KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 13

Some of the metrics are blocked by your
consent settings
WILDFIRE CLASSIFICATION WITH DEEP LEARNING MODEL
(2026-07-01)
Tankue, Puwanai
;
Kruekaew, Boonhatai
;
Kimpan, Warangkhana
This research aims to develop models for wildfire classification from images. The objective is to provide decision-making support for wildfire control planning, prevention and management. Convolutional Neural Networks techniques are used to analyze a dataset consisting of three image groups: no fire images, fire images that are not wildfires, and wildfire images. The experimental results compared ResNet group, DenseNet group, MobileNet group, and EfficientNet group. The research findings indicate the best-performing model in this study is ResNet152V2, which achieved an accuracy of 92.75%. Furthermore, Precision, Recall, and F1-Score are within a satisfactory range. A web application has also been developed to facilitate users to detect and classify wildfire more conveniently.
Some of the metrics are blocked by your
consent settings
Feature Selection Method Based on Hybrid Cuckoo Search and Firefly Algorithm for Breast Cancer Prediction
(2025-01-01)
Teerasarn, Chalanwich
;
Kimpan, Warangkhana
This research focuses on developing an efficient feature selection process using a hybrid technique combining Cuckoo Search algorithm and Firefly Algorithm. The Wisconsin Diagnostic Breast Cancer dataset is utilized to evaluate the capability of selecting significant features and eliminating irrelevant ones. The experimental results demonstrate that using the hybrid technique significantly improves the accuracy of machine learning compared to using the Cuckoo Search and Firefly Algorithm individually. Additionally, an analysis was conducted on the impact of splitting the dataset for training and testing, with splits of 70/30, 80/20, and 90/10. The experiments revealed a relationship between the number of selected features and the model accuracy. The findings from this study can serve as a guideline for developing appropriate feature selection process for complex data analysis problems.
Some of the metrics are blocked by your
consent settings
STRAWBERRY SEEDLING CULTIVATION WITH SMART FARM
(2025-01-01)
Kruekaew, Boonhatai
;
Palananda, Attapon
;
Kimpan, Warangkhana
In Thailand, the main planting areas are in the upper northern region. It is difficult for strawberry to be cultivated in the central region of Thailand due to inappropriate weather conditions. Because the strawberry seedlings are delicate and sensitive to alterations in temperature and weather, they require extra care than mature plants while cultivation before planting. Therefore, modifications to strawberry seedling cultivation may result in more strawberry plantings. Consequently, one of the most beneficial choices is a smart farming. This research focuses on the cultivation of strawberry seedlings in tropical areas of Thailand using an intelligence model and Internet of Things. The prototype system focuses on automatic watering and lighting, and an environmental adaptation system that combines sensors to control water, air, and lighting. The physical characteristics data from all devices in prototype system are collected, and then analyzed using machine learning methods to automatically control the environment within the prototype system to be suitable for growing strawberry seedlings. Moreover, the real-time data will be displayed on a dashboard with various notification systems. The experimental results indicated that using machine learning models can control the suitable temperature and humidity for strawberry seedlings cultivation. The appropriate temperature and soil moisture are between 31-32 degrees Celsius and 70 percent, respectively.
Some of the metrics are blocked by your
consent settings
Automation 4.0 for Water Level Monitoring System
(2023-01-01)
Kimpan, Warangkhana
;
Palananda, Attapon
;
Kruekaew, Boonhatai
This paper proposed the concept of using automation 4.0 for monitoring the water level. The water level warning system specifications are to measure the water level using ultrasonic sensors and measure the amount of rainfall using a weighing rain gauge. The system automatically controls the measurement of the level of the flood using a Programmable Logic Controller (PLC) via PROFINET. Then the water level is monitored, and the results will be displayed through HMI technology via Web panel trainer, Node-Red dashboard, and transfer data via PROFICLOUD. Moreover, the warning information will be sent via LINE notification on mobile to people who live near water sources or staff in charge of preventing disasters.
