Pub Date : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10202071
Kohei Terakawa, Sinan Chen, Masahide Nakamura
Obsolete software developed in the past is gradually phased out over time. However, the source code of such software contains a wealth of information that can be re-purposed and possesses a high value as an asset. Thus, understanding the characteristics of existing software can aid in developing new software. In a previous study, we proposed a method for inferring the architecture of an existing system using a project corpus and conducted preliminary experiments to verify its feasibility. The findings revealed that the project corpus could be used to infer a system's purpose, functionality, and technology. In this present study, we confirm the validity of the project corpus from a perspective that was not examined in previous studies. We established three verification items and conducted an experiment in which we employed project corpus to infer the functionality of the system, the technology utilized in the system, and the architecture of the system. From the experiment results, we confirmed that the accuracy of the project's inference depends on two factors: first, that the project corpus accurately reflects the system's information, and second, the participants' familiarity with the words in the corpus.
{"title":"Design and Evaluating a Method Using Project Corpus for Inferring Software Description","authors":"Kohei Terakawa, Sinan Chen, Masahide Nakamura","doi":"10.1109/JCSSE58229.2023.10202071","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10202071","url":null,"abstract":"Obsolete software developed in the past is gradually phased out over time. However, the source code of such software contains a wealth of information that can be re-purposed and possesses a high value as an asset. Thus, understanding the characteristics of existing software can aid in developing new software. In a previous study, we proposed a method for inferring the architecture of an existing system using a project corpus and conducted preliminary experiments to verify its feasibility. The findings revealed that the project corpus could be used to infer a system's purpose, functionality, and technology. In this present study, we confirm the validity of the project corpus from a perspective that was not examined in previous studies. We established three verification items and conducted an experiment in which we employed project corpus to infer the functionality of the system, the technology utilized in the system, and the architecture of the system. From the experiment results, we confirmed that the accuracy of the project's inference depends on two factors: first, that the project corpus accurately reflects the system's information, and second, the participants' familiarity with the words in the corpus.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129032068","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10201987
Prathan Phumphuang, Wirat Jareevongpiboon
This research uses support vector machines to forecast the direction of the GameFi market on a daily basis. GameFi is a blockchain-based product that combines gaming and trading. The customer can profit from any in-game activity. Machine learning is a computational method that is able to forecast the future using historical data or attribute information such as open, high, low, close, volume, and market capitalization. Like many researchers who are working to develop accurate machine learning models that can forecast the market value of stocks, commodities like gold and gas, and cryptocurrencies, the main goal of this study is to use SVM to predict the daily direction of GameFi's price. The experimental result shows that SVM's prediction performance is best at 57.6%.
{"title":"Predicting GameFi's Daily Market Direction Using Support Vector Machine","authors":"Prathan Phumphuang, Wirat Jareevongpiboon","doi":"10.1109/JCSSE58229.2023.10201987","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10201987","url":null,"abstract":"This research uses support vector machines to forecast the direction of the GameFi market on a daily basis. GameFi is a blockchain-based product that combines gaming and trading. The customer can profit from any in-game activity. Machine learning is a computational method that is able to forecast the future using historical data or attribute information such as open, high, low, close, volume, and market capitalization. Like many researchers who are working to develop accurate machine learning models that can forecast the market value of stocks, commodities like gold and gas, and cryptocurrencies, the main goal of this study is to use SVM to predict the daily direction of GameFi's price. The experimental result shows that SVM's prediction performance is best at 57.6%.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129810211","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10202106
Rizwanul Islam Rudra, Anilkumar Kothalil Gopalakrishnan
Sentiment analysis is a beneficial method in natural language processing to understand the sentiments of end consumers. This research is focused on sentiment analysis on the IMDb dataset based on a standard Recurrent Neural Network (RNN) with two, three, and four layers of Gated Recurrent Unit (GRU) and Bidirectional Gated Recurrent Unit (BiGRU). In addition, this research uses the GloVe word embedding model to transform raw sentiment texts into meaningful vectors for creating a seen dataset. The presented method has shown more accuracy in identifying sentiments with multiple negations than existing algorithms. Apart from the RNN-GRU and RNN-BiGRU models, this paper has also tested various algorithms such as standard RNN, SVM, LSTM, CNN, Random Forest, Naïve Bayes, Logistic Regression, and Neural Networks with the same dataset (IMDb). It is noticed that the presented RNN-GRU and RNN-BiGRU models outperform other models in polarizing unlabeled sentiments.
