Pub Date : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157319
V. Asha, Binju Saju, Singh Navnit Dhirendra, Yuvraj Kaswan, Prajwal G C, S. Sreeja
One way to hike consumer satisfies the services of the company provides is through an use of the customer relationship management (CRM) system. It can be difficult to determine the proper info what customer requires from data in your CRM system. Businesses can use data mining processes to segment and retrieve important customer information. Basis of consumer's RFM (Recency, Frequency, and Monetary) score, we can classify the customer segmentation. The RFM model has been utilised as the foundation for client segmentation in a number of research. However, the approaches suggested in earlier research are extremely particular to particular businesses, the score of RFM range employed as likewise more arbitrary. Additionally, is organizations grow, problems arise with RFM scoring. Measurements of RFM scores require periodic corrections, and current techniques make these corrections difficult. Determine a correct RFM score range, this study provided a unique technique that used a combination of K-Means and the Davies-Bouldin Index (DBI), circumventing the shortcomings of previous methods. As the amount of data rises, the suggested technique makes it easier to calculate RMF ratings. This is based on research conducted in the telecom industry. The K-Means method used in this study also produced the correct RFM score range which is depended on the ideal K values of the K-Means algorithm. The proposed solution only depends on each customer's RFM value from the corresponding data, so it can be used in different industries.
提高消费者对公司提供的服务的满意度的一种方法是使用客户关系管理(CRM)系统。很难从CRM系统中的数据中确定客户需要的适当信息。企业可以使用数据挖掘过程来分割和检索重要的客户信息。根据消费者的RFM (recent, Frequency, and Monetary)得分，我们可以对客户细分进行分类。在许多研究中，RFM模型已被用作客户细分的基础。然而，在早期的研究中提出的方法是非常特定于特定的业务，RFM范围的得分同样是比较武断的。此外，随着组织的发展，RFM评分也会出现问题。RFM分数的测量需要定期修正，而当前的技术使这些修正变得困难。为了确定正确的RFM评分范围，本研究提供了一种独特的技术，将K-Means和Davies-Bouldin指数(DBI)结合使用，避免了以往方法的缺点。随着数据量的增加，建议的技术使计算RMF评级变得更容易。这是基于在电信行业进行的研究。本研究中使用的K- means方法也产生了正确的RFM评分范围，该范围取决于K- means算法的理想K值。所提出的解决方案仅依赖于每个客户对应数据中的RFM值，因此可以在不同的行业中使用。
{"title":"Machine Learning based prototype for Customer Segmentation using RFM","authors":"V. Asha, Binju Saju, Singh Navnit Dhirendra, Yuvraj Kaswan, Prajwal G C, S. Sreeja","doi":"10.1109/ICEEICT56924.2023.10157319","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157319","url":null,"abstract":"One way to hike consumer satisfies the services of the company provides is through an use of the customer relationship management (CRM) system. It can be difficult to determine the proper info what customer requires from data in your CRM system. Businesses can use data mining processes to segment and retrieve important customer information. Basis of consumer's RFM (Recency, Frequency, and Monetary) score, we can classify the customer segmentation. The RFM model has been utilised as the foundation for client segmentation in a number of research. However, the approaches suggested in earlier research are extremely particular to particular businesses, the score of RFM range employed as likewise more arbitrary. Additionally, is organizations grow, problems arise with RFM scoring. Measurements of RFM scores require periodic corrections, and current techniques make these corrections difficult. Determine a correct RFM score range, this study provided a unique technique that used a combination of K-Means and the Davies-Bouldin Index (DBI), circumventing the shortcomings of previous methods. As the amount of data rises, the suggested technique makes it easier to calculate RMF ratings. This is based on research conducted in the telecom industry. The K-Means method used in this study also produced the correct RFM score range which is depended on the ideal K values of the K-Means algorithm. The proposed solution only depends on each customer's RFM value from the corresponding data, so it can be used in different industries.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115187085","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157099
M. Maheswari, A. Aloysius, P. Purusothaman
The widespread adoption of electronic health record (EHR) systems in response to a diverse array of requirements for primary and secondary healthcare, there is now an abundance of clinical data that can be accessed with relative ease. This is a significant improvement over the situation that existed previously. The widespread implementation of EHR systems is directly responsible for this effect. Unstructured clinical reports that were either transcribed or dictated by clinicians make up a sizeable percentage of these data, and they were collected in that format. In this paper, we develop an ensemble model to classify cancer disease from EHR using several convolutional neural network (CNN). The simulation is conducted to test the efficacy of the model and the results show that the proposed method achieves higher classification rate than other methods.
