Pub Date : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558914
Emre Sasmaz, F. Tek
Many traders believe in and use Twitter tweets to guide their daily cryptocurrency trading. In this project, we investigated the feasibility of automated sentiment analysis for cryptocurrencies. For the study, we targeted one cryptocurrency (NEO) altcoin and collected related data. The data collection and cleaning were essential components of the study. First, the last five years of daily tweets with NEO hashtags were obtained from Twitter. The collected tweets were then filtered to contain or mention only NEO. We manually tagged a subset of the tweets with positive, negative, and neutral sentiment labels. We trained and tested a Random Forest classifier on the labeled data where the test set accuracy reached 77%. In the second phase of the study, we investigated whether the daily sentiment of the tweets was correlated with the NEO price. We found positive correlations between the number of tweets and the daily prices, and between the prices of different crypto coins. We share the data publicly.
{"title":"Tweet Sentiment Analysis for Cryptocurrencies","authors":"Emre Sasmaz, F. Tek","doi":"10.1109/UBMK52708.2021.9558914","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558914","url":null,"abstract":"Many traders believe in and use Twitter tweets to guide their daily cryptocurrency trading. In this project, we investigated the feasibility of automated sentiment analysis for cryptocurrencies. For the study, we targeted one cryptocurrency (NEO) altcoin and collected related data. The data collection and cleaning were essential components of the study. First, the last five years of daily tweets with NEO hashtags were obtained from Twitter. The collected tweets were then filtered to contain or mention only NEO. We manually tagged a subset of the tweets with positive, negative, and neutral sentiment labels. We trained and tested a Random Forest classifier on the labeled data where the test set accuracy reached 77%. In the second phase of the study, we investigated whether the daily sentiment of the tweets was correlated with the NEO price. We found positive correlations between the number of tweets and the daily prices, and between the prices of different crypto coins. We share the data publicly.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126282312","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9559033
Cengiz Kına, Serkan Kaya, Berkan Aydilek
No-wait flowshop scheduling is an optimization problem that finds wide application in the chemical industry, pharmaceutical industry, steel melting and casting industries. Flight scheduling, operating room scheduling, train line scheduling are a few examples of no-wait scheduling problems. Such problems are called NP-Hard optimization problems in the literature. Researchers have developed various methods to solve such problems. In this study, a gray wolf optimization algorithm is presented to minimize the maximum completion time for nowait flow shop scheduling problems. The local search algorithm has been adapted and hybridized in order to prevent the algorithm from getting stuck in local optima and to enable it to search in the global area. In addition, in order to increase the solution variety and quality of the proposed algorithm, the majority of the initial populations were created with sorting rules instead of random generation. It has been observed that the algorithm tested with the problem sets in the literature gives effective results compared to other methods compared.
{"title":"Hybrid Gray Wolf Algorithm for No Wait Flow Shop Scheduling Problems","authors":"Cengiz Kına, Serkan Kaya, Berkan Aydilek","doi":"10.1109/UBMK52708.2021.9559033","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9559033","url":null,"abstract":"No-wait flowshop scheduling is an optimization problem that finds wide application in the chemical industry, pharmaceutical industry, steel melting and casting industries. Flight scheduling, operating room scheduling, train line scheduling are a few examples of no-wait scheduling problems. Such problems are called NP-Hard optimization problems in the literature. Researchers have developed various methods to solve such problems. In this study, a gray wolf optimization algorithm is presented to minimize the maximum completion time for nowait flow shop scheduling problems. The local search algorithm has been adapted and hybridized in order to prevent the algorithm from getting stuck in local optima and to enable it to search in the global area. In addition, in order to increase the solution variety and quality of the proposed algorithm, the majority of the initial populations were created with sorting rules instead of random generation. It has been observed that the algorithm tested with the problem sets in the literature gives effective results compared to other methods compared.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124998206","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558969
A. Öztaş, Dorukhan Boncukçu, Ege Özteke, M. Demir, A. Mirici, P. Mutlu
The Covid-19 virus has made a major impact on the world and is still spreading rapidly. A reliable solution to prevent further damage, early diagnosis of coronavirus patients are incredibly important. While chest X-Ray diagnosis is the easiest and fastest solution for this, an average radiologist has only a 75% to 85% accuracy when evaluating X-Ray data, thus it is desirable to achieve an accurate artificial network for this. Throughout this study, chest X-Ray data and blood routine test data are utilised and compared. X-Ray data consists of 5000 chest X-Ray images which are gathered from an open-source research and from a local hospital in which both have anonymous data. The blood test results were also taken from the same hospital. For the chest X-Ray diagnosis we utilised two of the popular convolutional neural networks, which are Resnet18 and Squeezenet and concluded that Resnet18 provided slightly more accurate results, while both having almost 98% accuracy. For blood test diagnosis, a feed-forward multi layer neural network was used. Even though it was worked on an insufficient dataset, 72% accuracy was obtained, thus making it a feasible option for further research. Hence, we concluded that in general chest X-Ray diagnosis is preferable over routine blood test diagnosis and the usage of AI yields better approximate results than humans.