Some of the metrics are blocked by your
consent settings
Classification of Adulterated Particle Images in Coconut Oil Using Deep Learning Approaches
(2022-01-01)
Palananda, Attapon
;
Kimpan, Warangkhana
In the production of coconut oil for consumption, cleanliness and safety are the first priorities for meeting the standard in Thailand. The presence of color, sediment, or impurities is an important element that affects consumers’ or buyers’ decision to buy coconut oil. Coconut oil contains impurities that are revealed during the process of compressing the coconut pulp to extract the oil. Therefore, the oil must be filtered by centrifugation and passed through a fine filter. When the oil filtration process is finished, staff inspect the turbidity of coconut oil by examining the color with the naked eye and should detect only the color of the coconut oil. However, this method cannot detect small impurities, suspended particles that take time to settle and become sediment. Studies have shown that the turbidity of coconut oil can be measured by passing light through the oil and applying image processing techniques. This method makes it possible to detect impurities using a microscopic camera that photographs the coconut oil. This study proposes a method for detecting impurities that cause the turbidity in coconut oil using a deep learning approach called a convolutional neural network (CNN) to solve the problem of impurity identification and image analysis. In the experiments, this paper used two coconut oil impurity datasets, PiCO_V1 and PiCO_V2, containing 1000 and 6861 images, respectively. A total of 10 CNN architectures were tested on these two datasets to determine the accuracy of the best architecture. The experimental results indicated that the MobileNetV2 architecture had the best performance, with the highest training accuracy rate, 94.05%, and testing accuracy rate, 80.20%.
Some of the metrics are blocked by your
consent settings
Multi-Objective Task Scheduling Optimization for Load Balancing in Cloud Computing Environment Using Hybrid Artificial Bee Colony Algorithm with Reinforcement Learning
(2022-01-01)
Kruekaew, Boonhatai
;
Kimpan, Warangkhana
Workload balancing in cloud computing is still challenging problem, especially in Infrastructure as a Service (IaaS) in the cloud model. A problem that should not occur during cloud access is a host or server being overloaded or underloaded, which may affect the processing time or may result in a system crash. Therefore, to prevent these problems, an appropriate schedule of access should be considered so that the system can distribute tasks across all available resources, which is called load balancing. The load balancing technique should ensure that all Virtual Machines (VMs) are used appropriately. In this paper, an independent task scheduling approach in cloud computing is proposed using a Multi-objective task scheduling optimization based on the Artificial Bee Colony Algorithm (ABC) with a Q-learning algorithm,which is a reinforcement learning technique that helps the ABC algorithm work faster, called the MOABCQ method. The proposed method aims to optimize scheduling and resource utilization, maximize VM throughput, and create load balancing between VMs based on makespan, cost, and resource utilization, which are limitations of concurrent considerations. Performance analysis of the proposed method was compared using CloudSim with the existing load balancing and scheduling algorithms: Max-Min, FCFS, HABC-LJF, Q-learning, MOPSO, and MOCS algorithms in three datasets: Random, Google Cloud Jobs (GoCJ), and Synthetic workload. The experimental results indicated that the algorithms used MOABCQ approach outperformed the other algorithms in terms of reducing makespan, reducing cost, reducing degree of imbalance, increasing throughput and average resource utilization.
Some of the metrics are blocked by your
consent settings
Define Stance and Swing pattern of gait cycle using motion sensor and K-Mean Clustering
(2022-01-01)
Santikan, Piyapon
;
Tangwongcharoen, Wisan
;
Kimpan, Warangkhana
This research studied walking patterns based on the gait cycle focused on the stance and swing period. We are interested in creating the pattern for representing a normal person and person with a walking disorder by distinguishing patterns. This research uses Razor-IMU to collect all walking data and transfer data from sensors via WIFI which helps gain data to be stable and accurate.After collecting the walking data, we transformed data into linear graphs to reference the gait cycle pattern. Because the graph in linear form can show the movement and distinguish between normal and abnormal people the difference. The aim is to obtain representative data of normal and abnormal people for further analysis. Therefore, the data were then grouped using K-mean Clustering. The data obtained from the clusters were able to distinguish between normal and abnormal walking distances.