{"title":"Ensemble Deep Learning Classifier for Optimal Detection of Melanoma Cancer","authors":"M. Maheswari, A. Aloysius, P. Purusothaman","doi":"10.1109/ICEEICT56924.2023.10157099","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157099","url":null,"abstract":"The widespread adoption of electronic health record (EHR) systems in response to a diverse array of requirements for primary and secondary healthcare, there is now an abundance of clinical data that can be accessed with relative ease. This is a significant improvement over the situation that existed previously. The widespread implementation of EHR systems is directly responsible for this effect. Unstructured clinical reports that were either transcribed or dictated by clinicians make up a sizeable percentage of these data, and they were collected in that format. In this paper, we develop an ensemble model to classify cancer disease from EHR using several convolutional neural network (CNN). The simulation is conducted to test the efficacy of the model and the results show that the proposed method achieves higher classification rate than other methods.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114758848","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157086
S. Bhatlawande, Dhawal Khapre, Akshay Khare, S. Shilaskar
This paper presents an Electromyography (EMG) based system for classification of sitting and standing postures. The posture is classified by a machine learning model applied on the lower limb EMG data of the user. The dataset is collected from eight subjects, each with 8000 samples per channel, where six are used for training and two for testing. Time-domain, frequency-domain, and time-frequency-domain features are extracted for classification of sitting and standing postures. An array of algorithms are used for classification. Among all the classifiers Random Forest provided the highest accuracy at 98.38%.
{"title":"An Electromyography Based Intelligent System for Classification of Sitting and Standing Posture","authors":"S. Bhatlawande, Dhawal Khapre, Akshay Khare, S. Shilaskar","doi":"10.1109/ICEEICT56924.2023.10157086","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157086","url":null,"abstract":"This paper presents an Electromyography (EMG) based system for classification of sitting and standing postures. The posture is classified by a machine learning model applied on the lower limb EMG data of the user. The dataset is collected from eight subjects, each with 8000 samples per channel, where six are used for training and two for testing. Time-domain, frequency-domain, and time-frequency-domain features are extracted for classification of sitting and standing postures. An array of algorithms are used for classification. Among all the classifiers Random Forest provided the highest accuracy at 98.38%.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127523729","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}
The creation of key sequences for cryptographic protocols such as digital signatures, hashing, encryption, seed vector, one-time passwords (OTP), etc. depends heavily on pseudo random number generators (PRNG). Simple chaotic (PRNG)s exhibit superior randomness and unpredictability. The primary factor increasing area usage is the initial seed, which requires some memory to retain. When compared to other techniques, a simple chaotic pseudo random number generator (CPRNG) has the quality of being very unpredictable and sensitized to the initial seed. If the original seed is compromised, CPRNG is however exposed. We introduced a novel PUF-CPRNG key generation method that is useful for cryptographic applications in this paper. The purpose of PUF in this study is to generate a secured first seed. The proposed PUF-CPRNG which contains dynamic refresh logic to assurance correctness of the generated random numbers at all times. The Proposed Design Consumes 37.5% less Power than the Existing method [1].
{"title":"Design of PUF Based Chaotic Random Number Generator","authors":"Uday Kiran Anchana, Manisha Mogireddy, Ekshith Kadavergu, Sangeeta Singh","doi":"10.1109/ICEEICT56924.2023.10157913","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157913","url":null,"abstract":"The creation of key sequences for cryptographic protocols such as digital signatures, hashing, encryption, seed vector, one-time passwords (OTP), etc. depends heavily on pseudo random number generators (PRNG). Simple chaotic (PRNG)s exhibit superior randomness and unpredictability. The primary factor increasing area usage is the initial seed, which requires some memory to retain. When compared to other techniques, a simple chaotic pseudo random number generator (CPRNG) has the quality of being very unpredictable and sensitized to the initial seed. If the original seed is compromised, CPRNG is however exposed. We introduced a novel PUF-CPRNG key generation method that is useful for cryptographic applications in this paper. The purpose of PUF in this study is to generate a secured first seed. The proposed PUF-CPRNG which contains dynamic refresh logic to assurance correctness of the generated random numbers at all times. The Proposed Design Consumes 37.5% less Power than the Existing method [1].","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126728734","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}
The research work presents a microwave access 4-port multiple input multiple (MIMO) output antenna designed on a FR4 epoxy substrate in this article. A single element design radiating electromagnetic energy is simulated and presented with the narrow bandwidth from 2.36-2.425GHz on $60times 60$ mm2 substrate. The proposed antenna releasing EM wave provides the refection coefficient of −22.447dB at 2.4GHz and peak S22 reaches up to −27.1261dB. It is evident that the given antenna resonates at 2.4 GHz frequency band with acceptable reflection coefficient, bandwidth, and gain, which supports Bluetooth and Wi-Fi technology. The proposed design is then converted to the MIMO structure for mobile users, which offers good real and imaginary impedance of 50 ohm and 0 ohm respectively. The proposed antenna offers a peak gain of 2.63 dBi and radiation efficiency of more than 90% in communication technologies. The MIMO antenna exhibits good diversity performance, with to diversity gain ($text{DG}_{2.4} > 9.999text{dB}$), Envelope Correlation Coefficient $(text{ECC}_{2.4} < 1.12times 10^{-7})$, Total Active Reflection Coefficient $(text{TARC}_{2.4} < -6text{dB}), text{CCL}2.4 < 0.30mathrm{b}/mathrm{s}/text{Hz}$ and $text{MEG}_{mathrm{a}}-text{MEG}_{mathrm{b}}approx 0dB$.
