Pub Date : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576802
M. K. Awang, Muhd Aliff Haiqal Mohd Marzuki, Nurul Kamilah Mat Kamil
Multimodal image fusion is a method of fusing images of different modalities into one image without losing the overall meaning of the input images. The homomorphic method is one method to enhance digital images by increasing the high-frequency image signals and reducing the low-frequency of unwanted illumination. This paper will demonstrate how the homomorphic fusion method can improve the fused image quality compared to basic fusion methods such as principal component analysis (PCA) and discrete wavelet transform (DWT) methods. The design and simulation are carried out by MATLAB software on selected medical modalities, MR-PET and MRI. The results are compared using Mutual Information (MI) with aforementioned methods. The results showed that the homomorphic method has higher efficiency than DWT and PCA methods.
{"title":"Design and Optimization of Homomorphic Medical Image Fusion Algorithm","authors":"M. K. Awang, Muhd Aliff Haiqal Mohd Marzuki, Nurul Kamilah Mat Kamil","doi":"10.1109/ICSIPA52582.2021.9576802","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576802","url":null,"abstract":"Multimodal image fusion is a method of fusing images of different modalities into one image without losing the overall meaning of the input images. The homomorphic method is one method to enhance digital images by increasing the high-frequency image signals and reducing the low-frequency of unwanted illumination. This paper will demonstrate how the homomorphic fusion method can improve the fused image quality compared to basic fusion methods such as principal component analysis (PCA) and discrete wavelet transform (DWT) methods. The design and simulation are carried out by MATLAB software on selected medical modalities, MR-PET and MRI. The results are compared using Mutual Information (MI) with aforementioned methods. The results showed that the homomorphic method has higher efficiency than DWT and PCA methods.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115855427","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576807
R. Gurajala, P. Choppala, J. Meka, Paul D. Teal
The particle filter is an important approximation method for online state estimation in nonlinear nonGaussian scenarios. The resampling step in the particle filter is critical because it eliminates the wasteful use of particles that do not contribute to the posterior (degeneracy). The fully stochastic resamplers, despite being unbiased in approximating the posterior density, involve exhaustive and sequential communication within the particles and thus are computationally expensive. The alternate partial deterministic resamplers overcome this problem by reducing the communication within particles but this leads to approximation bias. This paper proposes a fast resampling procedure that gives an accurate approximation of the posterior and tracks as accurately as the conventional resamplers.
{"title":"A Fast and Unbiased Minimalistic Resampling Approach for the Particle Filter","authors":"R. Gurajala, P. Choppala, J. Meka, Paul D. Teal","doi":"10.1109/ICSIPA52582.2021.9576807","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576807","url":null,"abstract":"The particle filter is an important approximation method for online state estimation in nonlinear nonGaussian scenarios. The resampling step in the particle filter is critical because it eliminates the wasteful use of particles that do not contribute to the posterior (degeneracy). The fully stochastic resamplers, despite being unbiased in approximating the posterior density, involve exhaustive and sequential communication within the particles and thus are computationally expensive. The alternate partial deterministic resamplers overcome this problem by reducing the communication within particles but this leads to approximation bias. This paper proposes a fast resampling procedure that gives an accurate approximation of the posterior and tracks as accurately as the conventional resamplers.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114466973","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576774
Y. Jusman, M. K. Anam, Sartika Puspita, Edwyn Saleh, S. N. A. Kanafiah, Rhesezia Intan Tamarena
This study aims to build a dental caries level classification system based on image processing (i.e. to extract texture features) and machine learning methods. The first step was to analyze and discover the extraction results from Gray Level Co-Occurrence Matrix algorithm. After successfully extracting the features, the classification was carried out using a Support Vector Machine (SVM) and K-Nearest Neighbors (KNN). Both machine learnings are analyzed and used to obtain the better alternatives of the classification results. This study employed radiographic images of four dental caries classes consisting of Class 1, 2, 3, and 4. Total of images used after pre-processing are 396 images. Training data is 90% of total images then the rest is the testing data. The classification obtained accuracy value of the SVM and KNN. The SVM classification method revealed the highest accuracy value generated by the Fine Gaussian SVM model was 95.7%. Conversely, the lowest accuracy value generated was 83.3%, derived from the Quadratic SVM model. Meanwhile, the highest accuracy by using KNN is 94.9% of accuracy using Fine and lowest accuracy value generated was 91.4%, derived from Weighted KNN models. The KNN classification results are better than the SVM results.
