in some cases declines in water quality evidenced by increases in FIB, one year after conservation practice implementation. The implementation of conservation practices did not improve the overall water quality to reference levels in the short-term, despite anticipated long-term benefits.

Registration of ‘Espresso’ lowland switchgrass

Abstract‘Espresso’ (Reg. no. CV‐293, PI 687202) is a lowland ecotype (Gulf sub‐population) of switchgrass (Panicum virgatum L.) developed and released by Mississippi State University and the Mississippi Agricultural and Forestry Experiment Station. Espresso was developed from seven cycles of phenotypic recurrent selection for rapid seed germination without stratification. All cycles of selection were made using seed harvested from open‐pollinated isolation blocks containing approximately 100 mother plants. Germination was assessed in laboratory conditions for 15 years, culminating in 2 years of field‐based germination and emergence evaluation at five locations across Mississippi. Over a 2‐year period, germination and emergence of Espresso was significantly greater (P < 0.05) than seed from ‘Alamo’ and ‘Kanlow’. Field testing revealed Espresso as having significantly (P < 0.05) greater establishment success (mean number of seedlings per linear 30 cm) when compared to Alamo. Espresso was released based on the need for southeastern US‐adapted germplasm that possesses enhanced germination characteristics.

GRAY FOR A DAY: A SIMULATION ACTIVITY IMPACTS PARTICIPANTS IN COMMUNITY-BASED HEALTH EDUCATION

Abstract Nearly 80% of older adults own their own homes; approximately two-thirds live within 25 miles of relatives; and most prefer to age-in-place, without access to geriatric-specific care. They rely on primary care clinics and family, friends, and other local support to meet their health needs. The healthcare workforce and lay public need to better understand age-related functional decline. Simulation activities can increase one’s understanding of new concepts or perspectives. One such activity, “Gray for a Day,” offers a didactic presentation and simulations of functional decline. In 2022, 10 Mississippi State University Cooperative Extension agents offered the “Gray for a Day” training to 550 individuals across 35 sessions in community colleges, high school career and technical education classes, and hospitals and nursing home food service departments. Participant evaluations were collected to assess learning outcomes and empathy development. Process evaluation data and descriptive statistics are reported. Nearly all (98%) of the participants reported increases in knowledge of age-related sensory and functional decline and increases in understanding how age-related sensory and functional decline can impact one’s daily life; and 95% reported an increase in considering how they can better interact with older adults experiencing functional decline. Understanding challenges facing older adults is essential to empathy and important both in a home environment and in the workplace. The “Gray for a Day” simulation activity is feasible for implementation in multiple settings and has the potential to increase knowledge, understanding, and possibly empathy for older adults experiencing functional decline.

Cataloging Electronic Theses and Dissertations: Updates and Perspectives from the Mississippi State University Libraries

Cataloging Electronic Theses and Dissertations: Updates and Perspectives from the Mississippi State University Libraries

Mississippi USA'S timber severance tax apportionments and their economic contributions

Eighty percent of Mississippi's timber severance tax collections were authorized by the state legislature in 1984 to fund the Forest Resource Development Program (FRDP), which is a cost share program through which landowners can receive up to 75% reimbursement for their forest management expenses. The remaining 20% is apportioned on shares to counties where timber was harvested. The regional economic contributions of the Mississippi timber severance tax funding spending for FRDP and county governments were determined. Social accounting matrixes (SAMs) were constructed and customized using 2019 data collected by Mississippi State University, IMPLAN, and Lightcast. The modified SAMs were rebalanced to ultimately obtain the total requirements matrix. Funds expended for FRDP and counties in 2019 were run as exogenous final demands. The total possible impact was estimated to be $6.00 million in industrial output and 222 full-time and part-time employment. Because actual expenses were only 72.8% of FRDP funds available that year, this lessened the actual contribution by $1.80 million in output, $1.39 million in value added, and 80 jobs. Regional output contributions correlated moderately with tax collections and forest inventory statistics. Evidence pointed to an inequality of economic flows regarding what South Mississippi paid in tax and the contributions it received in return via industrial activities.

