Image compression and encryption are two processes that enable telemedicine application of eHealth services. However, performing these operations on the whole content of an image is computationally expensive. This work proposes a method for selective compression and selective encryption of medical images. It is based on lossless compression and encryption of the region of interest (ROI) in medical images. The non-ROI part of the image is compressed in lossy mode and is stored or transmitted as plain data, in order to further reduce the image size and to avoid the computational cost of encrypting huge volumes of medical images. Our analysis shows that the proposed method provides the necessary security and is secured against various attacks. In addition, the compression savings achieved by the proposed method is about 28% while preserving the crucial information in the ROI for correct diagnosis. For a quality factor of 80%, the reconstructed image has a peak signal-to-noise ratio of 42.5 dB. The proposed method requires less computational resources and enables processing of huge volume of image data in low power network.
{"title":"Region-based Selective Compression and Selective Encryption of Medical Images","authors":"Ijaz Ahmad, Seokjoo Shin","doi":"10.1145/3426020.3426027","DOIUrl":"https://doi.org/10.1145/3426020.3426027","url":null,"abstract":"Image compression and encryption are two processes that enable telemedicine application of eHealth services. However, performing these operations on the whole content of an image is computationally expensive. This work proposes a method for selective compression and selective encryption of medical images. It is based on lossless compression and encryption of the region of interest (ROI) in medical images. The non-ROI part of the image is compressed in lossy mode and is stored or transmitted as plain data, in order to further reduce the image size and to avoid the computational cost of encrypting huge volumes of medical images. Our analysis shows that the proposed method provides the necessary security and is secured against various attacks. In addition, the compression savings achieved by the proposed method is about 28% while preserving the crucial information in the ROI for correct diagnosis. For a quality factor of 80%, the reconstructed image has a peak signal-to-noise ratio of 42.5 dB. The proposed method requires less computational resources and enables processing of huge volume of image data in low power network.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115256914","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}
Sumet Darapisut, Komate Amphawan, S. Rimcharoen, Nutthanon Leelathakul
Location-Based Recommender Systems (LBRSs) have gained popularity in recent years as users tend to make decisions based on what are shared in social medias. Such systems depend on each user's historical behavioral information (or user profile) to determine users’ interests. However, it is impossible for new users to have the profile, making it difficult and challenging to recommend interesting locations (also known as a cold start problem). In order to tackle this issue, we propose an enhanced method, called N-most interesting location-based recommender system (NILR), which effectively recommends the N-most preferred places for each user without leveraging her profile. We also introduce a novel metric (so called interestingness score) to measure locations’ attractiveness. The metric takes into account both check-in frequencies and number of return visits of previous users already in the system. The method ranks the top-N locations based on the combination of the traditional HITS-based model (Hypertext Induced Topic Search) [1] and the proposed NILR. The results of the experiments on Foursquare dataset reveal that our proposed location recommender system and raking method perform effectively and efficiently, and outperform the HITS model in terms of accuracies and rankings.
{"title":"NILR:N-Most Interesting Location-based Recommender System","authors":"Sumet Darapisut, Komate Amphawan, S. Rimcharoen, Nutthanon Leelathakul","doi":"10.1145/3426020.3426145","DOIUrl":"https://doi.org/10.1145/3426020.3426145","url":null,"abstract":"Location-Based Recommender Systems (LBRSs) have gained popularity in recent years as users tend to make decisions based on what are shared in social medias. Such systems depend on each user's historical behavioral information (or user profile) to determine users’ interests. However, it is impossible for new users to have the profile, making it difficult and challenging to recommend interesting locations (also known as a cold start problem). In order to tackle this issue, we propose an enhanced method, called N-most interesting location-based recommender system (NILR), which effectively recommends the N-most preferred places for each user without leveraging her profile. We also introduce a novel metric (so called interestingness score) to measure locations’ attractiveness. The metric takes into account both check-in frequencies and number of return visits of previous users already in the system. The method ranks the top-N locations based on the combination of the traditional HITS-based model (Hypertext Induced Topic Search) [1] and the proposed NILR. The results of the experiments on Foursquare dataset reveal that our proposed location recommender system and raking method perform effectively and efficiently, and outperform the HITS model in terms of accuracies and rankings.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126086757","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}
{"title":"Blind Image Watermarking Scheme for Image Authentication and Restoration with Improved Restoration Features","authors":"Rishi Sinhal, I. Ansari, C. Ahn","doi":"10.1145/3426020.3426074","DOIUrl":"https://doi.org/10.1145/3426020.3426074","url":null,"abstract":"","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121006292","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}
Vo Hoang Trong, Gwanghyun Yu, H. Nguyen, Ju-Hwan Lee, Dang Thanh Vu, Jinyoung Kim
In this paper, we use a homomorphic filter and Deep Neural Network (DNN) for apple trees diseases classification. The homomorphic filter is used as the preprocessing step to enhance appearances of low-level features in an image, which can improve performances of DNN for classification. We experiment on the Plant pathology dataset, which also the Plant Pathology Challenge on Kaggle. The result shows that using a homomorphic filter gets 0.9116 on the accuracy.