{"title":"Comparison of Dental Caries Level Images Classification Performance using KNN and SVM Methods","authors":"Y. Jusman, M. K. Anam, Sartika Puspita, Edwyn Saleh, S. N. A. Kanafiah, Rhesezia Intan Tamarena","doi":"10.1109/ICSIPA52582.2021.9576774","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576774","url":null,"abstract":"This study aims to build a dental caries level classification system based on image processing (i.e. to extract texture features) and machine learning methods. The first step was to analyze and discover the extraction results from Gray Level Co-Occurrence Matrix algorithm. After successfully extracting the features, the classification was carried out using a Support Vector Machine (SVM) and K-Nearest Neighbors (KNN). Both machine learnings are analyzed and used to obtain the better alternatives of the classification results. This study employed radiographic images of four dental caries classes consisting of Class 1, 2, 3, and 4. Total of images used after pre-processing are 396 images. Training data is 90% of total images then the rest is the testing data. The classification obtained accuracy value of the SVM and KNN. The SVM classification method revealed the highest accuracy value generated by the Fine Gaussian SVM model was 95.7%. Conversely, the lowest accuracy value generated was 83.3%, derived from the Quadratic SVM model. Meanwhile, the highest accuracy by using KNN is 94.9% of accuracy using Fine and lowest accuracy value generated was 91.4%, derived from Weighted KNN models. The KNN classification results are better than the SVM results.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114982722","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576797
Brandon Sean Kong, I. Hipiny, Hamimah Ujir
Assistive technology has been given more attention in recent years to help people with disabilities to perform common tasks. Rather than designing a specialised tool for the task, it is more cost-effective and less inhibitory to make use of existing hardware integrated with a smart interface. Towards this end goal, we present our work on assisting a visually impaired person playing an online chess game. We evaluated an invariant feature descriptor, i.e., SIFT, for the task of classifying individual chess pieces across multiple visual themes. We compared two strategies for building the visual codebook, i.e., k-means clustering vs. image blending. The proposed pipeline receives live screen feeds from the browser at a fixed interval and produces an output in the form of chess pieces’ label and board position. Our proposed pipeline, paired with a visual codebook built using k-means clustering, managed an average accuracy rate of 6/10.
{"title":"Classification of Digital Chess Pieces and Board Position using SIFT","authors":"Brandon Sean Kong, I. Hipiny, Hamimah Ujir","doi":"10.1109/ICSIPA52582.2021.9576797","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576797","url":null,"abstract":"Assistive technology has been given more attention in recent years to help people with disabilities to perform common tasks. Rather than designing a specialised tool for the task, it is more cost-effective and less inhibitory to make use of existing hardware integrated with a smart interface. Towards this end goal, we present our work on assisting a visually impaired person playing an online chess game. We evaluated an invariant feature descriptor, i.e., SIFT, for the task of classifying individual chess pieces across multiple visual themes. We compared two strategies for building the visual codebook, i.e., k-means clustering vs. image blending. The proposed pipeline receives live screen feeds from the browser at a fixed interval and produces an output in the form of chess pieces’ label and board position. Our proposed pipeline, paired with a visual codebook built using k-means clustering, managed an average accuracy rate of 6/10.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"162 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130717270","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576776
Loshini Thiruchelvam, S. Dass, Nirbhay Mathur, V. Asirvadam, B. Gill
This study aimed to build best dengue cases prediction model for Petaling district, in Selangor. Linear Least Square estimation method is used to build the models and Mean Square Error (MSE) and Akaike Information Criterion (AIC) value is used as tool of comparison between models. Prior to model development, the respective variables are first normalized, using 0–1 normalization procedure. Next, significant predictors are identified from weather variables namely mean temperature, relative humidity, and rainfall. Thirdly, feedback data was included and identified if could yield better prediction models. Few model orders of lag time are built simultaneously, and the most accurate prediction model was selected for Petaling district. Study found dengue prediction models including all three climate variables of mean temperature, relative humidity, cumulative rainfall and together with previous dengue cases to have the lowest MSE and AIC values. This is aligned with previous studies which selected model with climate and previous dengue cases models as best model fit. Thus, study proposed future studies to incorporate all three climate variables and previous dengue cases while developing dengue prediction models.
