Fathinul Syahir Ahmad Sa'ad
Thumbnail Image
Preferred name
Fathinul Syahir Ahmad Sa'ad
Official Name
Fathinul Syahir, Ahmad Sa'ad
Alternative Name
Ahmad Saad, Fathinul Syahir
Sa’ad, Fathinul Syahir Ahmad
Main Affiliation
Scopus Author ID
57205307496
Researcher ID
R-5360-2019

Research Output

Filters

20
Reset filters

Settings
Now showing 1 - 10 of 20
  • Publication
    Segmentation of Diabetic Retinopathy Using Entropy-Based Thresholding - A Review
    ( 2022-01-01)
    Qaid M.S.A.
    ;
    Shafriza Nisha Basah
    ;
    Haniza Yazid
    ;
    Fathinul Syahir Ahmad Sa'ad
    Synthetic data by various algorithms that resemble actual data in terms of statistical features. Computer-aided medical applications have been extensively applied to model specific scenarios, such as medical imaging of retinal images for diabetic retinopathy (DR) detection. The available data and annotated medical data are typically rare and costly due to the difficulties of conducting medical screening and rely on highly trained doctors to review and diagnose. The modelling of retinal images for DR analysis is essential since it will provide a model to guide and test DR detection algorithms. This paper aims to model normal retina and non-proliferative diabetic retinopathy (NPDR) stages (mild, moderate, and severe) data models with the variation of dynamic models. The Digital Retinal Images for Vessel Extraction (DRIVE), The Standard Diabetic Retinopathy Database, Calibration Level 1 (DIARETDB1), and E-OPHTHA datasets are analyzed to obtain the specification of the human retina and DR lesions. In the data modelling phases, the model includes the bright and dark retinal lesions with the variation of dynamic parameters. 4100 synthetic images are used where 200 normal images and 3900 NPDR images to test the performance of DR detection algorithms over the full range of parameters.
  • Publication
    Performance analysis of multi-level thresholding for microaneurysm detection
    ( 2022-09-01)
    Choong K.H.
    ;
    Shafriza Nisha Basah
    ;
    Haniza Yazid
    ;
    Muhammad Juhairi Aziz Safar
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Lim Chee Chin
    Diabetic retinopathy (DR) – one of the diabetes complications – is the leading cause of blindness among the age group of 20–74 years old. Fortunately, 90% of these cases (blindness due to DR) could be prevented by early detection and treatment via manual and regular screening by qualified physicians. The screening of DR is tedious, which can be subjective, time-consuming, and sometimes prone to misclassification. In terms of accuracy and time, many automated screening systems based on image processing have been developed to improve diagnostic performance. However, the accuracy and consistency of the developed systems are largely unaddressed, where a manual screening process is still the most preferred option. The main contribution of this paper is to analyse the accuracy and consistency of microaneurysm (MA) detection via image processing by focusing on Otsu’s multi-thresholding as it has been shown to work very well in many applications. The analysis was based on Monte Carlo statistical analysis using synthetic retinal images of retinal images under variation of all stages of DR, retinal, and image parameters – intensity difference between MAs and blood vessels (BVs), MA size, and measurement noise. Then, the conditions – in terms of obtainable retinal and image parameters – that guarantee accurate and consistent MA detection via image processing were extracted. Finally, the validity of the conditions to guarantee accurate and consistent MA detection was verified using real retinal images. The results showed that MA detection via image processing is guaranteed to be accurate and consistent when the intensity difference between MAs and BVs is at least 50% and the sizes of MAs are from 5 to 20 pixels depending on measurement noise values. These conditions are very important as a guideline of MA detection for DR.
  • Publication
    Analysis on Clustering Based Method for Diabetic Retinopathy Using Color Information
    ( 2022-01-01)
    Selvam S.A
    ;
    Haniza Yazid
    ;
    Shafriza Nisha Basah
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Muhamad Khairul Ali Hassan
    Diabetic Retinopathy (DR) is an important global health concern and it can causes blindness. Early detection and treatment can prevent the patients from loss their vision. This study presents an approach of color image segmentation for automatic exudate detection. The color retinal images are converted into four different color spaces and preprocessed by applying Contrast Limited Adaptive Histogram Equalization (CLAHE). Fuzzy C-Means (FCM) and K-means clustering (KMC) algorithms are applied on the preprocessed image for the segmentation purpose. Then, optic disc is detected and eliminated by using Circular Hough Transform (CHT). Performance evaluation of developed algorithm is done using Structured Analysis of the Retina (STARE) dataset. The proposed algorithm achieved sensitivity of 93.4% for STARE datasets for LUV color space with KMC.
  • Publication
    Cloud-based System for University Laboratories Air Monitoring
    ( 2020-09-21)
    Abu Hassan Abdullah
    ;
    Sukhairi Sudin
    ;
    Mustafa M.H.
