Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864702
Muhammed Said Zengin, Rabia Arslan, Mehmet Burak Akgün
The ever-increasing frequency of sharing on social media makes these platforms one of the primary sources of data for computational social science studies. Similarly, examining and analyzing large scale social media data-sets is crucial for governments as well as companies. However, as the amount of data increases, insights that need to be derived from the data using artificial intelligence based models becomes more and more demanding in terms of processing power. In fact, hardware requirements might dramatically increase if the insights are needed under real-time or near-real time constraints. In this study, we developed a distributed sentiment analysis model that utilizes a large social media data-set. 16 million tweets have been collected and grouped by the originating city. The sentiment analysis model was produced by fine-tuning the pre-trained BERT model. Distributed big data analytics engine, Apache Spark, is used to execute the trained model in a distributed fashion. For evaluation purposes, the prediction time on a single compute unit is compared with the distributed prediction time. Sentiment analysis model has been executed separately for each of the data-groups corresponding to 81 provinces. The data-set containing 16 million tweets used in this study, the Turkish sentiment analysis model produced, the distributed prediction code developed for Apache Spark and all the results of the study can be accessed from the address https://distributed-sentiment-analysis.github.io/.
{"title":"Distributed Sentiment Analysis for Geo-Tagged Twitter Data","authors":"Muhammed Said Zengin, Rabia Arslan, Mehmet Burak Akgün","doi":"10.1109/SIU55565.2022.9864702","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864702","url":null,"abstract":"The ever-increasing frequency of sharing on social media makes these platforms one of the primary sources of data for computational social science studies. Similarly, examining and analyzing large scale social media data-sets is crucial for governments as well as companies. However, as the amount of data increases, insights that need to be derived from the data using artificial intelligence based models becomes more and more demanding in terms of processing power. In fact, hardware requirements might dramatically increase if the insights are needed under real-time or near-real time constraints. In this study, we developed a distributed sentiment analysis model that utilizes a large social media data-set. 16 million tweets have been collected and grouped by the originating city. The sentiment analysis model was produced by fine-tuning the pre-trained BERT model. Distributed big data analytics engine, Apache Spark, is used to execute the trained model in a distributed fashion. For evaluation purposes, the prediction time on a single compute unit is compared with the distributed prediction time. Sentiment analysis model has been executed separately for each of the data-groups corresponding to 81 provinces. The data-set containing 16 million tweets used in this study, the Turkish sentiment analysis model produced, the distributed prediction code developed for Apache Spark and all the results of the study can be accessed from the address https://distributed-sentiment-analysis.github.io/.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130521812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864809
Ümmügülsüm Şener, S. Eker
Although concrete has a heterogeneous structure due to the material components it contains, it can also be a structural component such as a cable or rebar in the examined part due to being a component of any structure or infrastructure, and the concrete sample to be examined may also have a layered structure. In this study, non-destructive testing of a layered and cable-containing concrete structure is simulated by microwave radar non-destructive testing technique, which uses microwave and material interaction to determine material characterization and internal components of structures. A two-dimensional simulation setup of the examined structure is prepared, and the results are discussed by comparing the simulation results obtained at different frequencies and different modes.
{"title":"Non-Destructive Assessment of Planarly Layered Concrete Structures Using Electromagnetic Waves","authors":"Ümmügülsüm Şener, S. Eker","doi":"10.1109/SIU55565.2022.9864809","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864809","url":null,"abstract":"Although concrete has a heterogeneous structure due to the material components it contains, it can also be a structural component such as a cable or rebar in the examined part due to being a component of any structure or infrastructure, and the concrete sample to be examined may also have a layered structure. In this study, non-destructive testing of a layered and cable-containing concrete structure is simulated by microwave radar non-destructive testing technique, which uses microwave and material interaction to determine material characterization and internal components of structures. A two-dimensional simulation setup of the examined structure is prepared, and the results are discussed by comparing the simulation results obtained at different frequencies and different modes.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132440843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864696
Assim Ibrahim, E. Tetik, S. Karamzadeh
5G technology for health care is a considerable demand nowadays due to the revolution in IoT devices. A dual-band antenna on a Rogers RT/duroid 5880 substrate with a dielectric constant of 2.2 and thickness of 0.508 mm, is presented for 3.4 GHz and 5.85GHz frequency bands. The top layer of the proposed antenna consists of a square part that includes an inductive meander line. It is connected to an I-shape, which is merged with reversed L-shaped, the higher resonant frequency excited at 3.4 GHz. In the bottom layer, after the reflector is made, the inductive meander line is created and connected to a reversed U-shape. The proposed antenna of size 19 x 12 mm2 reveals to perform well for frequencies between 3.37 and 3.47 GHz. It has an axial ratio less than 3dBi and a peak gain of 1.7 dBi that increases after adding a multi-layer of a human hand up to 8 dBi.