{"title":"Covid19 Diagnosis: Comparative Approach Between Chest X-Ray and Blood Test Data","authors":"A. Öztaş, Dorukhan Boncukçu, Ege Özteke, M. Demir, A. Mirici, P. Mutlu","doi":"10.1109/UBMK52708.2021.9558969","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558969","url":null,"abstract":"The Covid-19 virus has made a major impact on the world and is still spreading rapidly. A reliable solution to prevent further damage, early diagnosis of coronavirus patients are incredibly important. While chest X-Ray diagnosis is the easiest and fastest solution for this, an average radiologist has only a 75% to 85% accuracy when evaluating X-Ray data, thus it is desirable to achieve an accurate artificial network for this. Throughout this study, chest X-Ray data and blood routine test data are utilised and compared. X-Ray data consists of 5000 chest X-Ray images which are gathered from an open-source research and from a local hospital in which both have anonymous data. The blood test results were also taken from the same hospital. For the chest X-Ray diagnosis we utilised two of the popular convolutional neural networks, which are Resnet18 and Squeezenet and concluded that Resnet18 provided slightly more accurate results, while both having almost 98% accuracy. For blood test diagnosis, a feed-forward multi layer neural network was used. Even though it was worked on an insufficient dataset, 72% accuracy was obtained, thus making it a feasible option for further research. Hence, we concluded that in general chest X-Ray diagnosis is preferable over routine blood test diagnosis and the usage of AI yields better approximate results than humans.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125270333","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558999
Zh.B. Sadirmekova, J. Tussupov, M.A. Sambetbaveva, A. Yerimbetova, Y.A. Zaeorulko
The purpose of this work is to develop methods, technologies and tools for creating and maintaining intelligent scientific and educational Internet resources (ISEIR) based on a service-oriented approach and Semantic Web technologies. The main purpose of ISEIR is to provide meaningful access to scientific and educational information resources of a given field of knowledge and integrated information processing services. According to the preliminary concept, an intelligent scientific and educational Internet resource will be an information system accessible via the Internet, which provides ontology-based systematization and integration of scientific knowledge, data and information resources into a single information space, meaningful effective access to them, as well as supporting their use in solving various scientific and educational tasks. ISEIR is equipped with an ergonomic web-based user interface and special editors designed to manage the knowledge integrated into it. The proposed approach to the construction of intelligent scientific and educational Internet resources is the basis of the developed technology for creating and maintaining information environments for distributed learning.
