ASEAN Journal of Scientific and Technological Reports

Efficient and Rapid Classification of Various Maize Seeds Using Transfer Learning and Advanced AI Techniques

2024-09-14T14:29:20+07:00

The classification of maize (Zea mays) is crucial for agricultural efficiency, breeding programs, and market specifications. The EfficientMaize dataset was utilized alongside Google’s Teachable Machine to develop a model separating maize varieties into classes: Bhihilifa, SanzalSima, and WangDataa. As a result, the study demonstrated that user-friendly machine learning tools are helpful in agriculture since they delivered high accuracy rates, such as 99% in Bhihilifa, 95% in SanzalSima, and 85% in WangDataa. This paper also emphasizes how modern machine-learning technologies can be accessible to farmers and researchers through tools such as Google’s Teachable Machine, which does not require coding knowledge or online expertise. To validate the results obtained with Google Teachable Machine, further analyses were conducted using RESNET-50. These findings add to previous studies on deep learning and hyperspectral imaging, leading to seed classification by increasing the potential of using machine learning to improve agricultural practices and food security.

Effect of Dietary Supplementation with Durio zibethinus Murr. cv. Monthong Rind on the Hematology and Innate Immune Response Against Aeromonas hydrophila in Red Tilapia (Oreochromis niloticus x Oreochromis mossambicus)

2024-09-06T12:55:23+07:00

In vivo activity study of durian rind (Durio zibethinus Murr. cv. Monthong) as a supplement in the tilapia diet was performed to evaluate its effect on the hematology and innate immune response against Aeromonas hydrophila. Red Tilapia (Oreochromis niloticus x Oreochromis mossambicus), with an initial average weight range of 40-45 g, were fed diets supplemented with durian rind at 0, 10, 15, and 20% for 140 days and were injected intraperitoneally with A. hydrophila. Blood was collected at 1 and 3 hours post-injection to determine the phagocytic activity and the durian rind efficiency in protecting fish red blood cells from hemolysin produced by A. hydrophila. Ensuing results showed that red blood cell counts in fish-fed durian rind supplemented diets, regardless of inclusion levels, did not decrease at 3 hours post bacterial infection and were significantly higher than control (P<0.05). Furthermore, we observed that the total white blood cells and phagocytic activity of all fish fed with diets supplemented with durian rind increased from 1-hour post-infection, indicating a heightened innate immune response. Taken together, durian rind supplementation of at least 10% in the tilapia diet can act as an immunostimulant and improve innate immune response and anti-hemolytic activity against A. hydrophila infection.

Enhancing Autonomous Driving: A Novel Approach of Mixed Attack and Physical Defense Strategies

2024-09-18T09:09:49+07:00

Adversarial attacks are a significant threat to autonomous driving safety, especially in the physical world where there is a prevalence of "sticker-paste" attacks on traffic signs. However, most of these attacks are single-category attacks with little interference effect. This paper builds an autonomous driving platform and conducts extensive experiments on five single-category attacks. Moreover, we proposed a new physical attack - a mixed attack consisting of different single-category physical attacks. The proposed method outperforms existing methods and can reduce the accuracy of traffic sign recognition of an autonomous driving platform by 38%. Furthermore, we proposed a new anti-jamming model for physical adversarial defense, CBAM-ResNet26 & CBAM-Alexnet, which improves an autonomous driving platform's traffic sign recognition accuracy to 63% under mixed attack. Finally, experiments were also conducted with datasets with different ratios of adversarial attack examples, and the experimental results showed that in adversarial training, the higher the ratio of adversarial examples, the higher the recognition accuracy. However, a too high ratio would reduce the accuracy of normal traffic signs. Finally, the optimal ratio for physical adversarial defense training is 1:2.

An An AIoT-based Air Quality Monitoring System for Real-time PM2.5 Prediction in Urban Environments

2024-09-28T17:01:24+07:00

This study aimed to develop an air quality monitoring and forecasting system focusing on PM_2.5 using a combination of AI of Things (AIoT) technology. The system was designed to provide warnings of PM_2.5 levels through a mobile application. Air pollution, particularly PM_2.5, is a significant health concern globally, with Southeast Asia being heavily affected. Bangkok, Thailand, experiences high PM_2.5 concentrations during cool weather. Existing research explores short-term PM_2.5 prediction using AIoT. Still, there is a need for improved software, hardware, and ML algorithms for user-friendly mobile applications with real-time data access and health advisories. The system was installed on a building next to a main road in Bangkok. It collected data on PM_2.5. The Air Quality Index (AQI) was used to categorize PM_2.5 levels and their health impacts. Time series analysis with moving averages and the Random Forest algorithm were employed in advance for PM_2.5 forecasting. A mobile application was developed to provide a user interface and data visualization. The MARF (Moving Average and Random Forest) model emerged as a success, achieving higher accuracy (average of 92.59%) for 1-hour advance forecasts compared to the Moving Average (MA) model (average of 84.16%). The developed system demonstrates the potential of AIoT for accurate PM_2.5 monitoring and forecasting. Future research could explore more advanced ML algorithms and integrate additional environmental factors for enhanced forecasting accuracy.