{"title":"A Four-Port Novel Inset-Fed, Rectangular MIMO-Antenna Designed for 2.40 GHz Bluetooth & Wi-Fi Applications","authors":"Vaishali Kikan, Anusha Dagar, Shreya Singh, Shweta Singh, Nishika Chandra Deo, Ashwni Kumar, Manish Sharma","doi":"10.1109/ICEEICT56924.2023.10156908","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10156908","url":null,"abstract":"The research work presents a microwave access 4-port multiple input multiple (MIMO) output antenna designed on a FR4 epoxy substrate in this article. A single element design radiating electromagnetic energy is simulated and presented with the narrow bandwidth from 2.36-2.425GHz on $60times 60$ mm2 substrate. The proposed antenna releasing EM wave provides the refection coefficient of −22.447dB at 2.4GHz and peak S22 reaches up to −27.1261dB. It is evident that the given antenna resonates at 2.4 GHz frequency band with acceptable reflection coefficient, bandwidth, and gain, which supports Bluetooth and Wi-Fi technology. The proposed design is then converted to the MIMO structure for mobile users, which offers good real and imaginary impedance of 50 ohm and 0 ohm respectively. The proposed antenna offers a peak gain of 2.63 dBi and radiation efficiency of more than 90% in communication technologies. The MIMO antenna exhibits good diversity performance, with to diversity gain ($text{DG}_{2.4} > 9.999text{dB}$), Envelope Correlation Coefficient $(text{ECC}_{2.4} < 1.12times 10^{-7})$, Total Active Reflection Coefficient $(text{TARC}_{2.4} < -6text{dB}), text{CCL}2.4 < 0.30mathrm{b}/mathrm{s}/text{Hz}$ and $text{MEG}_{mathrm{a}}-text{MEG}_{mathrm{b}}approx 0dB$.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127181496","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10156982
Bhaskara Naga Siresha, N. Swathi, R. Kiranmayi, K. Nagabhushanam
Brushless DC motor (BLDC) is a type of synchronous motor, gaining high popularity in various industries due to having high efficiency, dynamic response and long operating life. Closed loop control strategies were established for industrial drive applications, and PI, PID, FOPID, and FUZZY controllers were utilized in conjunction with power electronic converters. This paper presents controlling of speed and torque of BLDC motor using hybrid techniques. With the help of FOPID & Fuzzy controller, motor's reference current and inverter DC Bus voltage can be varied respectively. For tuning parameters of FOPID controller, a Modified Harmonic Search (HS) algorithm is employed. Hybrid Fuzzy-FOPID controller is used and implemented in simulink platform. BLDC motor is tested under three different operating conditions such as No-Load, Variable load, variable speed. Simulation results show that Fuzzy-FOPID controller gives greater steady-state error, rated starting torque and ripples throughout the speed profile. To overcome this drawback Hybrid ANFIS-FOPID controller with HS algorithm will be implemented and developed in MATLAB/Simulink platform and evaluates its performance under three different operating conditions. Simulation results show that proposed ANFIS-FOPID controller reduces steady-state error and ripples.