{"title":"A study on applying homomorphic filter and Deep Neural Network for apple trees diseases classification","authors":"Vo Hoang Trong, Gwanghyun Yu, H. Nguyen, Ju-Hwan Lee, Dang Thanh Vu, Jinyoung Kim","doi":"10.1145/3426020.3426042","DOIUrl":"https://doi.org/10.1145/3426020.3426042","url":null,"abstract":"In this paper, we use a homomorphic filter and Deep Neural Network (DNN) for apple trees diseases classification. The homomorphic filter is used as the preprocessing step to enhance appearances of low-level features in an image, which can improve performances of DNN for classification. We experiment on the Plant pathology dataset, which also the Plant Pathology Challenge on Kaggle. The result shows that using a homomorphic filter gets 0.9116 on the accuracy.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128143794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The adaptation of sensor-cloud in healthcare infrastructure has enabled the use of machine learning techniques for efficient healthcare provisioning. In the context of the smart healthcare system, biomedical wireless sensor networks (BWSNs) are one of the key infrastructure enabling the development of healthcare applications and services. With the increasing number of healthcare information collected through BWSN, different types of medical data can be exploited to design a predictive analytics system, thereby transforming the traditional healthcare system. In this paper, we propose and highlight smart healthcare monitoring framework using state-of-the-art technologies. In particular, we focus on sensor-cloud computing and machine learning as emerging technologies, which are suitable for a proactive healthcare system by the advancement in various aspects, including computational capability, data storage, and learning techniques. Besides, we describe the components of our proposed framework with data analysis techniques and sensor-cloud layered architecture.
{"title":"Towards Sensor-Cloud Based Efficient Smart Healthcare Monitoring Framework using Machine Learning","authors":"Khadak Singh Bhandari, Changho Seo, G. Cho","doi":"10.1145/3426020.3426138","DOIUrl":"https://doi.org/10.1145/3426020.3426138","url":null,"abstract":"The adaptation of sensor-cloud in healthcare infrastructure has enabled the use of machine learning techniques for efficient healthcare provisioning. In the context of the smart healthcare system, biomedical wireless sensor networks (BWSNs) are one of the key infrastructure enabling the development of healthcare applications and services. With the increasing number of healthcare information collected through BWSN, different types of medical data can be exploited to design a predictive analytics system, thereby transforming the traditional healthcare system. In this paper, we propose and highlight smart healthcare monitoring framework using state-of-the-art technologies. In particular, we focus on sensor-cloud computing and machine learning as emerging technologies, which are suitable for a proactive healthcare system by the advancement in various aspects, including computational capability, data storage, and learning techniques. Besides, we describe the components of our proposed framework with data analysis techniques and sensor-cloud layered architecture.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127289925","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}
This study is intended to develop an artificial intelligence model capable of recognizing and detecting heavy equipment to compensate for visual observation while driving. The total number of data collected was 6,700 images, of which 5,820 were used as training data and the remaining 880 images as a set of testing data. The YOLOv3 Network proposed in this study shows improved performance compared to the existing YOLOv3 and YOLOv2 with performance indicators for F1-Score 84.65%, Precision 86.94%, and Recall 82.47% in the detection of heavy equipment in the testing data set. Since the heavy equipment detection model proposed in the study is designed to enable real-time detection, it is expected to be helpful in patrolling construction sites in areas where gas pipes are buried.