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Khairul Azwan Ismail
    ;
    Muhammad Aizat Abu Bakar
    ;
    Mohamed Elshaikh Elobaid Said Ahmed
    ;
    Abdul Ghapar Ahmad
    ;
    Zahari Awang Ahmad
    ;
    Sara Yasina Yusuf
    Indoor air such as house, shopping complex, hospital, university, office and hotel should be monitor for human safety and wellbeing. These closed areas are prone to harmful air pollutants i.e. allergens, smoke, mold, particles radon and hazardous gas. Laboratories in university are special room in which workers (student, technician, teaching/research assistants, researcher and lecturer) conduct their works and experiment. The activities and the environment will generate specific air pollutant which concentration depending to their parameters. Anyone in the environment that exposure to these pollutants may affect safety and health issue. This paper proposes a study of development of a cloud-based electronic nose system for university laboratories air monitoring. The system consists of DSP33-based electronic nose (e-nose) as nodes which measure main indoor air pollutant along with two thermal comfort variables, temperature and relative humidity. The e-noses are placed at five different laboratories for acquiring data in real time. The data will be sent to a web server and the cloud-based system will process, analyse using Neuro-Fuzzy classifier and display on a website in real time. The system will monitor the laboratories air pollutants and thermal comfort by predict the pollutant concentration and dispersion in the area i.e. Air Pollution Index (API). In case of air hazard safety (e.g., gas spills detection and pollution monitoring), the system will alert the security by activate an alarm and through e-mail. The website will display the API of the area in real-time. Results show that the system performance is good and can be used to monitor the air pollutant in the university laboratories.
  • Publication
    Harumanis mango quality assessments technique based on high level features fusion of infra-red thermal and optical image
    ( 2017)
    Fathinul Syahir Ahmad Sa'ad
    Mangoes imported from other parts of the world, especially Malaysia, Thailand, Mexico and the Philippines, are usually available all year round but in Perlis, Malaysia there is one unique and famous mango is Harumanis mango and this fruit is seasonal. Every year, a large amount of mangoes are produced and need to be evaluated for quality assessments. Presently, the quality inspection was done manually by the quality expert as there are no automated grading system is available. Hence, by automating the procedure as well as developing new classification technique, it may solve these problems. This thesis presents the new method on the high level features fusion of visible and IR Thermal Image features for mango quality assessment. A shape and weight analysis was developed from visible imaging and a maturity analysis was developed from IR thermal imaging. A Fourier-Descriptor method was developed to grade mango by its shape and a cylinder analysis method was used to grade Harumanis mango by its weight and it give different accuracy result of classification. The spectrum of infrared image was used to distinguish and classify the level of maturity of the fruits and it gave low accuracy compare to shape and weight classification. To get high accuracy for quality assessment for Harumanis mango, high level data fusion was proposed. This method combined all three classifier of shape, weight and maturity and it was found to be able to achieve 98% accuracy classification.
  • Publication
    Development of a Common Waste Combustion System for Generating Electricity at Remote Area
    ( 2023-01-01)
    Zaini A.S.A.
    ;
    Sukhairi Sudin
    ;
    Haris F.A.
    ;
    Nurul Syahirah Khalid
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Muhamad Khairul Ali Hassan
    ;
    Ibrahim I.I.
    Malaysia's daily amount of municipal solid waste (MSW) has rapidly increased. This causes the landfills number to increase due to inadequate waste management systems. Apart from that, Malaysia depends on non-renewable resources for electricity generation which could have a significant effect on the environment. Therefore, this study proposed to reduce landfills in Malaysia in a proper way and supply electricity using municipal solid waste as a renewable resource. In this study, the combustion of municipal solid waste (MSW) produces steam, which will rotate a turbine that is connected to a dynamo. Then, the energized dynamo will supply electricity to appliances including a direct current motor. The motor shaft then rotates the dynamo shaft in the pulley system which causes the electricity to flow in a closed loop. In this system, a pressurized container is crucial to produce sufficient steam. Based on the experimental setup, it was observed that continuous electricity was successfully achieved by looping the system using a pulley on the dynamo and motor.
      1
  • Publication
    Holonomic Mobile Robot Planners: Performance Analysis
    ( 2022-01-01)
    Aljamali Y.S.
    ;
    Muhammad Juhairi Aziz Safar
    ;
    Khairul Salleh Basaruddin
    ;
    Yazid H.
    ;
    Basha S.N.
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Hassan M.K.A.
    Many algorithms have been proposed to tackle the path planning problem in mobile robots. Among the well-known and established algorithms are the Probabilistic Road Map (PRM) algorithm, A* algorithm, Genetic algorithm (GA), Rapidly-exploring random tree (RRT), and dual Rapidly-exploring random trees (RRT-connect). Hence, this paper will focus on the performance comparison between the aforementioned algorithms concerning computation time, path length, and fail and success rate for producing a path. For the sake of fair and conclusive results, simulation is conducted in two phases with four different environments, namely, free space environment, low cluttered environment, medium cluttered environment, and high cluttered environment. The results show that RRT-connect has a high success rate in producing a feasible path with the least computation time. Hence, RRTs-based sampling algorithms, in general, and RRT-connect, in specific, will be explored in-depth for possible optimization.
