Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.015
S. Omar, B. Musa, Sulaiman Wadi Harun, N. Ahmed, N. Zulkipli, Z. Jusoh, Husna Abdul Rahman
108 Abstract— In this paper, Titanium Aluminium Cloride (Ti3AlC2) is used as a saturable absorber (SA) to generate Q-switched pulses in an erbium-doped fibre laser (EDFL) cavity. The Ti3AlC2 film was prepared based mixing the material with polyvinyl alcohol (PVA) solution. It was added into the laser cavity by sandwiching the film between two fiber ferrules. By controlling the cavity's loss and gain, a stable Q-switched operation was achieved. When the pumping power was increased from 58.88 to 97.62 mW, the repetition frequency was increased from 59.52 to 67.52 kHz, resulting in the shortest pulse width of 3.3 μs. The Q-switched EDFL operated at a centre wavelength of 1560 nm, had a maximum pulse energy of 138.76 nJ, and a slope efficiency of 9.36 percent. The results demonstrate that Ti3AlC2 can be used as a SA for realizing a laser device for various applications including optical communications.
{"title":"Q-switching in EDFL Cavity using Ti3AlC2 as Saturable Absorber","authors":"S. Omar, B. Musa, Sulaiman Wadi Harun, N. Ahmed, N. Zulkipli, Z. Jusoh, Husna Abdul Rahman","doi":"10.24191/jeesr.v20i1.015","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.015","url":null,"abstract":"108 Abstract— In this paper, Titanium Aluminium Cloride (Ti3AlC2) is used as a saturable absorber (SA) to generate Q-switched pulses in an erbium-doped fibre laser (EDFL) cavity. The Ti3AlC2 film was prepared based mixing the material with polyvinyl alcohol (PVA) solution. It was added into the laser cavity by sandwiching the film between two fiber ferrules. By controlling the cavity's loss and gain, a stable Q-switched operation was achieved. When the pumping power was increased from 58.88 to 97.62 mW, the repetition frequency was increased from 59.52 to 67.52 kHz, resulting in the shortest pulse width of 3.3 μs. The Q-switched EDFL operated at a centre wavelength of 1560 nm, had a maximum pulse energy of 138.76 nJ, and a slope efficiency of 9.36 percent. The results demonstrate that Ti3AlC2 can be used as a SA for realizing a laser device for various applications including optical communications.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121166094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.007
Najwa Ahmad Suhaimi, A. I. Mohd Yassin, M. S. A. Megat Ali, A. H. Jahidin
51 Abstract—Intelligence is defined as the mental ability to learn, reason and solve problems. Recently, studies have characterized the different levels of intelligence quotient from the resting brainwaves. Various filter designs and power ratio equations have been proposed, all with unique strengths and limitations. Hence, further investigation is required to standardize the pre-processing algorithm using the established electroencephalogram database. The previously established Hamming and equiripple filter designs are evaluated in this study. The later are more superior for filtering the electroencephalogram into the respective brainwaves. Despite the limitations, the low-order Hamming filters are still recommended as the memory required is only 6% of the highorder equiripple filters. These greatly enhance the computational efficiency. The cross-correlation function tests further revealed the impact of filter designs on the resultant brainwaves. Hence, a new set of power ratio equations have been successfully formulated for dataset validation.