本研究旨在建立雪兰莪州Petaling地区登革热病例的最佳预测模型。采用线性最小二乘估计方法建立模型，采用均方误差(MSE)和赤池信息准则(Akaike Information Criterion, AIC)值作为模型间的比较工具。在模型开发之前，首先使用0-1规范化过程对各个变量进行规范化。接下来，从天气变量即平均温度、相对湿度和降雨量中确定重要的预测因子。第三，纳入反馈数据，并确定是否可以产生更好的预测模型。同时建立了几个滞后时间的模型阶数，选取了最准确的花瓣陵区预测模型。研究发现，登革热预测模型包括平均温度、相对湿度、累积降雨量这三个气候变量，并结合以往登革热病例，其MSE和AIC值最低。这与以前的研究一致，这些研究选择了气候模型和以前的登革热病例模型作为最佳模型拟合。因此，研究人员建议未来的研究在开发登革热预测模型时纳入所有三个气候变量和以前的登革热病例。
{"title":"Inclusion of Climate Variables for Dengue Prediction Model: Preliminary Analysis","authors":"Loshini Thiruchelvam, S. Dass, Nirbhay Mathur, V. Asirvadam, B. Gill","doi":"10.1109/ICSIPA52582.2021.9576776","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576776","url":null,"abstract":"This study aimed to build best dengue cases prediction model for Petaling district, in Selangor. Linear Least Square estimation method is used to build the models and Mean Square Error (MSE) and Akaike Information Criterion (AIC) value is used as tool of comparison between models. Prior to model development, the respective variables are first normalized, using 0–1 normalization procedure. Next, significant predictors are identified from weather variables namely mean temperature, relative humidity, and rainfall. Thirdly, feedback data was included and identified if could yield better prediction models. Few model orders of lag time are built simultaneously, and the most accurate prediction model was selected for Petaling district. Study found dengue prediction models including all three climate variables of mean temperature, relative humidity, cumulative rainfall and together with previous dengue cases to have the lowest MSE and AIC values. This is aligned with previous studies which selected model with climate and previous dengue cases models as best model fit. Thus, study proposed future studies to incorporate all three climate variables and previous dengue cases while developing dengue prediction models.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129126549","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576766
Joe Huei Ong, Kam Meng Goh, Li Li Lim
The COVID-19 outbreak brought a huge impact globally. Early studies show that the COVID-19 is manifested in chest X-rays of infected patients. Hence, these studies attract the attention of the computer vision community in integrating X-ray scans and deep-learning-based solutions to aid the diagnosis of COVID-19 infection. However, at present, efforts and information on implementing explainable artificial intelligence in interpreting deep learning model for COVID-19 recognition are scarce and limited. In this paper, we proposed and compared the LIME and SHAP model to enhance the interpretation of COVID diagnosis through X-ray scans. We first applied SqueezeNet to recognise pneumonia, COVID-19, and normal lung image. Through SqueezeNet, an 84.34% recognition rate success in testing accuracy was obtained. To better understand what the network “sees” a specific task, namely, image classification, Shapley Additive Explanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) were implemented to expound and interpret how Squeezenet performs classification. Results show that LIME and SHAP can highlight the area of interest where they can help to increase the transparency and the interpretability of the Squeezenet model.
