Texas A&M University-Kingsville: AKM Digital Repository
Not a member yet
1689 research outputs found
Sort by
Herbicidal management of Kleberg Bluestem (DIchanthium annulatum) in South Texas
No full textKleberg bluestem (Dichanthium annulatum) is a nonnative grass introduced in South Texas for forage production and erosion control, but now is invasive and out-competes more desirable species in rangelands, improved pastures, and hay meadows. This study evaluated fifteen herbicide treatments, rates, and timings for long-term management. Field trials were conducted near Banquete (2019) and Orange Grove, TX (2020). Treatments were applied at three timings (spring, summer, and fall). In 2019, at 47 days after summer application, a 43-55% decrease in bluestem density was observed with treatments of pendimethalin + glyphosate (2.77 + 1.26 kg ae ha-1; spring) followed by (fb) glyphosate (1.26 kg ae ha-1; summer) and indaziflam + glyphosate (43.87 or 73.12 g ai ha-1 + 1.26 kg ae ha-1; spring) fb glyphosate (1.26 kg ae ha-1; summer). At 25 days after fall application in 2019, treatments of indaziflam (43.87 and 73.12 g ai ha-1; spring) + glyphosate (43.87 or 73.12 g ai ha-1 + 1.26 kg ae ha-1; spring) fb glyphosate (1.26 kg ha-1; summer and fall), pendimethalin + glyphosate (2.77 + 1.26 kg ae ha-1; spring) fb glyphosate (1.26 kg ha-1; summer and fall), and glyphosate (1.26 kg ae ha-1; spring, summer, and fall) resulted in a 79-67% decrease in bluestem density. At 28 days after fall application in 2020, indaziflam + glyphosate (73.12 g ha-1 + 1.26 kg ae ha-1; spring) fb glyphosate (1.26 kg ha-1; summer and fall) and indaziflam + glyphosate (43.87 g ai ha-1 + 1.26 kg ae ha-1; spring and summer) fb glyphosate (1.26 kg ha-1; fall) resulted in a decrease in bluestem densities as well. Single applications of imazapyr (336.52 g ai ha-1), glyphosate (1.26 kg ae ha-1), and imazapyr + glyphosate (336.52 g ai ha-1+1.26 kg ae ha-1) applied in spring or sequential applications of nicosulfuron + metsulfuron-methyl (59.06+15.7622 g ai/ha-1, 39.37+10.51 g ai/ha-1, 19.67+5.26 g ai/ha-1) didn’t provide season-long control. This study indicates that the combination of preemergence herbicide indaziflam with sequential applications of glyphosate can provide effective levels of control for Kleberg bluestem on heavily invaded sites, while also suppressing later-emerging seedlings for later control
Efficacy of commercial acaricide with insect growth regulators against Amblyomma americanum (Acari:ixodidae) and Haemaphysalis longicornis (Acari: ixodidae)
No full textTicks are an important ectoparasite in livestock and are responsible for severe economic loss in the cattle industry. A major problem for many cattle ranches in southern Texas are Cattle Fever Ticks: Rhipicephalus annulatus and Rhipicephalus microplus. These ticks not only impact livestock but also wildlife, such as deer and nilgai. The most common and effective method to control tick populations is chemical control. Constant use of these insecticides has led to development of resistance to a variety of acaricides. Resistance to insecticides has made it harder to control tick populations to avoid loss of production. New insecticides need to be evaluated for their efficacy against ticks. Therefore, our objective was to conduct two in-vitro assays to determine effectiveness of insect growth regulators against lab-reared ticks, Amblyomma americanum and Haemaphysalis longicornis. Products tested included diflubenzuron, permethrin, fluazuron, and diflubenzuron plus permethrin. Results of Larval Packet Test (LPT) showed a 100% mortality rate for permethrin, fluazuron, and diflubenzuron plus permethrin for both tick species. Results of the engorged larvae test, for A. americanum, showed a high mortality rate for permethrin, fluazuron, and diflubenzuron plus permethrin. Therefore, a high total fecal count and molting for diflubenzuron and control group was obtained. In H. longicornis, diflubenzuron resulted in a 50% higher mortality rate compared to A. americanum. Overall, permethrin + diflubenzuron seems to be the best product to control cattle fever tick (CFT). An in- vivo assay might be our next step to determine effectiveness of treatments with more susceptible ticks and different environment conditions
Countering cybersecurity threats in smart buildings using machine learning
No full text“Smart Building technology” refers to all software and hardware that makes buildings “smarter.” Smart buildings make efficient and economical use of resources while creating a safe and comfortable environment for occupants. Buildings that are smart use resources efficiently and economically while ensuring occupants can live in a safe and comfortable environment. The Internet of Things (IoT) and sensors are the main things for smart buildings, as they capture data transmitted to the cloud. Artificial Intelligence (AI) is also essential for controlling and optimizing smart building performance. Analytical software that interprets the sensor data, a user interface, and connectivity are also necessary. All these technologies improve performance, optimize space, and use less energy by automating lighting, temperature, and security. Yet these technologies have exposed smart buildings to the threat of cyberattacks such as DoS attacks, data breaches, Session Hijacking, and false-injection attacks. Attackers can manipulate building software or hardware to cause instability or malfunction of some processes remotely or physically.