{"title":"EEG-based IQ Pattern Analysis: Considerations on Filter Design and Power Ratio Equations","authors":"Najwa Ahmad Suhaimi, A. I. Mohd Yassin, M. S. A. Megat Ali, A. H. Jahidin","doi":"10.24191/jeesr.v20i1.007","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.007","url":null,"abstract":"51 Abstract—Intelligence is defined as the mental ability to learn, reason and solve problems. Recently, studies have characterized the different levels of intelligence quotient from the resting brainwaves. Various filter designs and power ratio equations have been proposed, all with unique strengths and limitations. Hence, further investigation is required to standardize the pre-processing algorithm using the established electroencephalogram database. The previously established Hamming and equiripple filter designs are evaluated in this study. The later are more superior for filtering the electroencephalogram into the respective brainwaves. Despite the limitations, the low-order Hamming filters are still recommended as the memory required is only 6% of the highorder equiripple filters. These greatly enhance the computational efficiency. The cross-correlation function tests further revealed the impact of filter designs on the resultant brainwaves. Hence, a new set of power ratio equations have been successfully formulated for dataset validation.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121019395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.011
A. B. Roslan, R. L. A. Shauri
{"title":"Object Texture Recognition Based on Grasping Force Data Using Feedforward Neural Network","authors":"A. B. Roslan, R. L. A. Shauri","doi":"10.24191/jeesr.v20i1.011","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.011","url":null,"abstract":"","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127474950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.002
Siti Maisarah Zainorzuli, S. A. Che Abdullah, H. Zainol Abidin, Fazlina Ahmat Ruslan
11 Abstract— Deep Learning is an Artificial Intelligence (AI) function which can imitate the human brain to process data and deciding. It has networks that able to learn the unsupervised data that unlabeled or unstructured. It also identified as Deep Neural Network or Deep Neural Learning. Convolutional Neural Network (CNN) is a subset of Deep Neural Network which frequently used to analyse images. CNN also called as ConvNet which can be trained using an existing model that has been finetuned or trained from zero by using a large data set. CNN was often used in image classification due to its effectiveness and accuracy. However, there are several CNN architectures such as AlexNet, GoogleNet and ResNet-50. To select the appropriate architecture for our research in agriculture, a preliminary study to evaluate the architecture were conducted by using five different types of flower datasets that obtained from Matlab and Kaggle database. The three types of CNN architecture were compared in terms of accuracy in classifying the flowers. Result of this study indicated that the optimal configuration is by setting the number of epochs at 30, with the learning rate at 0.0005, to obtain the highest accuracy at 99.82%.
{"title":"Comparison Study on Convolution Neural Network (CNN) Techniques for Image Classification","authors":"Siti Maisarah Zainorzuli, S. A. Che Abdullah, H. Zainol Abidin, Fazlina Ahmat Ruslan","doi":"10.24191/jeesr.v20i1.002","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.002","url":null,"abstract":"11 Abstract— Deep Learning is an Artificial Intelligence (AI) function which can imitate the human brain to process data and deciding. It has networks that able to learn the unsupervised data that unlabeled or unstructured. It also identified as Deep Neural Network or Deep Neural Learning. Convolutional Neural Network (CNN) is a subset of Deep Neural Network which frequently used to analyse images. CNN also called as ConvNet which can be trained using an existing model that has been finetuned or trained from zero by using a large data set. CNN was often used in image classification due to its effectiveness and accuracy. However, there are several CNN architectures such as AlexNet, GoogleNet and ResNet-50. To select the appropriate architecture for our research in agriculture, a preliminary study to evaluate the architecture were conducted by using five different types of flower datasets that obtained from Matlab and Kaggle database. The three types of CNN architecture were compared in terms of accuracy in classifying the flowers. Result of this study indicated that the optimal configuration is by setting the number of epochs at 30, with the learning rate at 0.0005, to obtain the highest accuracy at 99.82%.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124893288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.014
Azlina Ahmad Khan, N. Zainal., M. Mokhtar, M. H. Mamat, M. H. Abdullah
102 Abstract— Dye-sensitized solar cell (DSSC) has been actively researched due to its lower fabrication cost. In DSSC, sensitizer imitates photosynthesis process in plants by absorbing the photon energy from the sunlight. This research investigated methanolic extracted chlorophyll dye sensitizer in comparison to water extracted and bare chlorophyll. Density Functional Theory (DFT) and Time-Dependent Density Functional Theory (TD-DFT) methods in Gaussian 03/Gaussview 03 software were used to investigate its optical and electronic properties. Each of the molecular structure (methanol and water) was individually attached to chlorophyll, and their performances were determined. Results demonstrated that the absorbance spectrum of chlorophyll shows a broad peak in the visible range from 400 to 700 nm with its highest peak at 430 nm. Energy gap of chll-m between the E(LUMO) and the conduction band of TiO2 semiconductor, EC(TiO2) or so called as VOC is slightly smaller compared to water (chll-w). Meanwhile, the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) bandgap of chll-m is slightly higher at 2.13 eV. Finally, position of LUMO to redox electrolyte in chll-m is the lowest, which is favourable for faster electron transition. From observation, chll-m demonstrated better optical and electronic properties compared to its counterparts and hence good for chlorophyll sensitizer extraction.