{"title":"Detection Model of Heavy Equipment Using YOLOv3 while Driving","authors":"Won-Seok Lee, H. Lee, Choong Kwon Lee, K. Ko","doi":"10.1145/3426020.3426082","DOIUrl":"https://doi.org/10.1145/3426020.3426082","url":null,"abstract":"This study is intended to develop an artificial intelligence model capable of recognizing and detecting heavy equipment to compensate for visual observation while driving. The total number of data collected was 6,700 images, of which 5,820 were used as training data and the remaining 880 images as a set of testing data. The YOLOv3 Network proposed in this study shows improved performance compared to the existing YOLOv3 and YOLOv2 with performance indicators for F1-Score 84.65%, Precision 86.94%, and Recall 82.47% in the detection of heavy equipment in the testing data set. Since the heavy equipment detection model proposed in the study is designed to enable real-time detection, it is expected to be helpful in patrolling construction sites in areas where gas pipes are buried.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126421905","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}
While most embedding methods in the Korean language focus on morpheme unit to alleviate the out of vocabulary problem, recent researches in the English use the subword unit for embedding. Considering that a word is composed of subwords, which have a partial role in a word, we hypothesize that a sequence of subwords enriches the meaning of a sentence than a sequence of words or morphemes. We propose a sentence embedding method based on a sequence of subwords in the Korean language. We evaluate the effectiveness of our sentence embedding method on binary sentiment classification using Naver Sentiment Movie Corpus. By comparing the performance of sentence embedding based on a sequence of words, morphemes, and subwords, we verify that sentence embedding based on a sequence of subwords is more robust to the out of vocabulary problem than the others.
{"title":"Subword-based Sentence Representation Model for Sentiment Classification","authors":"Danbi Cho, Hyunyoung Lee, Seungshik Kang","doi":"10.1145/3426020.3426046","DOIUrl":"https://doi.org/10.1145/3426020.3426046","url":null,"abstract":"While most embedding methods in the Korean language focus on morpheme unit to alleviate the out of vocabulary problem, recent researches in the English use the subword unit for embedding. Considering that a word is composed of subwords, which have a partial role in a word, we hypothesize that a sequence of subwords enriches the meaning of a sentence than a sequence of words or morphemes. We propose a sentence embedding method based on a sequence of subwords in the Korean language. We evaluate the effectiveness of our sentence embedding method on binary sentiment classification using Naver Sentiment Movie Corpus. By comparing the performance of sentence embedding based on a sequence of words, morphemes, and subwords, we verify that sentence embedding based on a sequence of subwords is more robust to the out of vocabulary problem than the others.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121484649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The increased penetration of renewable energy resources has become one of the driving forces with a rapid growth of the number of prosumers in the energy market. The aim of maximizing prosumer revenue and welfare has been one of the motivations due to a growing interest in the field of Transactive Energy (TE). This paper presents a novel method for determining and forecasting a prosumer's energy generation performance by using Monte Carlo Simulation (MCS) with Geometric Brownian Motion (GBM). With the implementation of GBM, the average generation of a prosumer is forecasted based on multiple timesteps to visualize multiple outcomes. In addition, the computation of potential payoffs/revenues to be received by the prosumers based on average energy generation and randomized prices has been carried out. Finally, the experimental results of this work provide beneficial insights into future decisions concerning the improvement of prosumer performance.