Effects of a Sprinkler and Cool Cell Combined System on Cooling Water Usage, Litter Moisture, and Indoor Environment of Broiler Houses.

Climate change is a serious challenge to food production around the world. Sustainability and water efficiency are critical to a poultry industry faced with global production concerns including increased demands for high-quality, affordable animal protein and greater environmental pressures resulting from rising global temperatures, flock heat stress, and limits on water availability. To address these concerns, a commercial sprinkler system used in combination with a cool cell system was evaluated against a cool cell system alone for two summer flocks of heavy broilers at Mississippi State University to determine effects of sprinkler technology on cooling water usage, litter moisture, and in-house environments. Environmental data were calculated and recorded throughout the flocks. The combination house exhibited a 2.2 °C (4 °F) increase in daily maximum temperature, lower coincident relative humidity, and a 64% (62,039 L/flock) reduction in average cooling water usage over the cool cell-only house. Litter moisture for the combination house tended to be numerically lower but showed no significant difference at several time points between and across flocks. A combined sprinkler/cool cell system reduced cooling water use by 64% over two flocks compared to a cool cell alone system and decreased in-house relative humidity levels.

Intrusion Detection for Industrial Control Systems Based on Improved Contrastive Learning SimCLR

Since supervised learning intrusion detection models rely on manually labeled data, the process often requires a lot of time and effort. To make full use of unlabeled network traffic data and improve intrusion detection, this paper proposes an intrusion detection method for industrial control systems based on improved comparative learning SimCLR. Firstly, a feature extraction network is trained on SimCLR using unlabeled data; a linear classification layer is added to the trained feature extraction network model; and a small amount of labeled data is used for supervised training and fine-tuning of the model parameters. The trained model is simulated on the Secure Water Treatment (SWaT) dataset and the publicly available industrial control dataset from Mississippi State University, and the results show that the method has better results in all evaluation metrics compared with the deep learning algorithm using supervised learning directly, and the comparative learning has research value in industrial control system intrusion detection.

CONSTRUCTION STUDENTS’ PERCEPTIONS OF THE CIRCADIAN RHYTHM

Circadian rhythms, as nychthemeron cycles, are a component of the body’s internal clock that regulate essential functions and processes, which are vividly reflected in sleep routines. While issues pertaining to sleeping patterns are general, the problem is underscored among students who may not have established norms due to internal and external motives. Construction students are among those who may even experience worsened situations emanating from the nature of construction jobs. The construction industry is commonly perceived with long working hours, especially for new employees. Therefore, it is crucial to explore the status of construction students’ routines, knowledge, and perceptions about sleep in the context of education and profession. This study aimed to explore how construction students perceive sleeping hygiene and its impacts. This paper addresses the results of the first stage of the study, conducted in 2021-22 in the Department of Building Construction Science at Mississippi State University. A quantitative method was employed to gather the data. The results highlighted students’ experiences and expectations about sleep deprivation and its impact on their performances. The findings of this study contribute to the body of knowledge by emphasizing the importance of sleep effects on construction students’ education and careers.

Industrial Internet Intrusion Detection Based on Res-CNN-SRU

Nowadays, the industrial Internet is developing rapidly, but at the same time it faces serious information security risks. At present, industrial Internet data generally have the problems of complex attack sample types, large numbers, and high feature dimensions. When training a model, the complexity and quantity of attack samples will result in a long detection time for the intrusion detection algorithm, which will fall short of the system’s real-time performance. Due to the high feature dimension of the data, shallow feature extraction will be unable to extract the data’s more significant features, which will render the model’s overall detection capacity insufficient. Aiming at the above problems, an industrial Internet intrusion detection method based on Res-CNN-SRU is proposed. This method not only considers the temporality of network traffic data but can also effectively capture the local features in the data. The dataset used in the experiment is the gas pipeline industry dataset proposed by Mississippi State University in 2014. Experiments show that the algorithm can effectively improve the recognition rate of the system and reduce the false-alarm rate. At the same time, the training time required for this method is also greatly shortened, and it can perform efficient intrusion detection on the industrial Internet.