{"title":"Features of The Development of Intelligent Scientific and Educational Internet Resources","authors":"Zh.B. Sadirmekova, J. Tussupov, M.A. Sambetbaveva, A. Yerimbetova, Y.A. Zaeorulko","doi":"10.1109/UBMK52708.2021.9558999","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558999","url":null,"abstract":"The purpose of this work is to develop methods, technologies and tools for creating and maintaining intelligent scientific and educational Internet resources (ISEIR) based on a service-oriented approach and Semantic Web technologies. The main purpose of ISEIR is to provide meaningful access to scientific and educational information resources of a given field of knowledge and integrated information processing services. According to the preliminary concept, an intelligent scientific and educational Internet resource will be an information system accessible via the Internet, which provides ontology-based systematization and integration of scientific knowledge, data and information resources into a single information space, meaningful effective access to them, as well as supporting their use in solving various scientific and educational tasks. ISEIR is equipped with an ergonomic web-based user interface and special editors designed to manage the knowledge integrated into it. The proposed approach to the construction of intelligent scientific and educational Internet resources is the basis of the developed technology for creating and maintaining information environments for distributed learning.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"496 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123068586","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9559031
Mandana Fasounaki, Emirhan Burak Yüce, Serkan öncül, G. Ince
With the widespread use of voice-controlling services and devices, the research for developing robust and fast systems for automatic speaker identification had accelerated. In this paper, we present a Convolutional Neural Network (CNN) architecture for text-independent automatic speaker identification. The primary purpose is to identify a speaker, among many others, using a short speech segment. Most of the current researches focus on deep CNNs, which were initially designed for computer vision tasks. Besides, most of the existing speaker identification methods require audio samples longer than 3 seconds in the query phase for achieving a high accuracy. We created a CNN architecture appropriate for voice and speech-related classification tasks. We propose an optimum model that achieves 99.5% accuracy on LibriSpeech and 90% accuracy on VoxCeleb 1 dataset using only 1-second test utterances in our experiments.
{"title":"CNN-based Text-independent Automatic Speaker Identification Using Short Utterances","authors":"Mandana Fasounaki, Emirhan Burak Yüce, Serkan öncül, G. Ince","doi":"10.1109/UBMK52708.2021.9559031","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9559031","url":null,"abstract":"With the widespread use of voice-controlling services and devices, the research for developing robust and fast systems for automatic speaker identification had accelerated. In this paper, we present a Convolutional Neural Network (CNN) architecture for text-independent automatic speaker identification. The primary purpose is to identify a speaker, among many others, using a short speech segment. Most of the current researches focus on deep CNNs, which were initially designed for computer vision tasks. Besides, most of the existing speaker identification methods require audio samples longer than 3 seconds in the query phase for achieving a high accuracy. We created a CNN architecture appropriate for voice and speech-related classification tasks. We propose an optimum model that achieves 99.5% accuracy on LibriSpeech and 90% accuracy on VoxCeleb 1 dataset using only 1-second test utterances in our experiments.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133347274","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558946
Melike Dokgöz, Y. Yaslan
With the increase of conventional vehicles and carbon emission from them boosted the need for electrical vehicles (EV). One of the major components of the EVs are their batteries and the commercialization of EVs are affected by their battery technology and performance. It is also obvious that the range of an EV is mainly affected by the lifetime of its battery. Estimation of the battery cycle life in the early cycles is one of the most important challenges for maximization of the EVs range. Charge-discharge cycles affect battery lifetime of the EV which also made the estimation of battery life cycle a matter of interest. In this study, different machine learning models are applied to predict the lifecycle of a battery at early stages of usage. Detailed experiments have been performed to analyze the prediction accuracy at early cycle numbers. Experimental results show that the error rate in cycle life estimation decreased from 9.2 to 2.4% using Adaptive Boosting method.
{"title":"A Comparison of Machine Learning Algorithms on Lithium-ion Battery Cycle Life Prediction","authors":"Melike Dokgöz, Y. Yaslan","doi":"10.1109/UBMK52708.2021.9558946","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558946","url":null,"abstract":"With the increase of conventional vehicles and carbon emission from them boosted the need for electrical vehicles (EV). One of the major components of the EVs are their batteries and the commercialization of EVs are affected by their battery technology and performance. It is also obvious that the range of an EV is mainly affected by the lifetime of its battery. Estimation of the battery cycle life in the early cycles is one of the most important challenges for maximization of the EVs range. Charge-discharge cycles affect battery lifetime of the EV which also made the estimation of battery life cycle a matter of interest. In this study, different machine learning models are applied to predict the lifecycle of a battery at early stages of usage. Detailed experiments have been performed to analyze the prediction accuracy at early cycle numbers. Experimental results show that the error rate in cycle life estimation decreased from 9.2 to 2.4% using Adaptive Boosting method.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125548174","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9559006
Emre Ertürk, Murat Doğan, Ümit Kadiroğlu, Enis Karaarslan
Cultural heritage assets are in danger of extinction or damage due to lack of publicity and financial problems. Technological advances can play a role in their preservation and promotion. This study aims to create a blockchain-based cultural property protection system which we named the Heirloom. The proposed system uses blockchain and IPFS. This system will allow foundations to receive funding to protect cultural assets without using an intermediary. The cultural assets are transformed into unique digital items using the NFT (Non-Fungible Token) technology. The metadata of the created NFTs is stored in the distributed file system IPFS (InterPlanetary File System). An autonomous working system is provided with smart contracts. The supporters give donations to earn their share of protection and maintenance rights. The proof of concept implementation is promising. A case study on protecting old olive trees in Milas has also started with a local foundation. Possible outcomes will be the ease of getting funds for preserving cultural heritage and increasing awareness. Future studies will include working on different methods for decreasing the costs of the system and integrating augmented and virtual reality technologies.