Cyber Attacks on IoT devices can compromise sensitive user data. As connected IoT devices are open to other malicious activities, protecting them has become one of the biggest challenges in Cybersecurity. Hence, it is necessary to create various safety measures to protect occupants from such attacks. Cybersecurity problems in smart cities have to be addressed by first identifying the threats and Challenges to citizen privacy. Moreover, Smart buildings can suffer from door issues that result from malfunctioning door mechanisms. To address the cybersecurity threat of smart buildings, we propose a smart door system framework that identifies cyber threats in real-time and enables immediate defense against them. By using AI techniques and various types of sensors, we will recognize any threats that may exist, and we will respond to them accordingly. An intelligent home security system based on multiple sensor devices provides several advantages in providing safety and security, such as visualizing and identifying people. These technologies will keep the home secure and help provide safety and security by monitoring and recognizing people. The proposed system will have two effective techniques to provide home security. First, detecting motion at the front door in real-time, regardless of whether anyone is inside, is the first step; the second is establishing communication between the door system and a smartphone device.
The main objective of the Thesis is to reduce risks by identifying possible attacks, and we aim to extend the door automation technique with artificial intelligence-based techniques to provide users with more security, ease, and comfort
Impact of Texas Title One priority schools (TTIPS) grant funding on student achievement
No full textSchool Improvement Grants (SIGs) have provided millions of dollars in federal funding to improve low-performing schools. Studying the Texas Title One Priority Schools (TTIPS) Cycle 3 Grant is important because not enough empirical evidence substantiates whether the grant had an
impact. The purpose of this study was to determine whether the TTIPS Cycle 3 Grant impacted student achievement in reading and mathematics in nine elementary schools in Texas (Regions 2, 3, 12, 16, 17, and 20) that received funding from 2014-2017 and utilized the Transformation Model. Human capital theory was the theoretical foundation utilized in this study as it pertains to the investing in individuals’ learned knowledge and skills by contributing to society as productive
citizens. Research questions focused on differences in overall student achievement as measured by the State of Texas Assessments of Academic Readiness (STAAR) All Grades (3-5) reading and mathematics scores before and after grant funding. The research design was a quasi-experimental ex post facto time series design using archived data. Statistically, a paired samples t-test
determined a statistical significance between reading and mathematics pre- and post-scores. Recommendations included monitoring sustainability of SIG funding to sustain best practices, investment in human capital, and improvements that increase student achievement in reading and mathematics
The impact of curriculum and instruction personnel on leadership and school performance
No full textWith high stake accountability testing, local education agencies are working diligently with their curriculum and instruction department to ensure leadership capacity building within school administration. This phenomenological qualitative study analyzed the practices of curriculum and instruction personnel to determine impact on leadership and school performance.