{"title":"Comparative Analysis of Explainable Artificial Intelligence for COVID-19 Diagnosis on CXR Image","authors":"Joe Huei Ong, Kam Meng Goh, Li Li Lim","doi":"10.1109/ICSIPA52582.2021.9576766","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576766","url":null,"abstract":"The COVID-19 outbreak brought a huge impact globally. Early studies show that the COVID-19 is manifested in chest X-rays of infected patients. Hence, these studies attract the attention of the computer vision community in integrating X-ray scans and deep-learning-based solutions to aid the diagnosis of COVID-19 infection. However, at present, efforts and information on implementing explainable artificial intelligence in interpreting deep learning model for COVID-19 recognition are scarce and limited. In this paper, we proposed and compared the LIME and SHAP model to enhance the interpretation of COVID diagnosis through X-ray scans. We first applied SqueezeNet to recognise pneumonia, COVID-19, and normal lung image. Through SqueezeNet, an 84.34% recognition rate success in testing accuracy was obtained. To better understand what the network “sees” a specific task, namely, image classification, Shapley Additive Explanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) were implemented to expound and interpret how Squeezenet performs classification. Results show that LIME and SHAP can highlight the area of interest where they can help to increase the transparency and the interpretability of the Squeezenet model.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134351709","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576784
Ziming Liu, Lauren Proctor, Parker N. Collier, Xiaopeng Zhao
With the increasing prevalence of Alzheimer’s disease (AD), it is important to develop detectable biomarkers to reliably identify AD in the early stage. Language deficit is one of the common signs that appear in the early stage of mild Alzheimer’s disease. Therefore, using natural language processing and related machine learning algorithms for AD diagnosis using patients’ speech recordings has drawn more attention in recent years. In this study, three approaches are proposed to extract features through speech recording in this model: (1) using fine-tuning pre-trained encoder model (BERT) for transcripts from automatic transcription, (2) hand-crafted linguistic features for transcripts from automatic transcription, and (3) selected acoustic features for denoised speech recordings. The three designed approaches are applied to three tasks: AD diagnosis, MMSE score prediction, and cognitive decline inference. The approach using BERT yields the best performance in all three challenge tasks based on cross-validation results using the training dataset. Specifically, in the AD diagnosis task, 5-fold cross-validation using encoded features based on transcripts generated from Deep Speech yields an average classification accuracy of 97.18%. In the MMSE score prediction task, 5-fold cross-validation using BERT encoded features based on transcripts generated from Deep Speech yields an average Root Mean Squared Error (RMSE) of 3.76. In the cognitive decline inference task, the leave-one-out cross-validation using BERT encoded features based on transcripts generated from Sphinx or Deep Speech yields an average classification accuracy of 100%. The analyses suggest that the combination of automatic transcription and BERT may produce a significant performance in AD related detection and prediction problems.
{"title":"Automatic Diagnosis and Prediction of Cognitive Decline Associated with Alzheimer’s Dementia through Spontaneous Speech","authors":"Ziming Liu, Lauren Proctor, Parker N. Collier, Xiaopeng Zhao","doi":"10.1109/ICSIPA52582.2021.9576784","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576784","url":null,"abstract":"With the increasing prevalence of Alzheimer’s disease (AD), it is important to develop detectable biomarkers to reliably identify AD in the early stage. Language deficit is one of the common signs that appear in the early stage of mild Alzheimer’s disease. Therefore, using natural language processing and related machine learning algorithms for AD diagnosis using patients’ speech recordings has drawn more attention in recent years. In this study, three approaches are proposed to extract features through speech recording in this model: (1) using fine-tuning pre-trained encoder model (BERT) for transcripts from automatic transcription, (2) hand-crafted linguistic features for transcripts from automatic transcription, and (3) selected acoustic features for denoised speech recordings. The three designed approaches are applied to three tasks: AD diagnosis, MMSE score prediction, and cognitive decline inference. The approach using BERT yields the best performance in all three challenge tasks based on cross-validation results using the training dataset. Specifically, in the AD diagnosis task, 5-fold cross-validation using encoded features based on transcripts generated from Deep Speech yields an average classification accuracy of 97.18%. In the MMSE score prediction task, 5-fold cross-validation using BERT encoded features based on transcripts generated from Deep Speech yields an average Root Mean Squared Error (RMSE) of 3.76. In the cognitive decline inference task, the leave-one-out cross-validation using BERT encoded features based on transcripts generated from Sphinx or Deep Speech yields an average classification accuracy of 100%. The analyses suggest that the combination of automatic transcription and BERT may produce a significant performance in AD related detection and prediction problems.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121800969","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576788
Afiq Aiman Ahmad Fairuz Rizal, N. Hashim
In recent years, the technology of emotion speech recognition has gradually become more important to the industries. This is proven by the integration of this system into many applications such as the interface with robots, audio surveillance, web-based E-learning, commercial applications, clinical studies and so on. Generally, speech emotion recognition (SER) is developed to help humans to understand and retrieve desired emotions. In this research, the analysis of using Bahasa Malaysia Language for three basic emotions of happy, sad and angry was analyzed. A total of 30 male and 30 female audio recordings were collected. Mel-frequency cepstral coefficient, chroma and mel spectrogram features were extracted. Feature dimensions were reduced using forward, backward and exhaustive selection methods before classification. Classification was performed using K-nearest neighbors, Support Vector Machine and Random Forest. The analysis demonstrated 78% accuracy for male speech and 78% for female speech.