{"title":"Speed Control Of Brushless DC Motor Using Hybrid ANFIS-FOPID Controller","authors":"Bhaskara Naga Siresha, N. Swathi, R. Kiranmayi, K. Nagabhushanam","doi":"10.1109/ICEEICT56924.2023.10156982","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10156982","url":null,"abstract":"Brushless DC motor (BLDC) is a type of synchronous motor, gaining high popularity in various industries due to having high efficiency, dynamic response and long operating life. Closed loop control strategies were established for industrial drive applications, and PI, PID, FOPID, and FUZZY controllers were utilized in conjunction with power electronic converters. This paper presents controlling of speed and torque of BLDC motor using hybrid techniques. With the help of FOPID & Fuzzy controller, motor's reference current and inverter DC Bus voltage can be varied respectively. For tuning parameters of FOPID controller, a Modified Harmonic Search (HS) algorithm is employed. Hybrid Fuzzy-FOPID controller is used and implemented in simulink platform. BLDC motor is tested under three different operating conditions such as No-Load, Variable load, variable speed. Simulation results show that Fuzzy-FOPID controller gives greater steady-state error, rated starting torque and ripples throughout the speed profile. To overcome this drawback Hybrid ANFIS-FOPID controller with HS algorithm will be implemented and developed in MATLAB/Simulink platform and evaluates its performance under three different operating conditions. Simulation results show that proposed ANFIS-FOPID controller reduces steady-state error and ripples.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126591041","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157385
Ajith Sankar R, S. Juliet
Machine learning Techniques are identified as the most suitable methods for mental health analysis and prediction. Mental illness among people has increased vastly around the world and has become a serious human problem to be solved. From much research work and research articles, it is evident that machine learning algorithms can be an effective approach to finding mental illness. In this paper, different machine learning algorithms are investigated to find the best model, suitable to predict the mental health of a person more accurately and at a faster rate. In order to create a system that operates effectively and quickly, this paper investigates the performance of various machine learning models, including KNN, Support Vector Machine, Random Forest, Logistic regression, Decision tree, etc. All the models are compared based on the accuracy that each method offers after successful execution.
{"title":"Investigations on Machine Learning Models for Mental Health Analysis and Prediction","authors":"Ajith Sankar R, S. Juliet","doi":"10.1109/ICEEICT56924.2023.10157385","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157385","url":null,"abstract":"Machine learning Techniques are identified as the most suitable methods for mental health analysis and prediction. Mental illness among people has increased vastly around the world and has become a serious human problem to be solved. From much research work and research articles, it is evident that machine learning algorithms can be an effective approach to finding mental illness. In this paper, different machine learning algorithms are investigated to find the best model, suitable to predict the mental health of a person more accurately and at a faster rate. In order to create a system that operates effectively and quickly, this paper investigates the performance of various machine learning models, including KNN, Support Vector Machine, Random Forest, Logistic regression, Decision tree, etc. All the models are compared based on the accuracy that each method offers after successful execution.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124866552","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157324
Robin George, T. A. Jones Mary
An eight shaped split ring resonator (ES-SRR) is proposed which can be used in size reduction and overall performance enhancement of patch antennas. A unit cell ES-SRR is designed using sea lion optimization technique which is convenient compared to conventional methods. Outer radius, gap and width of the concentric circles are optimized in order to achieve resonance frequency of 2.45 GHz. This proposed unit cell is fabricated and tested.
{"title":"Sea Lion Optimized Eight Shaped Split Ring Resonator","authors":"Robin George, T. A. Jones Mary","doi":"10.1109/ICEEICT56924.2023.10157324","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157324","url":null,"abstract":"An eight shaped split ring resonator (ES-SRR) is proposed which can be used in size reduction and overall performance enhancement of patch antennas. A unit cell ES-SRR is designed using sea lion optimization technique which is convenient compared to conventional methods. Outer radius, gap and width of the concentric circles are optimized in order to achieve resonance frequency of 2.45 GHz. This proposed unit cell is fabricated and tested.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"685 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122701559","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10157163
S. Salma, H. Khan, B. Madhav, B. Sathwik, S. S. D. Praveen Koushik, P. Arun Harsha Vardhan
An polyester substrate textile antenna is proposed for wearable body applications at 4.6 and 5.8 GHz frequencies. The antenna has the required compact design for wearable applications. The X-mass tree-shaped patch, supported by the staircase structure, aids the antenna in operating at treble bands. The antenna's polyester has hydrophobic properties, and the conductive ground patch layers were portrayed by means of the conductive adhesive copper film. A prototype was developed with the polyester material as the substrate, and the total footprints of the antenna are 30x20 mm2. This model was meant for wearable applications, so it was intensively tested in many horizontal and vertical bending positions. Thus the conformability of the antenna was validated. The specific absorption rate (SAR) analysis was also done on a three-level human phantom prototype; comprising muscle, fat and skin. The dual wearable application frequencies of 4.6 and 5.8 GHz applications are validated. The results from the SAR analysis conclude that the antenna is safe to use on the human body with a max SAR of 0.762 and 0.698 w/kg for 1 gram of tissue. The compact design with conformability and safe SAR thresholds aid the antenna for wearable body application.