{"title":"Sentiment Analysis of Consumer Reviews Using Machine Learning Approach","authors":"Rizwanul Islam Rudra, Anilkumar Kothalil Gopalakrishnan","doi":"10.1109/JCSSE58229.2023.10202106","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10202106","url":null,"abstract":"Sentiment analysis is a beneficial method in natural language processing to understand the sentiments of end consumers. This research is focused on sentiment analysis on the IMDb dataset based on a standard Recurrent Neural Network (RNN) with two, three, and four layers of Gated Recurrent Unit (GRU) and Bidirectional Gated Recurrent Unit (BiGRU). In addition, this research uses the GloVe word embedding model to transform raw sentiment texts into meaningful vectors for creating a seen dataset. The presented method has shown more accuracy in identifying sentiments with multiple negations than existing algorithms. Apart from the RNN-GRU and RNN-BiGRU models, this paper has also tested various algorithms such as standard RNN, SVM, LSTM, CNN, Random Forest, Naïve Bayes, Logistic Regression, and Neural Networks with the same dataset (IMDb). It is noticed that the presented RNN-GRU and RNN-BiGRU models outperform other models in polarizing unlabeled sentiments.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123438343","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10201986
Supanat Jintawatsakoon, Ekkapob Poonsawat
Gender is a crucial consideration in many fields of study. In the era of social networks, massive volumes of data are gathered and processed, enabling us to use this data for a variety of purposes. Our objective to build a gender classification model based on Thai text using natural language processing (NLP) and a machine learning approach. We collected the data on social media websites using web scraping. TF-IDF and n-gram were applied for feature extraction tasks. Logistic Regression, Naïve Bayes, and Random Forest have implemented classification models. Accuracy, precision, recall, and f1 score are used as evaluation metrics and demonstrate that the Logistic Regression model trained on the features derived from data received from texts longer than 200 words produces the best outcome. The dataset is available at https://github.com/supanat/gender-classification-thai-text.git.
{"title":"Gender Classification of Social Network Text Using Natural Language Processing and Machine Learning Approaches","authors":"Supanat Jintawatsakoon, Ekkapob Poonsawat","doi":"10.1109/JCSSE58229.2023.10201986","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10201986","url":null,"abstract":"Gender is a crucial consideration in many fields of study. In the era of social networks, massive volumes of data are gathered and processed, enabling us to use this data for a variety of purposes. Our objective to build a gender classification model based on Thai text using natural language processing (NLP) and a machine learning approach. We collected the data on social media websites using web scraping. TF-IDF and n-gram were applied for feature extraction tasks. Logistic Regression, Naïve Bayes, and Random Forest have implemented classification models. Accuracy, precision, recall, and f1 score are used as evaluation metrics and demonstrate that the Logistic Regression model trained on the features derived from data received from texts longer than 200 words produces the best outcome. The dataset is available at https://github.com/supanat/gender-classification-thai-text.git.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131583289","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10202060
Tanapat Samakit, Chutiporn Anutariya, M. Buranarach
Open Government Data (OGD) refers to the provision of data produced by the government to the general public, in a format that is readily readable and can be used by machines with ease. It can also promote transparency, improve decision-making, enhance accountability, create economic opportunities, and encourage civic engagement. The OGD can help citizens understand the government and its legitimacy and transparency. Thus, when the government shares its data with people, it helps to create trust by being transparent, accountable, and promoting innovative solutions that benefit everyone. However, each published dataset has no indication of its quality assessment at all; thus, making it difficult for citizens to assess the reliability of the data from the OGD. Therefore, a data quality assessment for OGD should be developed. This will help create effective datasets which in turn help users understand the data. This study proposes QUALYST, a system that assesses Thailand's OGD dataset and validates it for analytic and visualization purposes. The study focuses on designing the data storage and implementing the assessment system. Furthermore, the proposed data quality dimensions, the developed data pipeline, and the assessment process are elaborated. Finally, the prototype system is demonstrated using Thailand's OGD datasets with examples in a visualized format.
{"title":"QUALYST: Data Quality Assessment System for Thailand Open Government Data","authors":"Tanapat Samakit, Chutiporn Anutariya, M. Buranarach","doi":"10.1109/JCSSE58229.2023.10202060","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10202060","url":null,"abstract":"Open Government Data (OGD) refers to the provision of data produced by the government to the general public, in a format that is readily readable and can be used by machines with ease. It can also promote transparency, improve decision-making, enhance accountability, create economic opportunities, and encourage civic engagement. The OGD can help citizens understand the government and its legitimacy and transparency. Thus, when the government shares its data with people, it helps to create trust by being transparent, accountable, and promoting innovative solutions that benefit everyone. However, each published dataset has no indication of its quality assessment at all; thus, making it difficult for citizens to assess the reliability of the data from the OGD. Therefore, a data quality assessment for OGD should be developed. This will help create effective datasets which in turn help users understand the data. This study proposes QUALYST, a system that assesses Thailand's OGD dataset and validates it for analytic and visualization purposes. The study focuses on designing the data storage and implementing the assessment system. Furthermore, the proposed data quality dimensions, the developed data pipeline, and the assessment process are elaborated. Finally, the prototype system is demonstrated using Thailand's OGD datasets with examples in a visualized format.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126668772","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}
Nam Dok Mai See-Thong mango is a highly profitable export fruit for Thailand's economy. The grading of high-quality mangoes meeting international standards leads to higher export prices. This paper presents applied deep learning and transfer learning techniques to classify and grade Nam Dok Mai See-Thong mango. Four models, namely MobileNetV2, DenseNet121, InceptionV3, and Xception, were employed to classify mangoes into four categories based on images: perfect mango, ripe, moldy, and shape. The study utilized a large dataset of mango images for training the models and evaluated the results using accuracy, precision, recall, and F1-score. The study proposes the potential of machine learning to enhance the accuracy and efficiency of mango classification. The result showed that the MobileNetV2 model performed best in classifying ripe and shaped mangoes, achieving accuracies of 0.94 and 0.71, respectively. In contrast, the Xception model demonstrated superior performance in classifying moldy mangoes, attaining an accuracy of 0.96. This research highlights the importance of utilizing technology in the quality grading of export fruits to improve their economic value.