{"title":"NFT based Fundraising System for Preserving Cultural Heritage: Heirloom","authors":"Emre Ertürk, Murat Doğan, Ümit Kadiroğlu, Enis Karaarslan","doi":"10.1109/UBMK52708.2021.9559006","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9559006","url":null,"abstract":"Cultural heritage assets are in danger of extinction or damage due to lack of publicity and financial problems. Technological advances can play a role in their preservation and promotion. This study aims to create a blockchain-based cultural property protection system which we named the Heirloom. The proposed system uses blockchain and IPFS. This system will allow foundations to receive funding to protect cultural assets without using an intermediary. The cultural assets are transformed into unique digital items using the NFT (Non-Fungible Token) technology. The metadata of the created NFTs is stored in the distributed file system IPFS (InterPlanetary File System). An autonomous working system is provided with smart contracts. The supporters give donations to earn their share of protection and maintenance rights. The proof of concept implementation is promising. A case study on protecting old olive trees in Milas has also started with a local foundation. Possible outcomes will be the ease of getting funds for preserving cultural heritage and increasing awareness. Future studies will include working on different methods for decreasing the costs of the system and integrating augmented and virtual reality technologies.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124869159","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558976
Yasir Kilic, Ahmet Büyükeke
Graph Convolutional Neural Networks (GCNs) are highly popular in recent years. It gives very successful results for various natural language processing (NLP) tasks such as sentiment classification. It has recently been shown to be effective and successful models to solve sentiment classification problem of texts. However, there is no research demonstrating the performance of this model on Turkish texts. In this study, we observe performance of the GCN model on the sentiment classification problem of Turkish texts as first research. Since the structure of Turkish language is agglutinative, different preprocessing approaches are presented and performance results on three real-world Turkish sentiment datasets are shown. It is observed that the TripAdv dataset, which was used in this study, yielded a 0.76 F-measure value. This can be considered a reasonable success for a sentiment classification with three sentiment classes. On the other hand, this study is presented as an exploratory case study in preparation for more detailed and extensive research in the future.
{"title":"An Exploratory Case Study for Turkish Sentiment Classification Using Graph Convolutional Neural Networks","authors":"Yasir Kilic, Ahmet Büyükeke","doi":"10.1109/UBMK52708.2021.9558976","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558976","url":null,"abstract":"Graph Convolutional Neural Networks (GCNs) are highly popular in recent years. It gives very successful results for various natural language processing (NLP) tasks such as sentiment classification. It has recently been shown to be effective and successful models to solve sentiment classification problem of texts. However, there is no research demonstrating the performance of this model on Turkish texts. In this study, we observe performance of the GCN model on the sentiment classification problem of Turkish texts as first research. Since the structure of Turkish language is agglutinative, different preprocessing approaches are presented and performance results on three real-world Turkish sentiment datasets are shown. It is observed that the TripAdv dataset, which was used in this study, yielded a 0.76 F-measure value. This can be considered a reasonable success for a sentiment classification with three sentiment classes. On the other hand, this study is presented as an exploratory case study in preparation for more detailed and extensive research in the future.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132960458","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558995
I. Akman, Ç. Turhan, Tuna Hacaloglu
Software development involves a significant amount of team effort where collaboration and communication of the team members are crucial. The team meetings are core activities in all stages of the software development process. Even though these meetings often are conducted face-to-face (F2F) with a lack of technology utilization, changing global conditions such as the COVID-19 pandemic require other solutions urgently without interrupting the software development schedule. For this purpose, online collaborative tools provide new opportunities for software teams to work together avoiding waste in time and resources and the relevant literature is immature. This study aims to assess the factors affecting the integration of online collaborative tools to SE practices with a special reference to review meetings. For this purpose, a sample of 73 SE sophomore and junior students who are future software professionals participated in experimental review meetings based on predefined scenarios. The findings indicate that the utilization of OCT’s has positive effects on the participants’ actual performance and improves the interaction between team members compared to F2F meetings.