In vitro antibacterial efficacy of autologous conditioned plasma and amniotic membrane eye drops.

To determine the in vitro antibacterial efficacy of equine and canine autologous conditioned plasma (ACP) and amniotic membrane extract eye drops (AMEED) against aerobic bacteria common to the corneal surface. Canine (n = 4) and equine (n = 4) anticoagulated whole blood samples were sterilely collected, pooled for each species, and processed using the Arthrex ACP® Double-Syringe System. Platelet counts were performed on ACP and pooled blood. AMEED were obtained from a commercial source. An electronic medical records search (2013-2022) identified aerobic bacteria cultured from canine and equine corneal ulcers at Mississippi State University College of Veterinary Medicine (MSU-CVM). Ten commonly isolated bacteria for each species were collected from cultures submitted to the MSU-CVM Microbiology Diagnostic Service and frozen at -80°C. The Kirby-Bauer disk diffusion method was used to determine the sensitivities of these isolates to ACP and AMEED. Bacterial isolates were plated onto Mueller-Hinton +5% sheep blood agar and blank sterile discs saturated with 20 μL of ACP or AMEED were tested in duplicate. Imipenem discs served as positive controls and blank discs as negative controls. Zones of inhibition were measured at 18 h. ACP platelet counts were 1.06 and 1.65 times higher than blood for equine and canine samples, respectively. Growth of a multi-drug resistant Enterococcus faecalis was partially inhibited by canine and equine ACP. AMEED did not inhibit growth of any examined bacteria. Canine and equine ACP partially inhibited E. faecalis growth in vitro. Further studies using varying concentrations of ACP against bacterial isolates from corneal ulcers are warranted.

An integrated weed management approach in tomato using soil steaming, mulching, and winter cover crops

One of the most significant yield losses in tomato (Solanum lycopersicum L.) is due to weeds. Yellow and purple nutsedge, large crabgrass, and Palmer amaranth are the most troublesome weed species in tomato production throughout the southeastern United States. This study aimed to determine the impact of soil steaming, plastic mulching, and cover crops on weed suppression, tomato height, and fruit yield. The cover crops used were hairy vetch (Vicia villosa), crimson clover (Trifolium incarnatum), and cereal rye (Secale cereale). The study was conducted at the Mississippi State University Truck Crops Experiment Station in Crystal Springs, Mississippi, USA. The experiment used a completely randomized block design with three fall cover crop treatments, including fallow, and each was replicated three times and repeated in two years. Each plot was broadcasted with a mixture of yellow nutsedge (Cyperus esculentus L.), large crabgrass (Digitaria sanguinalis L.), barnyardgrass (Echinochloa crus-galli), and Palmer amaranth [Amaranthus palmeri (S.) Watson] at a density of 20 plants m-2 for each weed species. Two days after sowing the weed seeds, the soil surface was steamed according to its assigned treatment until it reached 61°C for either 0, 5, or 20 min. After steaming, drip irrigation tubing was laid on each row, and covered by black, 0.0254-mm plastic mulch. Data were recorded in both years, including weed cover, plant height, and fruit yield. The lowest weed cover was observed at 5 min of soil steaming in mulched treatment, and the highest cover was noted at 0 min of soil steaming in the absence of mulching. Yellow nutsedge was the dominant weed species, even under steam and mulch treatments. The use of cover crops did not show a difference compared to fallow treatments. However, hairy vetch showed the lowest weed cover, followed by crimson clover. Tomato plants in steamed soil were up to 13 cm taller than those in unsteamed soils. Additionally, steaming at 5 or 20 min in combination with plastic mulch increased the marketable and cull yield. Soil steaming and mulching increased tomato plant height and yield while decreasing weed population and can, therefore, be effectively incorporated into an integrated weed management program in tomato.