{"title":"Determining Prosumer Energy Generation Performance as Basis for Peer-to-Peer Energy Trading Decisions using Monte Carlo Simulation","authors":"Ralph Voltaire J. Dayot, Hyuntae Kim, In-ho Ra","doi":"10.1145/3426020.3426086","DOIUrl":"https://doi.org/10.1145/3426020.3426086","url":null,"abstract":"The increased penetration of renewable energy resources has become one of the driving forces with a rapid growth of the number of prosumers in the energy market. The aim of maximizing prosumer revenue and welfare has been one of the motivations due to a growing interest in the field of Transactive Energy (TE). This paper presents a novel method for determining and forecasting a prosumer's energy generation performance by using Monte Carlo Simulation (MCS) with Geometric Brownian Motion (GBM). With the implementation of GBM, the average generation of a prosumer is forecasted based on multiple timesteps to visualize multiple outcomes. In addition, the computation of potential payoffs/revenues to be received by the prosumers based on average energy generation and randomized prices has been carried out. Finally, the experimental results of this work provide beneficial insights into future decisions concerning the improvement of prosumer performance.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122042729","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}
This case study describes the evaluation of a portfolio of investments in IT projects in a manufacturing SME in the Czech Republic. The most important 13 of the 80 projects were assessed and evaluated, based on this procedure. These projects at a cost of 572 thousand CZK (1 Euro is approx. 26 Czech crowns (CZK)), proposed projects can create a net present value of 1,576 thousands CZK. The output is a RACI table of assigned responsibilities for the IT manager. The whole workflow is based on the Val IT 2.0 framework. As the company is in the initial phase of ITG maturity, it is recommended to continue working on these principles with evaluation and project management and to develop them according to individual needs. The results of the work were achieved through semi-standardized interviews with internal stakeholders, a study of the literature and a qualified estimate. This paper is based on the thesis [1].
{"title":"Case Study: Continual Evaluation of IT Process Portfolio in SME based on Val IT 2.0","authors":"Jan Lacina, Libor Mesícek, H. Ko, S. Pan","doi":"10.1145/3426020.3426023","DOIUrl":"https://doi.org/10.1145/3426020.3426023","url":null,"abstract":"This case study describes the evaluation of a portfolio of investments in IT projects in a manufacturing SME in the Czech Republic. The most important 13 of the 80 projects were assessed and evaluated, based on this procedure. These projects at a cost of 572 thousand CZK (1 Euro is approx. 26 Czech crowns (CZK)), proposed projects can create a net present value of 1,576 thousands CZK. The output is a RACI table of assigned responsibilities for the IT manager. The whole workflow is based on the Val IT 2.0 framework. As the company is in the initial phase of ITG maturity, it is recommended to continue working on these principles with evaluation and project management and to develop them according to individual needs. The results of the work were achieved through semi-standardized interviews with internal stakeholders, a study of the literature and a qualified estimate. This paper is based on the thesis [1].","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123329890","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}
In this study, image search techniques are suggested based on dual rearrangement of object corner. Suggested algorithm is proceeded in the following stages. First, edges and corner points are extracted in the image. Then, the dispersion value and the number of white level are extracted in each of the corner areas. In addition, correlogram feature table is configured by using the dispersion value and the number of white level measuring the similarity of them. Suggested technique turned out to be outstanding in performance from objects with distinct structure in the image and strong against movement or rotation of an object. In addition, it represented an improved recall by about 0.05 compared to searching form in the use of corner patch histogram.
{"title":"Image Retrieval Based on Dual Rearrangement of Object Corner","authors":"Youngeun An, Sungbum Pan, Taeyeun Kim","doi":"10.1145/3426020.3426059","DOIUrl":"https://doi.org/10.1145/3426020.3426059","url":null,"abstract":"In this study, image search techniques are suggested based on dual rearrangement of object corner. Suggested algorithm is proceeded in the following stages. First, edges and corner points are extracted in the image. Then, the dispersion value and the number of white level are extracted in each of the corner areas. In addition, correlogram feature table is configured by using the dispersion value and the number of white level measuring the similarity of them. Suggested technique turned out to be outstanding in performance from objects with distinct structure in the image and strong against movement or rotation of an object. In addition, it represented an improved recall by about 0.05 compared to searching form in the use of corner patch histogram.","PeriodicalId":305132,"journal":{"name":"The 9th International Conference on Smart Media and Applications","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131479856","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：