{"title":"Properties of Chlorophyll Based Organic Dye Sensitizer Molecule Extracted in Various Solvents","authors":"Azlina Ahmad Khan, N. Zainal., M. Mokhtar, M. H. Mamat, M. H. Abdullah","doi":"10.24191/jeesr.v20i1.014","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.014","url":null,"abstract":"102 Abstract— Dye-sensitized solar cell (DSSC) has been actively researched due to its lower fabrication cost. In DSSC, sensitizer imitates photosynthesis process in plants by absorbing the photon energy from the sunlight. This research investigated methanolic extracted chlorophyll dye sensitizer in comparison to water extracted and bare chlorophyll. Density Functional Theory (DFT) and Time-Dependent Density Functional Theory (TD-DFT) methods in Gaussian 03/Gaussview 03 software were used to investigate its optical and electronic properties. Each of the molecular structure (methanol and water) was individually attached to chlorophyll, and their performances were determined. Results demonstrated that the absorbance spectrum of chlorophyll shows a broad peak in the visible range from 400 to 700 nm with its highest peak at 430 nm. Energy gap of chll-m between the E(LUMO) and the conduction band of TiO2 semiconductor, EC(TiO2) or so called as VOC is slightly smaller compared to water (chll-w). Meanwhile, the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) bandgap of chll-m is slightly higher at 2.13 eV. Finally, position of LUMO to redox electrolyte in chll-m is the lowest, which is favourable for faster electron transition. From observation, chll-m demonstrated better optical and electronic properties compared to its counterparts and hence good for chlorophyll sensitizer extraction.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131649722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.006
M. Islam, H. Mohamad, S. Z. Mohammad Noor
41 Abstract— Solar Home System (SHS) is a stand-alone power solution in isolated places where there is no connectivity to the national power grid. Currently, an off-grid SHS has a solar panel, a lead-acid battery, a Pulse Width Modulation (PWM) solar charge controller, and 12V DC power operated electrical home appliances; in some cases, a DC to DC converter or a DC to AC inverter has been connected. So far, various SHS packages have been introduced, each with a set of product features and prices. Since the majority of SHS are small in size, producing energy is inadequate to satisfy the customer's demand. This study introduces a new feature that incorporates both analog PWM solar charging techniques and a DC to DC (12V to 120V) converter to satisfy the needs of existing off-grid SHS users. The design of solar charging includes improved charging techniques i.e surge voltage and lightning protection. The result produced is a noisefree switching gate drive signal and high-performance fixedfrequency current mode controllers. The new controller's design also reduces the number of devices in the SHS; thus reduces the controller's total production cost.
{"title":"Development of a New Controller for Solar Home System: PWM Charge Controller & DC to DC Converter (12V to 120V)","authors":"M. Islam, H. Mohamad, S. Z. Mohammad Noor","doi":"10.24191/jeesr.v20i1.006","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.006","url":null,"abstract":"41 Abstract— Solar Home System (SHS) is a stand-alone power solution in isolated places where there is no connectivity to the national power grid. Currently, an off-grid SHS has a solar panel, a lead-acid battery, a Pulse Width Modulation (PWM) solar charge controller, and 12V DC power operated electrical home appliances; in some cases, a DC to DC converter or a DC to AC inverter has been connected. So far, various SHS packages have been introduced, each with a set of product features and prices. Since the majority of SHS are small in size, producing energy is inadequate to satisfy the customer's demand. This study introduces a new feature that incorporates both analog PWM solar charging techniques and a DC to DC (12V to 120V) converter to satisfy the needs of existing off-grid SHS users. The design of solar charging includes improved charging techniques i.e surge voltage and lightning protection. The result produced is a noisefree switching gate drive signal and high-performance fixedfrequency current mode controllers. The new controller's design also reduces the number of devices in the SHS; thus reduces the controller's total production cost.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116359646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.012
N. F. Naim, M. F. Ibrahim, S. S. Sarnin, Norsuzila Ya’acob
This paper presents on the design of multiwavelength laser for hybrid Raman-erbium doped fiber gains in three different cavities; single pass linear, bidirectional linear and ring laser. Multiwavelength laser has the potential for its simplicity and lower in cost. The 50% arm coupling ratio of ring laser structure with 0.2 nm line width, 20 lasing line and 5 mW lasing threshold is achieved and shown to be the best result for multiwavelength laser design compared to the single pass and bidirectional linear cavity. With an input power of 240 mW (EDFA) and 2000 mW (FRA), a 96% power conversion efficiency is obtained and produced a stable multiwavelength for L-band source with peak power of -10 dBm for ring laser, -25 dBm for single pass linear and -5 dBm for bidirectional linear laser. The lasing threshold obtained at 50% arm coupling ratio for ring laser is also slightly lower than other value reported previously. The interesting features of the proposed laser systems is the low threshold power, high output power, high number of lasing line and high pump conversion efficiency.