{"title":"Classification Grading of Nam Dok Mai See-Thong Mango by Deep Learning and Transfer Learning","authors":"Pomboon Pomboomee, Poommipat Lonlue, Papitchaya Praha, Pongsakron Mungmor, Sanya Khruahong","doi":"10.1109/JCSSE58229.2023.10202009","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10202009","url":null,"abstract":"Nam Dok Mai See-Thong mango is a highly profitable export fruit for Thailand's economy. The grading of high-quality mangoes meeting international standards leads to higher export prices. This paper presents applied deep learning and transfer learning techniques to classify and grade Nam Dok Mai See-Thong mango. Four models, namely MobileNetV2, DenseNet121, InceptionV3, and Xception, were employed to classify mangoes into four categories based on images: perfect mango, ripe, moldy, and shape. The study utilized a large dataset of mango images for training the models and evaluated the results using accuracy, precision, recall, and F1-score. The study proposes the potential of machine learning to enhance the accuracy and efficiency of mango classification. The result showed that the MobileNetV2 model performed best in classifying ripe and shaped mangoes, achieving accuracies of 0.94 and 0.71, respectively. In contrast, the Xception model demonstrated superior performance in classifying moldy mangoes, attaining an accuracy of 0.96. This research highlights the importance of utilizing technology in the quality grading of export fruits to improve their economic value.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128117210","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10202013
Dittaphong Prapunwattana, Aurawan Imsombut, Picha Suwannahitatorn, Thammakorn Saethang
The Healthcare Accreditation Institute has an assessment and certification process for hospitals applying for Healthcare Accreditation. The assessment process requires a large number of text-based reports. The purpose of this research was to study the text analysis of the self-assessment reports of healthcare facilities and surveyor reports on issues related to the pharmaceutical system to evaluate and rate the accreditation of medical facilities. The natural language text vector analysis technique, together with the Universal Sentence Encoder (USE) was compared to Learning Lightweight Language-agnostic Sentence Embeddings (LEALLA) for encoding data into a high-dimensional format. Next the sentence encoding feature was fed through a machine learning procedure, including artificial neural networks, logistic regression, and support vector machines to classify nursing facility accreditation ratings. The experimental results showed that the USE embedding yielded better performance than the LEALLA embedding across all models with a precision of 0.70 but took slightly longer to encode feature sentences. This research could improve the performance of the analysis and scoring.
{"title":"An Evaluation Of Hospital Accreditation From The Survey With Text Vector Analysis Techniques","authors":"Dittaphong Prapunwattana, Aurawan Imsombut, Picha Suwannahitatorn, Thammakorn Saethang","doi":"10.1109/JCSSE58229.2023.10202013","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10202013","url":null,"abstract":"The Healthcare Accreditation Institute has an assessment and certification process for hospitals applying for Healthcare Accreditation. The assessment process requires a large number of text-based reports. The purpose of this research was to study the text analysis of the self-assessment reports of healthcare facilities and surveyor reports on issues related to the pharmaceutical system to evaluate and rate the accreditation of medical facilities. The natural language text vector analysis technique, together with the Universal Sentence Encoder (USE) was compared to Learning Lightweight Language-agnostic Sentence Embeddings (LEALLA) for encoding data into a high-dimensional format. Next the sentence encoding feature was fed through a machine learning procedure, including artificial neural networks, logistic regression, and support vector machines to classify nursing facility accreditation ratings. The experimental results showed that the USE embedding yielded better performance than the LEALLA embedding across all models with a precision of 0.70 but took slightly longer to encode feature sentences. This research could improve the performance of the analysis and scoring.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114727193","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 : 2023-06-28DOI: 10.1109/jcsse58229.2023.10202129
{"title":"Sponsors","authors":"","doi":"10.1109/jcsse58229.2023.10202129","DOIUrl":"https://doi.org/10.1109/jcsse58229.2023.10202129","url":null,"abstract":"","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122077596","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10201947
Bhattaraprot Bhabhatsatam, Sucha Smanchat
This research applies Bit-Partition hybrid quantum encoding methods to store and process classical data in quantum systems efficiently. By combining amplitude encoding for representing the index and basis encoding for the data values, we introduce a novel technique that leverages the strengths of both encoding methods. We describe the process of encoding and decoding the hybrid states, highlighting the potential benefits of this approach in terms of data storage and computational efficiency. Furthermore, we explore the decoding process, addressing the inherent uncertainty associated with quantum measurements and discussing strategies to minimize such uncertainty. Our findings suggest that hybrid encoding can improve quantum information processing tasks, making it a promising technique for future quantum computing applications. Further research is needed to optimize the encoding and decoding processes and explore the full potential of this approach in various quantum algorithms.