{"title":"Emotion Recognition Using Bahasa Malaysia Natural Speech","authors":"Afiq Aiman Ahmad Fairuz Rizal, N. Hashim","doi":"10.1109/ICSIPA52582.2021.9576788","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576788","url":null,"abstract":"In recent years, the technology of emotion speech recognition has gradually become more important to the industries. This is proven by the integration of this system into many applications such as the interface with robots, audio surveillance, web-based E-learning, commercial applications, clinical studies and so on. Generally, speech emotion recognition (SER) is developed to help humans to understand and retrieve desired emotions. In this research, the analysis of using Bahasa Malaysia Language for three basic emotions of happy, sad and angry was analyzed. A total of 30 male and 30 female audio recordings were collected. Mel-frequency cepstral coefficient, chroma and mel spectrogram features were extracted. Feature dimensions were reduced using forward, backward and exhaustive selection methods before classification. Classification was performed using K-nearest neighbors, Support Vector Machine and Random Forest. The analysis demonstrated 78% accuracy for male speech and 78% for female speech.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"41 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125779864","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576767
S. S. Sarnin, M. Yusuf, Ros Shilawani S. Abdul Kadir, N. F. Naim, W. N. W. Mohamad, Mohd Nor Md Tan
This research focuses on providing a solution for a mobile service provider with Multi Network Operators (MNOs) using a single multi-beam antenna via a hybrid circuit to provide an excellent service attended by thousands of Mobile Subscribers (MS) at Nasional Bukit Jalil Kuala Lumpur Stadium. The combination circuit design using the Hybrid Combiner (HC) is the solution used to combine multiple MNOs towards minimizing space and cost besides ensuring the aesthetical value of the national stadium. During a significant incident, MS users may have trouble accessing the service due to unavailability of the service due to network congestion. In this situation, the MNOs must have additional capacity to meet the demand for data transmission and voice call transactions. Improving the output of the network and the quality of service should reflect customer loyalty as it automatically produces. The implementation of the proposed solution, MS users will be able to access the network and will also enjoy live feeds via Facebook (FB) and other software applications without delay and interruption as well as voice call congestion. The results of the suggested solution will be compared with the walk test results and the coverage simulation analysis using the planning methods. Data statistics taken from MNOs will explain the effectiveness of the solution in term of Signal quality level where the Signal to Noise Ratio (SINR) recorded at −95 dBm below the threshold of −85 dBm to prevent interference with MS users. The Resource Block (RB) Utilization shows the utilization of all sectors are at below 70% of total available capacity which means that the congestion level is manageable and MS user able to access the network without interruption. Fast deployment, less maintenance and a shared solution between MNOs is a key factor in the proposed study and is known as Hybrid Combiner Circuit of Multi Network Operator for Capacity Enhancement Solution in Indoor Environment.
{"title":"Hybrid Combiner Circuit Of Multi Network Operator For Capacity Enhancement Solution In Indoor Environment","authors":"S. S. Sarnin, M. Yusuf, Ros Shilawani S. Abdul Kadir, N. F. Naim, W. N. W. Mohamad, Mohd Nor Md Tan","doi":"10.1109/ICSIPA52582.2021.9576767","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576767","url":null,"abstract":"This research focuses on providing a solution for a mobile service provider with Multi Network Operators (MNOs) using a single multi-beam antenna via a hybrid circuit to provide an excellent service attended by thousands of Mobile Subscribers (MS) at Nasional Bukit Jalil Kuala Lumpur Stadium. The combination circuit design using the Hybrid Combiner (HC) is the solution used to combine multiple MNOs towards minimizing space and cost besides ensuring the aesthetical value of the national stadium. During a significant incident, MS users may have trouble accessing the service due to unavailability of the service due to network congestion. In this situation, the MNOs must have additional capacity to meet the demand for data transmission and voice call transactions. Improving the output of the network and the quality of service should reflect customer loyalty as it automatically produces. The implementation of the proposed solution, MS users will be able to access the network and will also enjoy live feeds via Facebook (FB) and other software applications without delay and interruption as well as voice call congestion. The results of the suggested solution