{"title":"Performance of Multiwavelength Laser Utilizing Fabry-Pérot Interferometer in Different Cavity Configuration using Hybrid Raman-EDF Gains","authors":"N. F. Naim, M. F. Ibrahim, S. S. Sarnin, Norsuzila Ya’acob","doi":"10.24191/jeesr.v20i1.012","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.012","url":null,"abstract":"This paper presents on the design of multiwavelength laser for hybrid Raman-erbium doped fiber gains in three different cavities; single pass linear, bidirectional linear and ring laser. Multiwavelength laser has the potential for its simplicity and lower in cost. The 50% arm coupling ratio of ring laser structure with 0.2 nm line width, 20 lasing line and 5 mW lasing threshold is achieved and shown to be the best result for multiwavelength laser design compared to the single pass and bidirectional linear cavity. With an input power of 240 mW (EDFA) and 2000 mW (FRA), a 96% power conversion efficiency is obtained and produced a stable multiwavelength for L-band source with peak power of -10 dBm for ring laser, -25 dBm for single pass linear and -5 dBm for bidirectional linear laser. The lasing threshold obtained at 50% arm coupling ratio for ring laser is also slightly lower than other value reported previously. The interesting features of the proposed laser systems is the low threshold power, high output power, high number of lasing line and high pump conversion efficiency.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114792778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.001
M. F. Khyasudeen, N. Buniyamin
1 Abstract—In this Covid -19 pandemic era, where the negative impact of the virus is felt worldwide, it is crucial to reduce the rate of infection. This paper presents the development of a vehicle to minimize the spread of the disease. The Coronavirus-Autonomous Ground Vehicle (COR-AGV) uses Ultraviolet Light (UV) rather than liquid sanitiser. The use of UV light as a disinfectant is proven in the medical field, where many medical appliances were disinfected using UV light without the need for alcohol substances. Previous research showed thatsimilarCoronavirus such as SARS and MERS lost their ability to infect when exposed to the UV for a particular durationasthe exposure causeddamageto the virus's DNA sequence. The COR-AGV presented herewith uses UV Light to carry out the disinfection task autonomously. The COR-AGV, equipped with a far-UVC light array, navigates autonomously within the targeted areas to sweep systematically, scan and disinfect. The COR-AGV is equipped with camera and sonar sensors to enable path planning and obstacle avoidance in confined spaces and outdoor environments. A GPS is included to assist the COR-AGV navigation system. Simulation test runs were performed to check the adequacy of the rover in performing disinfection. The COR-AGV perform disinfection on the wall surface with more than 50% of the wall surface were covered for disinfection in various floor layouts, while being able to perform obstacle avoidance within the test area. Thus, COR-AGV can be adapted prior to the actual pandemic event to enhance its level of performance. The performance of COR-AGV can be improved with additional sensors such as Computer Vision or LiDAR sensor.
{"title":"Autonomous Ground Vehicle (COR-AGV) Disinfectant System Using Far-UVC Light Exposure","authors":"M. F. Khyasudeen, N. Buniyamin","doi":"10.24191/jeesr.v20i1.001","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.001","url":null,"abstract":"1 Abstract—In this Covid -19 pandemic era, where the negative impact of the virus is felt worldwide, it is crucial to reduce the rate of infection. This paper presents the development of a vehicle to minimize the spread of the disease. The Coronavirus-Autonomous Ground Vehicle (COR-AGV) uses Ultraviolet Light (UV) rather than liquid sanitiser. The use of UV light as a disinfectant is proven in the medical field, where many medical appliances were disinfected using UV light without the need for alcohol substances. Previous research showed thatsimilarCoronavirus such as SARS and MERS lost their ability to infect when exposed to the UV for a particular durationasthe exposure causeddamageto the virus's DNA sequence. The COR-AGV presented herewith uses UV Light to carry out the disinfection task autonomously. The COR-AGV, equipped with a far-UVC light array, navigates autonomously within the targeted areas to sweep systematically, scan and disinfect. The COR-AGV is equipped with camera and sonar sensors to enable path planning and obstacle avoidance in confined spaces and outdoor environments. A GPS is included to assist the COR-AGV navigation system. Simulation test runs were performed to check the adequacy of the rover in performing disinfection. The COR-AGV perform disinfection on the wall surface with more than 50% of the wall surface were covered for disinfection in various floor layouts, while being able to perform obstacle avoidance within the test area. Thus, COR-AGV can be adapted prior to the actual pandemic event to enhance its level of performance. The performance of COR-AGV can be improved with additional sensors such as Computer Vision or LiDAR sensor.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122928336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.005
Nurul Farhana Mohamad Zamri, N. Md. Tahir, M. S. A. Megat Ali, Nur Dalila Khirul Ashar
34 Abstract— Learning algorithms related to deep learning use bells and whistles, called hyperparameters. Hence, this study conducted numerical analysis, specifically backpropagation gradients and gradient-based optimization for snatch-theft detection. Here, snatch theft images and augmented images were used to perform the experimental study to determine the optimum hyperparameter values. Next, the value of epoch and learning rate was obtained after careful analysis based on each training option. Results achieved showed that epoch value of 20 and learning rate corresponding to 0.0001 was the optimum values. Findings from this study can be used as a practical guide in determining the importance of the most optimum hyperparameters.