{"title":"A Wearable Circularly Polarized Antenna for 5G Applications","authors":"Assim Ibrahim, E. Tetik, S. Karamzadeh","doi":"10.1109/SIU55565.2022.9864696","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864696","url":null,"abstract":"5G technology for health care is a considerable demand nowadays due to the revolution in IoT devices. A dual-band antenna on a Rogers RT/duroid 5880 substrate with a dielectric constant of 2.2 and thickness of 0.508 mm, is presented for 3.4 GHz and 5.85GHz frequency bands. The top layer of the proposed antenna consists of a square part that includes an inductive meander line. It is connected to an I-shape, which is merged with reversed L-shaped, the higher resonant frequency excited at 3.4 GHz. In the bottom layer, after the reflector is made, the inductive meander line is created and connected to a reversed U-shape. The proposed antenna of size 19 x 12 mm2 reveals to perform well for frequencies between 3.37 and 3.47 GHz. It has an axial ratio less than 3dBi and a peak gain of 1.7 dBi that increases after adding a multi-layer of a human hand up to 8 dBi.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125419434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864666
Talha Ince, Sertac Cakir
The Inertial Navigation System(INS) and Doppler Velocity Logs(DVL) which are used frequently on autonomous underwater vehicles can be fused under different types of integration architectures. These architectures differ in terms of algorithm requirements and complexity. DVL may experience acoustic beam losses during operation due to environmental factors and abilities of the sensor. In these situations, radial velocity information cannot be received from lost acoustic beam. In this paper, the performances of INS and DVL integration under tightly and loosely coupled architectures are comparatively presented with simulations. In the tightly coupled approach, navigation filter is updated with solely available beam measurements by using sequential measurement update method, and the sensitivity of this method is investigated for acoustic beam losses.
{"title":"Tightly and Loosely Coupled Architectures for Inertial Navigation System and Doppler Velocity Log Integration at Autonomous Underwater Vehicles","authors":"Talha Ince, Sertac Cakir","doi":"10.1109/SIU55565.2022.9864666","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864666","url":null,"abstract":"The Inertial Navigation System(INS) and Doppler Velocity Logs(DVL) which are used frequently on autonomous underwater vehicles can be fused under different types of integration architectures. These architectures differ in terms of algorithm requirements and complexity. DVL may experience acoustic beam losses during operation due to environmental factors and abilities of the sensor. In these situations, radial velocity information cannot be received from lost acoustic beam. In this paper, the performances of INS and DVL integration under tightly and loosely coupled architectures are comparatively presented with simulations. In the tightly coupled approach, navigation filter is updated with solely available beam measurements by using sequential measurement update method, and the sensitivity of this method is investigated for acoustic beam losses.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125509895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864786
Baturay Sağlam, Onat Dalmaz, Kaan Gonc, S. Kozat
The Batch-Constrained Q-learning algorithm is shown to overcome the extrapolation error and enable deep reinforcement learning agents to learn from a previously collected fixed batch of transitions. However, due to conditional Variational Autoencoders (VAE) used in the data generation module, the BCQ algorithm optimizes a lower variational bound and hence, it is not generalizable to environments with large state and action spaces. In this paper, we show that the performance of the BCQ algorithm can be further improved with the employment of one of the recent advances in deep learning, Generative Adversarial Networks. Our extensive set of experiments shows that the introduced approach significantly improves BCQ in all of the control tasks tested. Moreover, the introduced approach demonstrates robust generalizability to environments with large state and action spaces in the OpenAI Gym control suite.