Some of the metrics are blocked by your
consent settings
Network Architecture of ETAT Education and Training Centers for Automation 4.0
(2022-01-01)
De Marchi, Matteo
;
Jitngernmadan, Prajaks
;
Singsri, Pongpat
;
Putpuek, Narongsak
;
Kumpakeaw, Saman
Automation 4.0 comprises the latest technologies that enhance classic automation to communicate with machines in a connected manner and to enable more flexible and more intelligent automation in smart factories. Industry 4.0 has become a global trend over the past ten years to combine the benefits of automation with the needs for flexible production. This trend has reached, in addition to industrialized countries, also newly industrialized countries like Thailand, with the aim to strengthen the global competitiveness. In the ETAT project, partners from Europe as well as Thailand combine their competences to build up a network of so-called Smart Labs around Bangkok area and thus take an important step towards Automation 4.0 in Thailand. Using Thailand as an example, this work shows how industrially emerging countries can sustainably strengthen their qualification with regard to Automation 4.0 by setting up such networks of education and training centers. In addition to explaining the general architecture of the network, the specific characteristics of all six established ETAT Smart Labs are presented and discussed.
Some of the metrics are blocked by your
consent settings
Thai food recommendation system using hybrid of particle swarm optimization and K-means algorithm
(2021-04-23)
Puraram, Tanakorn
;
Chaovalit, Pimwadee
;
Peethong, Apatha
;
Tiyanunti, Pongsak
;
Charoensiriwath, Supiya
A food recommendation system is an information filtering tool that helps suggest appropriate food menus to users based on their dietary behavior, nutrition, health, or activity. In this paper, a hybrid method of Particle Swarm Optimization (PSO) and K-Means algorithm is proposed to improve the user's dietary behavior clustering and using Principal Component Analysis (PCA) to reduce the data dimension. Moreover, the User-Based Collaborative Filtering technique is used to predict the rating of relevant Thai food menus and recommendation. The experimental result shows the hybrid method improves the clustering performance from 3 models: Hierarchical Clustering, K-Means, and K-Means with PCA, in terms of silhouette coefficient score. In addition, the hybrid method improves the Davies-Bouldin index score by 44%, 19%, and 17% compared to those models, respectively. The rating prediction result shows the hybrid method outperforms the other methods.
Some of the metrics are blocked by your
consent settings
Turbidity of Coconut Oil Determination Using the MAMoH Method in Image Processing
(2021-01-01)
Palananda, Attapon
;
Kimpan, Warangkhana
In general, considering standard production, as well as coconut oil production, in oil consumption industries is an important factor. Oil color is an important element, as it is an important factor for consumers or buyers in selecting coconut oil. In the process of producing coconut oil, the cold-pressed method has been chosen to maintain the essential quality of coconut oil. The quality of the coconut oil is inspected from the production process by means of light passing through the coconut oil. Then, the production staff compares the turbidity of coconut oil with the master sample. The turbidity of coconut oil in every production must be compared with a master sample to maintain standards control. According to previous studies, there are many methods for determining coconut oil turbidity. One method that has been utilized is determining turbidity from light passing through the medium in which the transmitted light can be absorbed through the turbidity of the variable medium. This process is applied together with image processing to determine the coconut oil turbidity. In this research, we propose a method for measuring coconut oil turbidity by the Moving Average Median of Hue (MAMoH), which is better in detecting the coconut oil turbidity than the Median of Gray Scale (MoGS) method, Median of Hue (MoH) method, and Random Position Median of Hue (RPMoH) method. In terms of the percentage accuracy of the efficiency test; the MAMoH method has 99 percent accuracy, while the MoGS method is not applicable, the MoH method has 88.04 percent accuracy, and the RPMoH method has 85.91 percent accuracy. Thus, the MAMoH method is considered an appropriate method for measuring coconut oil turbidity.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results