In the process, this study will bring clarity to the roles and responsibilities of the curriculum and instruction team. The research further sought to explore the efficacy of instructional approaches applied by curriculum and instruction personnel to support campus leadership. Analyzing the impact curriculum and instruction personnel have on school leadership and school performance
will aid other districts in facilitating assistance to school leaders while meeting the demands of both the state and federal accountability systems. The outcome of the study entails the use of the curriculum and instruction personnel as a support system to build leadership capacity through the use of best practice as instructional practice. Prevalent themes identified within the study were support service, instructional practices, professional development, and data reviews. The utilization of professional learning communities, data plans, coaching, and data analysis were some of the instructional practices utilized to impact school district and leadership performance
English as a foreign language pre-service teachers' perceptions of task-based language teaching in Northern Mexico
No full textThis quantitative case study research explored English as a foreign language (EFL) pre-service teachers’ perceptions of the Task-Based Language Teaching (TBLT) for facilitating written language competency for learners of English in northern Mexico. EFL pre-service teachers are trained to serve in basic education secondary school, teaching English. The participants in the study (n = 193) were from two locations: Universidad Autónoma de Nuevo León (UANL) with 112 EFL Pre-service teachers (Location A) and The Normal Superior School with 81 EFL pre-service teachers (Location B). All participants took part in the study voluntarily and responded to a survey via email.
The study included a survey developed for this study with a five-point scale (1. Strongly disagree, 2. Disagree, 3. Neutral, 4. Agree, 5. Strongly agree). The survey contained a demographic section followed by three subscales with a total of 35 items: Subscale one, The Teaching Process (items 1-10); Subscale two, The Role of the Task (items 11-18), and Subscale 3, The Teaching of Language Competencies (items 19-35). The last five items of the survey (31-35) asked participants about the facilitation of written expression competency through TBLT. The items of the survey responded to the three research questions (RQ) of the study, regarding perceptions about TBLT approach to the teaching of English; the role of task in English language learning; and the contribution of TBLT for teaching of the written expression competency in Mexico.
The construction of the survey was informed by contributions of Prabhu (1987), Nunan (2004), Swan (2005), Ellis (2009), Long (2015), Jeon and Hahn (2006), and SEP (2011). The survey was administered using the SurveyMonkey platform. Findings showed that the majority of participants (n = 152, 79%) responding with the two highest points of the Likert scale (4. Agree and 5 Strongly agree), perceived that the TBLT approach facilitated EFL teaching process, specifically competency in the written expression of learners of English. However, there were 29 (15%) participants who provided a neutral response and 12 (6%), who by indicating the two lowest points of the Likert scale (1 Strongly disagree and 2 Disagree), had a negative response that TBLT facilitates the teaching of the writing skills
Autonomous vehicle number plate recognition using a YOLO and Esrgan framework
No full textA vehicle number plate recognition framework supports law enforcement for recognizing and tracking vehicles that are lost or used for criminal activities. In this project, to identify the vehicle number plates, You Look Only Once (YOLO) v4 method is employed in combination with
Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN). ESRGAN enhances the images and gives a clear view of the number, which facilitates the detection. Finally, Optical Character Recognition (OCR) is run on the extracted image of the vehicle number plate. This OCR has three steps: It takes the input image. It preprocesses the image to improve its quality for
character recognition. After this preprocessing, it performs text recognition in the license plate image via matrix matching, which involves pattern matching, pattern recognition, and image correlation. After this text recognition, it does the post-processing to increase accuracy. This method also reduces the time complexity for detecting a number plate on the vehicle. The proposed
method in this work using YOLO v4 and ESRGAN achieved an average test accuracy of 98.3% for license plate character recognition on the same test dataset as compared to 96.1% average test accuracy achieved by a previous YOLO and OCR method
Franklin's randomized response model with correlated scrambling variables