{"title":"Improving the Performance of Batch-Constrained Reinforcement Learning in Continuous Action Domains via Generative Adversarial Networks","authors":"Baturay Sağlam, Onat Dalmaz, Kaan Gonc, S. Kozat","doi":"10.1109/SIU55565.2022.9864786","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864786","url":null,"abstract":"The Batch-Constrained Q-learning algorithm is shown to overcome the extrapolation error and enable deep reinforcement learning agents to learn from a previously collected fixed batch of transitions. However, due to conditional Variational Autoencoders (VAE) used in the data generation module, the BCQ algorithm optimizes a lower variational bound and hence, it is not generalizable to environments with large state and action spaces. In this paper, we show that the performance of the BCQ algorithm can be further improved with the employment of one of the recent advances in deep learning, Generative Adversarial Networks. Our extensive set of experiments shows that the introduced approach significantly improves BCQ in all of the control tasks tested. Moreover, the introduced approach demonstrates robust generalizability to environments with large state and action spaces in the OpenAI Gym control suite.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126795393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864706
Canberk Tatli, Egemen Denizeri, D. Kumlu, I. Erer, B. Yalçin, F. Işık
The clutter encountered in Ground Penetrating Radar(GPR) systems is an important area of research since it decreases target detection rates. Real-time radar applications and hardware implementation of clutter removal methods in autonomous systems are crucial. In this study, robust non-negative matrix factorization (RNMF) is used, which requires simple mathematical operations and suitable for hardware implementations. FPGA was chosen as the hardware implementation environment due to its re-programmable feature. It has been shown that the hardware implementation results have the same performance as the clutter removal results obtained in the MATLAB environment.
{"title":"Clutter Removal for Ground Penetrating Radars on FPGA: Design and Implementation","authors":"Canberk Tatli, Egemen Denizeri, D. Kumlu, I. Erer, B. Yalçin, F. Işık","doi":"10.1109/SIU55565.2022.9864706","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864706","url":null,"abstract":"The clutter encountered in Ground Penetrating Radar(GPR) systems is an important area of research since it decreases target detection rates. Real-time radar applications and hardware implementation of clutter removal methods in autonomous systems are crucial. In this study, robust non-negative matrix factorization (RNMF) is used, which requires simple mathematical operations and suitable for hardware implementations. FPGA was chosen as the hardware implementation environment due to its re-programmable feature. It has been shown that the hardware implementation results have the same performance as the clutter removal results obtained in the MATLAB environment.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123182965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864743
Hazal Mogultay, Sinan Kalkan, Fatoş T. Yarman Vural
Learnt representations by Deep autoencoders is not capable of decomposing the complex information into simple notion. In other words, attributes of samples are entangled in the basis vectors spanning the learned space. This leads to significant errors in deep learning algorithms. In order to avoid these errors, it is necessary to separate the feature space according to the common features shared between classes and to define a simple subspace for each feature. This approach has led to the birth of a new paradigm in Machine Learning, called disentanglement.Roughly, disentangled models can be defined as models that can independently learn the different components of the probability density function that produces the dataset in the feature space. Unfortunately, it is not always possible to learn these models. For this reason, there is still no easily applicable mathematical definition of disentanglement in the literature. In this study, a mathematical definition of the concept of disentanglement will be made and methods and metrics related to this approach will be discussed.
{"title":"An Analysis on Disentanglement in Machine Learning","authors":"Hazal Mogultay, Sinan Kalkan, Fatoş T. Yarman Vural","doi":"10.1109/SIU55565.2022.9864743","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864743","url":null,"abstract":"Learnt representations by Deep autoencoders is not capable of decomposing the complex information into simple notion. In other words, attributes of samples are entangled in the basis vectors spanning the learned space. This leads to significant errors in deep learning algorithms. In order to avoid these errors, it is necessary to separate the feature space according to the common features shared between classes and to define a simple subspace for each feature. This approach has led to the birth of a new paradigm in Machine Learning, called disentanglement.Roughly, disentangled models can be defined as models that can independently learn the different components of the probability density function that produces the dataset in the feature space. Unfortunately, it is not always possible to learn these models. For this reason, there is still no easily applicable mathematical definition of disentanglement in the literature. In this study, a mathematical definition of the concept of disentanglement will be made and methods and metrics related to this approach will be discussed.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126362332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864865
R. Ö. Dogan, H. Ture, T. Kayikçioglu
The pectoral muscle region on MLO mammography images appears prominently similar to suspicious areas. For this reason, Computer-Aided Detection (CAD) systems remove this region to reduce false-positive rates in the mass detection process. In some cases, the pectoral muscle region is exposed to distortions due to the superposition effects caused by the mammography technique. As a result, segmentation error rates of the pectoral muscle region, whose characteristic features are deteriorated, appear. In this study, a method to identify impaired pectoral muscle regions with MobileNetV2 backboned U-Net Deep Learning method is proposed. The proposed method was tested on 84 and 201 mammography images taken from both MIAS and InBreast databases and segmented with 1.81% and 1.92% false-negative (FN) and 0.25% and 0.37% false positive (FP) rates, respectively. Particularly for distorted pectoral muscle regions, the proposed method has been shown to outperform some pioneering studies in this area.