Technology Transfer Needs Assessment for Turmeric Farmers in Paphayom District, Phatthalung Province

2024-10-08T10:47:56+07:00

These research findings indicated that most farmers in the study were female, with an average age of 55.88 years and an average of 10.80 years of experience in cultivating turmeric. Often, turmeric is grown in conjunction with rubber cultivation. Farmers typically cultivated turmeric on 1.38 rai, preferring sandy loam soil conditions. Their preferred turmeric variety was Trang 84-2. The planting method involves digging holes 10 centimeters deep, with a spacing of 25 cm x 25 cm. Turmeric cultivation usually commences in June and concludes with its harvest in April. During planting, farmers apply a combination of chemical and biological fertilizers, along with organic manures. Weed control is primarily achieved by manual removal, while chemical treatments are employed to manage diseases and insect pests. During the 2021/2022 production season, farmers achieved an average turmeric yield of 1,675.81 kg. Farmers encounter various problems in turmeric production, particularly in maintenance, disease, and insect management. These issues adversely affect the planting area, often reducing production yields. The most pressing need for farmers is to acquire turmeric production technology, particularly in pest mitigation and elimination. Farmers want to receive knowledge transfer, such as with individualized guidance through farm visits. The media that farmers require include brochures and the assistance of extension officers. Additionally, the level of experience in turmeric production significantly influenced the farmers' technology transfer needs at a 0.01 significance level.

Induction of Mutation in Toothbrush Orchids using Ethylmethane Sulphonate (EMS) and Detection of Genetic Variation by SSR (Simple Sequence Repeat) Marker

2024-10-07T14:46:03+07:00

The toothbrush orchid is a monocotyledonous plant. It is classified in the Dendrobium genus, which is an important and outstanding economic flowering plant. It is an orchid that is sold both domestically and internationally due to its overall toothbrush-like design, both in the flower arrangement and stem. Recently, there are fewer toothbrush orchids now. Therefore, the objective of this study was to study the effects of Ethylmethane sulfonate (EMS) and detection of genetic variation in toothbrush orchids using simple sequence repeat (SSR) marker. EMS solution was used to soak a 0.5 cm piece of the protocorm like bodies (PLBs) at concentrations of 0.5, 1.0, and 2.0%. Thereafter, they were cultured on VW medium supplemented with 1 mg/l BA, 20 g/l sucrose, adjusted to pH 5.7 and solidified with 0.2 % phytagel. The cultures were maintained at 26 ±2°C under light at intensity of 3,000 lux for 14 hours per day. After culturing for 30 days, the results showed that PLBs immersed in EMS solution at 1.7% for 90 minutes gave the highest average survival rates at 50% (LD₅₀). For the development of PBLs, PBLs were derived with immersed in 0.5 % EMS for 60 minutes gave the highest PLBs induction (7.37%) and average number of PLBs (1.41 PLBs/explant) after culturing for 8 weeks. For detection of genetic variation, a total of 9 SSR primers were used, including EgCIR0409, EgCIR0905, EgCIR0781, EgCIR0446, EgCIR1772, EgCIR0337, EgCIR0337, EgCIR0243, mEgCIR0465, and mEgCIR008. The results found that only one primer gave polymorphic banding. EgCIR0905 primer gave polymorphic banding at 50% and specific DNA banding size at 200 bp.

Effects of Different Manures on the Growth Performance of Pechay (Brassica rapa L.) Under Visayas State University Condition

2024-10-08T13:30:29+07:00

The sudden rise in the demand for food in global markets has led farmers to adopt sustainable agricultural practices to reduce the negative impacts of synthetic fertilizers on soil, water, and ecosystems. This study assessed the effectiveness of vermicast, cattle manure, and goat manure and their combinations on the growth performance, nutrient content, and heavy metal accumulation in peachy. The study used a Completely Randomized Design (CRD) with six treatments and three replicates each and collected the data on plant height, number of leaves, leaf area, chlorophyll content, root biomass, and fresh and dry weight. Proximate tissue analysis was done to determine the NPK content and accumulated heavy metals. Data were analyzed using ANOVA in CRD, and a post-hoc analysis using Tukey’s HSD test was conducted after significance was found. The results indicated significant differences (P<0.05) in several growth parameters like plant height with cattle manure, number of leaves with vermicast, leaf area with vermicast, and chlorophyll content with mixed manures. Significant differences (P<0.05) were also found in fresh weight with vermicast and root biomass with mixed manures. These treatments improved soil nutrient profiles post-harvest, although there were reductions in available phosphorus and exchangeable potassium. The findings highlight the potential of organic fertilizers to enhance pechay growth performance, supporting sustainable agricultural practices that align with the Sustainable Development Goals (SDGs). Further research on organic fertilizer applications and their long-term effects on soil health and crop nutrition is recommended.