No full textThe problem of estimating population proportion of a sensitive characteristic by making use of a randomization device has been well established in the field of survey sampling. The innovative idea of protecting respondents while questioning them about their membership in a particular sensitive group was introduced by Warner (1965). Since that time, several techniques have been introduced in literature which leads to estimators which are more efficient than the initial estimator of Warner (1965). Franklin (1989a) introduced a new randomization device which possesses certain distinctive qualities over the other existing randomized response techniques. The uniqueness of the Franklin (1989a) model is that several existing randomized response models can be derived from it. Therefore, it seems worthy to explore Franklin’s randomized response model for further developments in the field of survey sampling. In this thesis, we have made improvements over the Franklin (1989a) model by making use of correlated outcomes reported by the respondents participating in surveys. New estimators of the population proportions were developed, and their bias and variance expressions derived. At the end, our findings are supported by numerical illustrations and simulation studies. SAS and R programming have been utilized for producing simulated results
Evaluation of field deployable leaf sensor for rapid and asymptomatic screening of Huanglongbing (HLB) in citrus
No full textHuanglongbing (HLB), a bacterial disease associated with Candidatus Liberibacter asiaticus (CLas), is a devastating citrus disease that compromises the tree’s health, leading to yield reduction and significant losses to the citrus industry worldwide. Long latency period for symptom development of infected trees makes it a challenging task to diagnose the disease at the pre-symptomatic stage. Currently, quantitative real-time polymerase chain reaction (qPCR) is the most widely used technique for HLB diagnosis in citrus. However, qPCR methods require expensive instruments, trained personnel, and a laboratory setup lacking field detection capabilities. Rapid in-field asymptomatic screening of HLB would facilitate implementation of disease management strategies in a timely manner. The study presented here aims to evaluate the efficacy of a portable spectrometer equipped with G-Fresnel lens for HLB screening using asymptomatic leaves from mature (12 years old) and young (1 year old) Valencia sweet orange trees under field conditions. This device utilizes the principle of diffuse reflectance spectroscopy in a wavelength ranging from 400nm to 1000nm, with a wavelength resolution of ~5nm that can monitor the internal structural changes in the leaf and is equipped with cloud-based data analysis capability. The spectral data obtained from the smartphone spectrometer is analyzed using an algorithm of a statistical model, where the statistical model for analysis was selected from pre-established models based on the HLB incidence in the field. The HLB incidence rate in the field was estimated by CLas detection using qPCR assay. Improving the sensitivity of CLas detection by qPCR improves the estimation of the HLB incidence, leading to accuracy in selection of statistical model for spectral data analysis using spectrometer. To improve the CLas detection by qPCR diagnosis, triplex qPCR using both RNRfpr and LJ900fpr for CLas detection together with COXfpr was conducted targeting two CLas and one citrus genomic locus in a single reaction. The results obtained were compared with duplex qPCR using either RNRfpr or LJ900fpr together with COXfpr. The preliminary data showed that there is a decrease in Ct values in triplex qPCR when compared to duplex qPCR using either RNRfpr or LJ900fpr together with COXfpr
New Developments in randomized response sampling utilizing two responses per respondent
No full textWarner (1965) came up with an ingenious idea of estimating the population proportion of a sensitive characteristic by asking questions indirectly through a randomization device in face-to-face interviews. The point of this methodology is to ask sensitive questions in such a way that the respondents should not feel threatened while revealing truth about themselves. He uses a simple randomization device, such as a spinner, with two mutually exclusive outcomes; one related to the membership of a sensitive group and the other on its complement. If the outcome from the spinner matches with the status of the experiencing respondent, then he/she is asked to report “yes”. If the outcome from the spinner does not match with the status of the respondent,
then he/she reports “no”. This simplicity within the respondent’s answer allows them to protect
his/her privacy from the interviewer. Without a doubt, Warner’s (1965) idea has spread in the vast literature of survey sampling and has sprung the curiosity of many to be keen in developing an estimator better than his. Odumade and Singh (2009) came up with an idea of using a pair of deck of cards while collecting information on a sensitive characteristic in a face-to-face survey.
The Odumade and Singh (2009) model has been observed to be an improvement over the Warner (1965) randomized response model and differs in the fact that it uses two randomization devices instead of one. In this thesis, we will further explore the possibilities of improving the Odumade and Singh (2009) model. A few new estimators of the population proportion will be developed and their properties such as unbiasedness, variances, and relative efficiency will be investigated analytically, empirically and through simulation studies. Lee, Sedory and Singh (2013) extended the idea to estimate the prevalence of two sensitive characteristics and their overlap. In this
thesis we also consider the problem of estimating a proportion of two sensitive characteristics and potential overlap by making use of single deck of cards which certainly reduces burden on respondents. SAS or R codes will be provided and used in supporting the proposed estimators