{"title":"Deep Learning Optimisation Algorithms for Snatch Theft Detection","authors":"Nurul Farhana Mohamad Zamri, N. Md. Tahir, M. S. A. Megat Ali, Nur Dalila Khirul Ashar","doi":"10.24191/jeesr.v20i1.005","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.005","url":null,"abstract":"34 Abstract— Learning algorithms related to deep learning use bells and whistles, called hyperparameters. Hence, this study conducted numerical analysis, specifically backpropagation gradients and gradient-based optimization for snatch-theft detection. Here, snatch theft images and augmented images were used to perform the experimental study to determine the optimum hyperparameter values. Next, the value of epoch and learning rate was obtained after careful analysis based on each training option. Results achieved showed that epoch value of 20 and learning rate corresponding to 0.0001 was the optimum values. Findings from this study can be used as a practical guide in determining the importance of the most optimum hyperparameters.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114033649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-01DOI: 10.24191/jeesr.v20i1.009
Ummu Umairah Mohamad Ali, Z. Jusoh, Md Ashadi Md Johari, Husna Abdul Rahman, Huda Adnan Zain, N. Zulkipli, Sulaiman Wadi Harun
66 Abstract—In this paper, the effect of microfibers diameter on the micro-sphere resonators (MSR) ability to detect relative humidity changes is analyzed. Microspheres show whispering gallery mode (WGM) characteristics when light is properly coupled into them. The interaction between that coupled light and the surroundings can be used for sensing applications. Three different diameters of microfiber were used to couple light into the MSR, 5 μm, 8 μm and 10 μm. The MSR with the sphere diameter of 200 μm was categorized by its quality factor and spectrum transmission modes. Then, the structure was used as a humidity sensor. The transmission modes of the MSR were compared at three different tapering sizes. Each of these structures was tested as a humidity sensor in the relative humidity range 50-90%. In addition, sensitivity, and linearity of the three proposed sensors were calculated. According to the compared results, the 5 μm microfiber has a much better sensitivity than the 8 μm and 10 μm microfiber when using the MSR.
{"title":"Effect of Microfiber Diameters on Micro-sphere Resonator Based Humidity Sensor","authors":"Ummu Umairah Mohamad Ali, Z. Jusoh, Md Ashadi Md Johari, Husna Abdul Rahman, Huda Adnan Zain, N. Zulkipli, Sulaiman Wadi Harun","doi":"10.24191/jeesr.v20i1.009","DOIUrl":"https://doi.org/10.24191/jeesr.v20i1.009","url":null,"abstract":"66 Abstract—In this paper, the effect of microfibers diameter on the micro-sphere resonators (MSR) ability to detect relative humidity changes is analyzed. Microspheres show whispering gallery mode (WGM) characteristics when light is properly coupled into them. The interaction between that coupled light and the surroundings can be used for sensing applications. Three different diameters of microfiber were used to couple light into the MSR, 5 μm, 8 μm and 10 μm. The MSR with the sphere diameter of 200 μm was categorized by its quality factor and spectrum transmission modes. Then, the structure was used as a humidity sensor. The transmission modes of the MSR were compared at three different tapering sizes. Each of these structures was tested as a humidity sensor in the relative humidity range 50-90%. In addition, sensitivity, and linearity of the three proposed sensors were calculated. According to the compared results, the 5 μm microfiber has a much better sensitivity than the 8 μm and 10 μm microfiber when using the MSR.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133473912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：