{"title":"On-Body and SAR analysis of a Polyester Textile antenna for Wearable Applications","authors":"S. Salma, H. Khan, B. Madhav, B. Sathwik, S. S. D. Praveen Koushik, P. Arun Harsha Vardhan","doi":"10.1109/ICEEICT56924.2023.10157163","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10157163","url":null,"abstract":"An polyester substrate textile antenna is proposed for wearable body applications at 4.6 and 5.8 GHz frequencies. The antenna has the required compact design for wearable applications. The X-mass tree-shaped patch, supported by the staircase structure, aids the antenna in operating at treble bands. The antenna's polyester has hydrophobic properties, and the conductive ground patch layers were portrayed by means of the conductive adhesive copper film. A prototype was developed with the polyester material as the substrate, and the total footprints of the antenna are 30x20 mm2. This model was meant for wearable applications, so it was intensively tested in many horizontal and vertical bending positions. Thus the conformability of the antenna was validated. The specific absorption rate (SAR) analysis was also done on a three-level human phantom prototype; comprising muscle, fat and skin. The dual wearable application frequencies of 4.6 and 5.8 GHz applications are validated. The results from the SAR analysis conclude that the antenna is safe to use on the human body with a max SAR of 0.762 and 0.698 w/kg for 1 gram of tissue. The compact design with conformability and safe SAR thresholds aid the antenna for wearable body application.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126321044","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 : 2023-04-05DOI: 10.1109/ICEEICT56924.2023.10156922
Neeraj Singla
American Sign Language (ASL) is a complex and diverse language used by millions of individuals with hearing impairments or disabilities. Accurate and efficient recognition of ASL letters from images is crucial for effective communication and accessibility.[1] However, this is a difficult task due to different hand shapes, orientations, and lighting conditions.In this study, we present a deep learning-based approach for accurately recognizing ASL characters from images. We trained three convolutional neural network (CNN) models, namely VGG16, InceptionV3, and MobileNetV2, on a large dataset of ASL letter images. These models were chosen because they have shown impressive performance in image classification tasks in various contexts. After training, we evaluated the models on a test set of ASL letter images, achieving classification accuracies of 90.7%, 95.7%, and 98% for VGG16, InceptionV3, and MobileNetV2 respectively.Our research provides significant contributions to the field of computer vision, particularly in the recognition of ASL letters from images. Our findings highlight the potential of deep learning-based research development for improving communication technology and accessibility for individuals with hearing impairments, by providing accurate and efficient recognition of ASL letters from images.
{"title":"American Sign Language Letter Recognition from Images Using CNN","authors":"Neeraj Singla","doi":"10.1109/ICEEICT56924.2023.10156922","DOIUrl":"https://doi.org/10.1109/ICEEICT56924.2023.10156922","url":null,"abstract":"American Sign Language (ASL) is a complex and diverse language used by millions of individuals with hearing impairments or disabilities. Accurate and efficient recognition of ASL letters from images is crucial for effective communication and accessibility.[1] However, this is a difficult task due to different hand shapes, orientations, and lighting conditions.In this study, we present a deep learning-based approach for accurately recognizing ASL characters from images. We trained three convolutional neural network (CNN) models, namely VGG16, InceptionV3, and MobileNetV2, on a large dataset of ASL letter images. These models were chosen because they have shown impressive performance in image classification tasks in various contexts. After training, we evaluated the models on a test set of ASL letter images, achieving classification accuracies of 90.7%, 95.7%, and 98% for VGG16, InceptionV3, and MobileNetV2 respectively.Our research provides significant contributions to the field of computer vision, particularly in the recognition of ASL letters from images. Our findings highlight the potential of deep learning-based research development for improving communication technology and accessibility for individuals with hearing impairments, by providing accurate and efficient recognition of ASL letters from images.","PeriodicalId":345324,"journal":{"name":"2023 Second International Conference on Electrical, Electronics, Information and Communication Technologies (ICEEICT)","volume":"155 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116515404","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：