{"title":"Utilization of Online Collaborative Tools in Software Engineering: An Empirical Study on Review Meetings","authors":"I. Akman, Ç. Turhan, Tuna Hacaloglu","doi":"10.1109/UBMK52708.2021.9558995","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558995","url":null,"abstract":"Software development involves a significant amount of team effort where collaboration and communication of the team members are crucial. The team meetings are core activities in all stages of the software development process. Even though these meetings often are conducted face-to-face (F2F) with a lack of technology utilization, changing global conditions such as the COVID-19 pandemic require other solutions urgently without interrupting the software development schedule. For this purpose, online collaborative tools provide new opportunities for software teams to work together avoiding waste in time and resources and the relevant literature is immature. This study aims to assess the factors affecting the integration of online collaborative tools to SE practices with a special reference to review meetings. For this purpose, a sample of 73 SE sophomore and junior students who are future software professionals participated in experimental review meetings based on predefined scenarios. The findings indicate that the utilization of OCT’s has positive effects on the participants’ actual performance and improves the interaction between team members compared to F2F meetings.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"306 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116272336","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 : 2021-09-15DOI: 10.1109/UBMK52708.2021.9558923
M. Aktaş, H. Ateş
Object detection and tracking from airborne imagery draws attention to the parallel development of UAV systems and computer vision technologies. Aerial imagery has its own unique challenges that differ from the training set of modern-day object detectors, since it is made of images of larger areas compared to the regular datasets and the objects are very small on the contrary. These problems do not allow us to use common object detection models. The main purpose of this paper is to make modifications to the Faster-RCNN (FRCNN) model, then leverage it for small object detection and tracking from the aerial imagery. It is aimed to use both spatial and temporal information from the image sequence, as appearance information alone is insufficient. The anchors in the Region Proposal Network (RPN) stage will be adjusted for small objects. Also, intersection over union (IoU) is optimized for small objects. After improving detection performance, The DeepSORT algorithm is inserted right after the Region of Interest (ROI Head) to track the objects. The results show that the proposed model has good performance on the VisDrone-2019 dataset. Detection performance becomes considerably better than the original FRCNN and the algorithms that are evaluated in the VisDrone-2019 VID challenge. After completing the proposed modifications, the AP-AP50 values reached 14.07-29.41 from 8.08-18.70, which means approximately 75% improvement.
{"title":"Small Object Detection and Tracking from Aerial Imagery","authors":"M. Aktaş, H. Ateş","doi":"10.1109/UBMK52708.2021.9558923","DOIUrl":"https://doi.org/10.1109/UBMK52708.2021.9558923","url":null,"abstract":"Object detection and tracking from airborne imagery draws attention to the parallel development of UAV systems and computer vision technologies. Aerial imagery has its own unique challenges that differ from the training set of modern-day object detectors, since it is made of images of larger areas compared to the regular datasets and the objects are very small on the contrary. These problems do not allow us to use common object detection models. The main purpose of this paper is to make modifications to the Faster-RCNN (FRCNN) model, then leverage it for small object detection and tracking from the aerial imagery. It is aimed to use both spatial and temporal information from the image sequence, as appearance information alone is insufficient. The anchors in the Region Proposal Network (RPN) stage will be adjusted for small objects. Also, intersection over union (IoU) is optimized for small objects. After improving detection performance, The DeepSORT algorithm is inserted right after the Region of Interest (ROI Head) to track the objects. The results show that the proposed model has good performance on the VisDrone-2019 dataset. Detection performance becomes considerably better than the original FRCNN and the algorithms that are evaluated in the VisDrone-2019 VID challenge. After completing the proposed modifications, the AP-AP50 values reached 14.07-29.41 from 8.08-18.70, which means approximately 75% improvement.","PeriodicalId":106516,"journal":{"name":"2021 6th International Conference on Computer Science and Engineering (UBMK)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114723978","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：