      2
  • Publication
    Deep CNN-based Planthopper classification using a high-density image dataset
    (MDPI, 2023)
    Mohd Firdaus Ibrahim
    ;
    Siti Khairunniza-Bejo
    ;
    Marsyita Hanafi
    ;
    Mahirah Jahari
    ;
    Mohammad Aufa Mhd Bookeri
    ;
    Fathinul Syahir Ahmad Sa'ad
    Rice serves as the primary food source for nearly half of the global population, with Asia accounting for approximately 90% of rice production worldwide. However, rice farming faces significant losses due to pest attacks. To prevent pest infestations, it is crucial to apply appropriate pesticides specific to the type of pest in the field. Traditionally, pest identification and counting have been performed manually using sticky light traps, but this process is time-consuming. In this study, a machine vision system was developed using a dataset of 7328 high-density images (1229 pixels per centimetre) of planthoppers collected in the field using sticky light traps. The dataset included four planthopper classes: brown planthopper (BPH), green leafhopper (GLH), white-backed planthopper (WBPH), and zigzag leafhopper (ZIGZAG). Five deep CNN models—ResNet-50, ResNet-101, ResNet-152, VGG-16, and VGG-19—were applied and tuned to classify the planthopper species. The experimental results indicated that the ResNet-50 model performed the best overall, achieving average values of 97.28% for accuracy, 92.05% for precision, 94.47% for recall, and 93.07% for the F1-score. In conclusion, this study successfully classified planthopper classes with excellent performance by utilising deep CNN architectures on a high-density image dataset. This capability has the potential to serve as a tool for classifying and counting planthopper samples collected using light traps.
      11  3
  • Publication
    Cycling performance prediction based on cadence analysis by using multiple regression
    ( 2021-12-01)
    Sukhairi Sudin
    ;
    Aziz Naim Abdul Aziz
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Nurul Syahirah Khalid
    ;
    Ismail Ishaq Ibrahim
    This project examined the influence of the cadence, speed, heart rate and power towards the cycling performance by using Garmin Edge 1000.Any change in cadence will affect the speed, heart rate and power of the novice cyclist and the changes pattern will be observed through mobile devices installed with Garmin Connect application. Every results will be recorded for the next task which analysis the collected data by using machine learning algorithm which is Regression analysis. Regression analysis is a statistical method for modelling the connection between one or more independent variables and a dependent (target) variable. Regression analysis is required to answer these types of prediction problems in machine learning. Regression is a supervised learning technique that aids in the discovery of variable correlations and allows for the prediction of a continuous output variable based on one or more predictor variables. A total of forty days' worth of events were captured in the dataset. Cadence act as dependent variable, (y) while speed, heart rate and power act as independent variable, (x) in prediction of the cycling performance. Simple linear regression is defined as linear regression with only one input variable (x). When there are several input variables, the linear regression is referred to as multiple linear regression. The research uses a linear regression technique to predict cycling performance based on cadence analysis. The linear regression algorithm reveals a linear relationship between a dependent (y) variable and one or more independent (y) variables, thus the name. Because linear regression reveals a linear relationship, it determines how the value of the dependent variable changes as the value of the independent variable changes. This analysis use the Mean Squared Error (MSE) expense function for Linear Regression, which is the average of squared errors between expected and real values. Value of R squared had been recorded in this project. A low R-squared value means that the independent variable is not describing any of the difference in the dependent variable-regardless of variable importance, this is letting know that the defined independent variable, although meaningful, is not responsible for much of the variance in the dependent variable's mean. By using multiple regression, the value of R-squared in this project is acceptable because over than 0.7 and as known this project based on human behaviour and usually the R-squared value hardly to have more than 0.3 if involve human factor but in this project the R-squared is acceptable.
      1
  • Publication
    Effect of Image Thresholding on the Homogenized Properties of Trabecular Bone Model
    ( 2022-01-01)
    Khairul Salleh Basaruddin
    ;
    Haniza Yazid
    ;
    Muhammad Juhairi Aziz Safar
    ;
    Mohd Hafif Basha Mohamad Jamel Basha
    ;
    Fathinul Syahir Ahmad Sa'ad
    ;
    Muhamad Khairul Ali Hassan
    This paper presents a numerical study to determine the homogenized (apparent) properties of vertebral trabecular bone with different threshold values using homogenization method. Series set of micro-CT images of vertebral trabecular bone was used in the present digital image-based modeling technique to reconstruct the microstructure model. Three image thresholding values were selected based on Otsu’s method. The homogenized properties that include the Young’s moduli, Poisson’s ratio and shear moduli was obtained in this study. The results showed there is significant effect of image threshold on the homogenized properties of vertebral trabecular bone model.
      2  19