Shahrul Fazly Man@Sulaiman
Thumbnail Image
Preferred name
Shahrul Fazly Man@Sulaiman
Official Name
Shahrul Fazly, Man@Sulaiman
Main Affiliation

Research Output

Filters

3
3
3
Reset filters

Settings
Now showing 1 - 3 of 3
No Thumbnail Available
Publication

AI Assisted and IOT Based Fertilizer Mixing System

2024-06-03 , Wan Mohd Faizal Wan Nik , Shahrul Fazly Man@Sulaiman , Muhammad Imran Ahmad , Shafie Omar , Tan Shie Chow , Mohd Nazri Abu Bakar , Fadhilnor Abdullah , Muhammad Khamil Akbar

Agriculture techniques, particularly fertilizer mixing, have significant impacts on crop productivity. Introducing IoT technology to agriculture can enhance productivity, and machine learning offers a mechanism to gain insights from data, making agricultural practices more efficient. This research aims to design an AI-assisted and IoT-based fertilizer mixing system for greenhouses. This system utilizes sensor data and AI algorithms, specifically the Support Vector Machine (SVM), to optimize fertilizer application. Results from the SVM classifier showed a 100% accuracy rate for temperature and humidity, 65% accuracy for phosphorus, 86% for nitrogen, and 100% for potassium. These findings demonstrate the potential of the proposed system to improve fertilizer efficiency while reducing labor and resource waste.

View
No Thumbnail Available
Publication

Plant Disease Classification Using Image Processing Technique

2024-06-03 , Shahrul Fazly Man@Sulaiman , Shafie Omar , Muhammad Imran Ahmad , Wan Mohd Faizal Wan Nik , Tan Shie Chow , Mohd Nazri Abu Bakar , Fadhilnor Abdullah , Asbhir Yuusuf Omar

Agriculture remains pivotal to our economy, with farming playing a central role in revenue generation. Challenges such as pests, plant diseases, and evolving climate patterns pose threats to crop yield and production. Addressing these challenges, timely and accurate detection of plant diseases emerges as imperative. Manual detection, however, remains resource-intensive and often lags. Addressing this gap, this project proposes an innovative image processing-based system for rapidly detecting plant diseases. The system proficiently identifies specific diseases by analyzing images of plant leaves against a curated dataset. The emphasis of this study was on three major diseases: Bacterial Blight (with an accuracy of 98.6%), Alternaria Alternata (98.5714%), and Cercospora Leaf Spot (97.5%). The compelling results underline the system's capacity to swiftly and effectively categorize diseases, offering monoculture farmers an indispensable tool for obtaining prompt, disease-specific insights.

View
No Thumbnail Available
Publication

IoT Enabled Mushroom Farm Automation with Machine Learning

2024-06-03 , Shafie Omar , Wan Mohd Faizal Wan Nik , Muhammad Imran Ahmad , Tan Shie Chow , Mohd Nazri Abu Bakar , Shahrul Fazly Man@Sulaiman , Fadhilnor Abdullah , Vikneshwara Ram Suppiah

Mushroom farming has gained prominence due to its significant contribution to the global market. One major challenge for mushroom cultivation is maintaining optimal environmental conditions, specifically temperature and humidity. Traditional farming methods, prevalent in many parts of the world, lack precise control over these parameters, often leading to poor yield. This paper presents an innovative approach combining the Internet of Things (IoT) and Machine Learning (ML) for mushroom farm automation. The proposed system employs the ESP8266 microcontroller with specific agricultural sensors for smart monitoring. To regulate the farm's environmental conditions, ML algorithms predict mushroom farm weather states: mild, normal, and hot. The ensemble ML model, comprising five classifiers – Decision Tree, Logistic Regression, K-nearest neighbor, Support Vector Machine, and Random Forest – delivers a commendable accuracy of 100% when combining predictions, surpassing the performance of individual classifiers. This integrated IoT and ML approach promises to revolutionize real-time automation and cultivation practices in the mushroom industry.

View