{"title":"Segmentation of Pectoral Muscle Region in MLO Mammography Images by Backboned U-Net","authors":"R. Ö. Dogan, H. Ture, T. Kayikçioglu","doi":"10.1109/SIU55565.2022.9864865","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864865","url":null,"abstract":"The pectoral muscle region on MLO mammography images appears prominently similar to suspicious areas. For this reason, Computer-Aided Detection (CAD) systems remove this region to reduce false-positive rates in the mass detection process. In some cases, the pectoral muscle region is exposed to distortions due to the superposition effects caused by the mammography technique. As a result, segmentation error rates of the pectoral muscle region, whose characteristic features are deteriorated, appear. In this study, a method to identify impaired pectoral muscle regions with MobileNetV2 backboned U-Net Deep Learning method is proposed. The proposed method was tested on 84 and 201 mammography images taken from both MIAS and InBreast databases and segmented with 1.81% and 1.92% false-negative (FN) and 0.25% and 0.37% false positive (FP) rates, respectively. Particularly for distorted pectoral muscle regions, the proposed method has been shown to outperform some pioneering studies in this area.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126467995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-15DOI: 10.1109/SIU55565.2022.9864793
H. Akçay, Emrah Onat
In this paper, the fusion of distance and velocity measurements using radar and GPS is discussed. A Kalman Filter (KF) was designed for the fusion of the measurement results obtained with these different systems. The designed model was used to estimate the position and velocity of a runner. Different scenarios were produced and tested, such as error-free measurements for the entire time interval, unexpected measurements from radar or GPS satellites for a certain period of time. Root Mean Square Error values were calculated and the success of position and velocity estimations were examined. It has been observed that the designed Kalman Filter predictions are more successful than radar and GPS systems.
{"title":"Position and Velocity Detection with RADAR and GPS Fusion","authors":"H. Akçay, Emrah Onat","doi":"10.1109/SIU55565.2022.9864793","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864793","url":null,"abstract":"In this paper, the fusion of distance and velocity measurements using radar and GPS is discussed. A Kalman Filter (KF) was designed for the fusion of the measurement results obtained with these different systems. The designed model was used to estimate the position and velocity of a runner. Different scenarios were produced and tested, such as error-free measurements for the entire time interval, unexpected measurements from radar or GPS satellites for a certain period of time. Root Mean Square Error values were calculated and the success of position and velocity estimations were examined. It has been observed that the designed Kalman Filter predictions are more successful than radar and GPS systems.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126566619","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}
Photoacoustic microscopy is a new medical imaging technique that has begun to be used for imaging skin tissue. The tissue viewed in photoacoustic microscopy is divided into pixels and the photoacoustic signal from each pixel is displayed, and depth information cannot be observed in these images. In this study, in order to determine the depths of the pigments from the data obtained by photoacoustic microscopy of the pigmented regions in the three-layer skin environment, a method was developed that first determines the boundaries of the pigmented regions. Photoacoustic microscopy data were obtained by making measurements from the skin model produced by placing inks in a three-layer PDMS phantom in a laboratory environment. Depth information, location and boundary information of the pigmented regions were determined from the image obtained by interpreting the maximum amplitude photoacoustic signal value and the time information of this value with the gradient calculation.
{"title":"Fotoakustik Mikroskopi Doku Görüntülerinde Pigment Sınırı ve Derinlik Belirleme Tekniği Pigment Boundary and Depth Determination Technique for Photoacoustic Microscopy Image of Tissue","authors":"Sıla Köksal, Başak Zeynep Ergüven, Alper Güzel, Özgür Özdemir","doi":"10.1109/SIU55565.2022.9864739","DOIUrl":"https://doi.org/10.1109/SIU55565.2022.9864739","url":null,"abstract":"Photoacoustic microscopy is a new medical imaging technique that has begun to be used for imaging skin tissue. The tissue viewed in photoacoustic microscopy is divided into pixels and the photoacoustic signal from each pixel is displayed, and depth information cannot be observed in these images. In this study, in order to determine the depths of the pigments from the data obtained by photoacoustic microscopy of the pigmented regions in the three-layer skin environment, a method was developed that first determines the boundaries of the pigmented regions. Photoacoustic microscopy data were obtained by making measurements from the skin model produced by placing inks in a three-layer PDMS phantom in a laboratory environment. Depth information, location and boundary information of the pigmented regions were determined from the image obtained by interpreting the maximum amplitude photoacoustic signal value and the time information of this value with the gradient calculation.","PeriodicalId":115446,"journal":{"name":"2022 30th Signal Processing and Communications Applications Conference (SIU)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126084805","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：