will be compared with the walk test results and the coverage simulation analysis using the planning methods. Data statistics taken from MNOs will explain the effectiveness of the solution in term of Signal quality level where the Signal to Noise Ratio (SINR) recorded at −95 dBm below the threshold of −85 dBm to prevent interference with MS users. The Resource Block (RB) Utilization shows the utilization of all sectors are at below 70% of total available capacity which means that the congestion level is manageable and MS user able to access the network without interruption. Fast deployment, less maintenance and a shared solution between MNOs is a key factor in the proposed study and is known as Hybrid Combiner Circuit of Multi Network Operator for Capacity Enhancement Solution in Indoor Environment.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"151 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132706301","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 : 2021-09-13DOI: 10.1109/ICSIPA52582.2021.9576804
Boon Kai Law, Lih Poh Lin
Pneumonia is commonly seen in several diseases, including Covid-19 that has put countries under lockdown today [1]. Other than antigen rapid test kit (RTK) and reverse transcription-polymerase chain reaction (RT-PCR), an alternative method to detect COVID-19 is through the examination of patients’ chest radiography (CXR). However, the results of manual inspections may be false and the misdiagnosis could lead to fatal consequences such as delayed treatment and death. The manual inspection can be inconsistent, inaccurate and may differ from different individuals due to different perspectives. Often, Covid-19 Xrays are misinterpreted as bacterial pneumonia. With the advancement of technology, this issue can be overcome by developing a Convolutional Neural Network (CNN) model to categorize X-ray of normal, pneumonia-affected and COVID-19 patients via deep learning. In this work, various CNN models (ResNet-50, ResNet-101, Vgg-16, Vgg-19 and SqueezeNet) were trained with the public databases that contain a combination of 1345 viral pneumonia, 1200 COVID-19 in addition to 1341 regular CXR images. The transfer learning method was employed, aided by image augmentation for training and validation of ResNet-50, ResNet-101, Vgg-16 and Vgg-19 architectures. Meanwhile, SqueezeNet was trained from scratch to investigate the importance of transfer learning to the model. The highest training accuracy achieved in this study was 97.38% by the VGG-16 model using a learning rate of 0.01 whereas the highest weighted average accuracy achieved was 94% by the VGG-16 model using a learning rate of 0.01 and the VGG-19 model using a learning rate of 0.001. The reliability and high accuracy of the CNN model would open a new avenue for the diagnosis of Covid-19.
{"title":"Development Of A Deep Learning Model To Classify X-Ray Of Covid-19, Normal And Pneumonia-Affected Patients","authors":"Boon Kai Law, Lih Poh Lin","doi":"10.1109/ICSIPA52582.2021.9576804","DOIUrl":"https://doi.org/10.1109/ICSIPA52582.2021.9576804","url":null,"abstract":"Pneumonia is commonly seen in several diseases, including Covid-19 that has put countries under lockdown today [1]. Other than antigen rapid test kit (RTK) and reverse transcription-polymerase chain reaction (RT-PCR), an alternative method to detect COVID-19 is through the examination of patients’ chest radiography (CXR). However, the results of manual inspections may be false and the misdiagnosis could lead to fatal consequences such as delayed treatment and death. The manual inspection can be inconsistent, inaccurate and may differ from different individuals due to different perspectives. Often, Covid-19 Xrays are misinterpreted as bacterial pneumonia. With the advancement of technology, this issue can be overcome by developing a Convolutional Neural Network (CNN) model to categorize X-ray of normal, pneumonia-affected and COVID-19 patients via deep learning. In this work, various CNN models (ResNet-50, ResNet-101, Vgg-16, Vgg-19 and SqueezeNet) were trained with the public databases that contain a combination of 1345 viral pneumonia, 1200 COVID-19 in addition to 1341 regular CXR images. The transfer learning method was employed, aided by image augmentation for training and validation of ResNet-50, ResNet-101, Vgg-16 and Vgg-19 architectures. Meanwhile, SqueezeNet was trained from scratch to investigate the importance of transfer learning to the model. The highest training accuracy achieved in this study was 97.38% by the VGG-16 model using a learning rate of 0.01 whereas the highest weighted average accuracy achieved was 94% by the VGG-16 model using a learning rate of 0.01 and the VGG-19 model using a learning rate of 0.001. The reliability and high accuracy of the CNN model would open a new avenue for the diagnosis of Covid-19.","PeriodicalId":326688,"journal":{"name":"2021 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133721990","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：