Methane Oxidation Rates and Efficiencies Across Four Distinct Soil Environments: Implications for Greenhouse Gas Mitigation

2024-10-05T18:26:02+07:00

Methane oxidation by soil microorganisms is crucial in mitigating greenhouse gas emissions. This study investigated methane oxidation potential across four distinct soil environments through standardized laboratory enrichment cultures. Soil samples were collected from landfill-cover soils, rice fields, cattle farms, and pond sediments, with environmental parameters monitored to understand their influence on oxidation rates and efficiencies. Using gas chromatography analysis, we quantified methane oxidation under controlled conditions. Statistical analysis revealed significant differences in oxidation rates across soil types. Landfill cover soils exhibited the highest oxidation rate of 0.39 μmol-CH₄/g-soil dry weight/h and efficiency of 66.5 %. Pond sediments, cattle farm soils, and rice field soils followed with rates of 0.29, 0.28, and 0.27 μmol-CH₄/g-soil dry weight/h, respectively. Oxidation efficiencies for these environments ranged from 46.1% to 48.4%. pH and organic matter content showed strong positive correlations with oxidation rates across all soil types, while environmental moisture content effects varied. The superior performance of landfill soils was attributed to optimal environmental conditions and stable substrate availability. This analysis revealed significant potential for enhancing oxidation efficiencies: landfill soils from 66.5% to 75-85%, rice fields from 46.1% to 60-70%, cattle farms from 47.0% to 55-65%, and pond sediments from 48.4% to 60-75%. Implementing optimized management strategies could reduce methane emissions by 70-90% in landfills, 30-50% in agricultural systems, and 40-60% in aquatic environments compared to current practices. This study highlights the substantial potential for enhancing biological methane oxidation across diverse ecosystems and emphasizes the need for targeted management approaches to optimize methane mitigation strategies.

Preparation and Characterization of Sulfonated CaO Catalyst for Biodiesel Production from Waste Cooking Oil

2024-09-25T18:42:52+07:00

Biodiesel production depends on raw materials. Low-quality oils, such as cooking oil, crude palm oil, and sludge oil, are used to reduce costs, and they contain free fatty acids (FFA) and water. Soap can be produced when using the alkaline catalyst during transesterification. In this work, the sulfonation method prepared the esterification of waste cooking oil by sulfonated CaO as a bifunctional catalyst. The sulfonated CaO catalysts were characterized by X-ray diffraction spectroscopy (XRD), Fourier transform infrared spectroscopy (FTIR), temperature-programmed desorption of carbon dioxide and ammonia (TPD), BET surface area, and scanning electron microscopy (SEM). It was observed that the specific surface area, pore volume, and pore diameter of the CaO increased after being sulfonated with a 2 M sulfuric acid solution. It showed a high total surface acidity and basicity, 7.22 and 3.86 mmol/g, respectively. The optimal FFA conversion (84.94 %) from the waste cooking oil was acquired at a reaction temperature of 65 ˚C, a 9:1 MeOH: Oil molar ratio, and 5 wt% catalyst loading for a 3 h reaction time. The 2 M sulfonated CaO catalyst can be reused twice with a high FFA conversion without further treatment under optimized reaction conditions. The 2 M sulfonated CaO catalyst has potential treatment for biodiesel production from high-FFA oils due to its lower production cost and high catalytic activity.

Biological Oxidation of Dissolved Methane in Palm Oil Mill Biogas Effluents Using an Anoxic Methane-Oxidizing Consortium

2024-09-30T17:31:14+07:00

This study investigated the potential of anoxic methane-oxidizing consortia for mitigating dissolved methane in palm oil mill biogas effluents. Microbial consortia from five soil sources were evaluated under various conditions. The cattle farm effluent-derived consortium demonstrated the highest methane reduction efficiency of 76.89% after a 3-week incubation period, with a methane consumption rate of 49.37 mg-CH₄/m²/d. The landfill soil consortium showed the second-highest performance with a 75.48% reduction efficiency under shaking conditions. Environmental factors significantly influenced methane oxidation performance. Optimal conditions were identified as 35°C, pH 7.0, 0.5 mg/L dissolved oxygen, 55 mg/L nitrate concentration, and 5 g/L NaCl. Plastic media enhanced methane reduction efficiency for most microbial sources, particularly for the cattle farm effluent consortium (67.07% efficiency). Characterization of the palm oil mill biogas effluent revealed a COD of 13.15 g/L, BOD of 7.11 g/L, and total Kjeldahl nitrogen of 0.73 g/L. Carbon mass balance analysis confirmed biological methane oxidation, with 45% converted to CO₂, 38% incorporated into biomass, and 12% as dissolved organic carbon. The developed system can potentially mitigate up to 23,067 t CO₂e/year for an average palm oil mill, with associated cost savings of approximately 115,335 USD/year through carbon credits, assuming a credit value of 5 USD/t CO₂e. These findings demonstrate the potential of anoxic methane-oxidizing consortia for greenhouse gas mitigation in the palm oil industry. The study provides insights into optimal conditions and microbial sources for efficient methane oxidation, paving the way for developing effective biological treatment systems for palm oil mill effluents.