{"title":"Hybrid Quantum Encoding: Combining Amplitude and Basis Encoding for Enhanced Data Storage and Processing in Quantum Computing","authors":"Bhattaraprot Bhabhatsatam, Sucha Smanchat","doi":"10.1109/JCSSE58229.2023.10201947","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10201947","url":null,"abstract":"This research applies Bit-Partition hybrid quantum encoding methods to store and process classical data in quantum systems efficiently. By combining amplitude encoding for representing the index and basis encoding for the data values, we introduce a novel technique that leverages the strengths of both encoding methods. We describe the process of encoding and decoding the hybrid states, highlighting the potential benefits of this approach in terms of data storage and computational efficiency. Furthermore, we explore the decoding process, addressing the inherent uncertainty associated with quantum measurements and discussing strategies to minimize such uncertainty. Our findings suggest that hybrid encoding can improve quantum information processing tasks, making it a promising technique for future quantum computing applications. Further research is needed to optimize the encoding and decoding processes and explore the full potential of this approach in various quantum algorithms.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117167310","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 : 2023-06-28DOI: 10.1109/JCSSE58229.2023.10201996
Panupong Sunkom, D. Worasawate, C. Srisurangkul, M. Nakayama
Remote photoplethysmography (rPPG) is a non-contact method that can be used to estimate heart rate (HR) from facial video. Many regions of interest (ROI) on the face were suggested for the rPPG signal extraction, such as the forehead and cheek. However, for the forehead ROI, some parts of the skin area may be covered by hair. This could lead to incorrect rPPG signals and HR. This paper proposes a method to improve the quality of forehead ROI by detecting and removing parts covered by hair based on the green color signal extracted from the forehead area. Any change in ambient light could introduce spikes of spurious frequencies in the interested interval. These spurious frequencies might be stronger than extracted rPPG signals. To overcome these spikes, subintervals are considered. The Short Time Fourier Transform (STFT) is applied to the rPPG signal of the interested interval to obtain HR for each subinterval. The representative HR for the interested interval is selected by majority vote. The estimated HR on the interested interval is then computed based on the representative HR. These experiments were performed on a public dataset, UBFC-RPPG, and the results show that the mean absolute error (MAE) of HR is improved by the proposed method.
{"title":"Improved Heart Rate Estimation From Facial Videos Using Hair Detection and Majority Vote in Subintervals","authors":"Panupong Sunkom, D. Worasawate, C. Srisurangkul, M. Nakayama","doi":"10.1109/JCSSE58229.2023.10201996","DOIUrl":"https://doi.org/10.1109/JCSSE58229.2023.10201996","url":null,"abstract":"Remote photoplethysmography (rPPG) is a non-contact method that can be used to estimate heart rate (HR) from facial video. Many regions of interest (ROI) on the face were suggested for the rPPG signal extraction, such as the forehead and cheek. However, for the forehead ROI, some parts of the skin area may be covered by hair. This could lead to incorrect rPPG signals and HR. This paper proposes a method to improve the quality of forehead ROI by detecting and removing parts covered by hair based on the green color signal extracted from the forehead area. Any change in ambient light could introduce spikes of spurious frequencies in the interested interval. These spurious frequencies might be stronger than extracted rPPG signals. To overcome these spikes, subintervals are considered. The Short Time Fourier Transform (STFT) is applied to the rPPG signal of the interested interval to obtain HR for each subinterval. The representative HR for the interested interval is selected by majority vote. The estimated HR on the interested interval is then computed based on the representative HR. These experiments were performed on a public dataset, UBFC-RPPG, and the results show that the mean absolute error (MAE) of HR is improved by the proposed method.","PeriodicalId":298838,"journal":{"name":"2023 20th International Joint Conference on Computer Science and Software Engineering (JCSSE)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115456236","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：