Pub Date : 2019-04-08DOI: 10.1109/ICDEW.2019.000-4
Bin Yang, Haiwei Pan, Jieyao Yu, Kun Han, Yanan Wang
Medicine has always been an important area of concern for people's lives. Medical images, as an important basis for doctors to diagnose diseases, has its own particularity. For example, many medical images are often difficult to distinguish due to intra-class variation and inter-class similarity, and medical images have high requirements for processing accuracy. A synergic graph convolutional networks (SGCN) model is proposed for image classification. This model is based on convolutional neural networks on graphs with fast localized spectral filtering. In our model, two graph convolutional networks (GCN) can learn from each other. We choose the Kth-order Chebyshev polynomials of the Laplacian to control K-localized of spectral filters conveniently. Specifically, we concatenate the image representation learned by both GCNs as the input of our synergic deep learning framework to predict whether the pair of input images belong to the same class. The intra-class similarity and inter-class variability of the dataset itself makes the performance of a single graph convolutional neural network better. We evaluated our SGCN model on MNIST and some Brain MRI image classification dataset and achieved advanced performance.
{"title":"Classification of Medical Images with Synergic Graph Convolutional Networks","authors":"Bin Yang, Haiwei Pan, Jieyao Yu, Kun Han, Yanan Wang","doi":"10.1109/ICDEW.2019.000-4","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.000-4","url":null,"abstract":"Medicine has always been an important area of concern for people's lives. Medical images, as an important basis for doctors to diagnose diseases, has its own particularity. For example, many medical images are often difficult to distinguish due to intra-class variation and inter-class similarity, and medical images have high requirements for processing accuracy. A synergic graph convolutional networks (SGCN) model is proposed for image classification. This model is based on convolutional neural networks on graphs with fast localized spectral filtering. In our model, two graph convolutional networks (GCN) can learn from each other. We choose the Kth-order Chebyshev polynomials of the Laplacian to control K-localized of spectral filters conveniently. Specifically, we concatenate the image representation learned by both GCNs as the input of our synergic deep learning framework to predict whether the pair of input images belong to the same class. The intra-class similarity and inter-class variability of the dataset itself makes the performance of a single graph convolutional neural network better. We evaluated our SGCN model on MNIST and some Brain MRI image classification dataset and achieved advanced performance.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126171824","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}
A rapid growth in the smart-wearable industry is making it increasingly important to cater to the Quality of Experience (QoE) requirements of the end-users. In this work, we try to model the relationship between human experience and quality perception in relation to the smart-wearable segment. For this, the concepts of Quality of Data (QoD) and Quality of Information (QoI) are used. Step-counts and heart-rate measurement readings by the wearables are the parameters considered for evaluating the QoD, whereas perceived ease of use, perceived usefulness, and richness in information are the ones taken for evaluating the QoI. A subjective experiment comprising of 40 participants and 5 wearable devices is performed in a free-living condition in order to create the QoE model. We hypothesize QoE to be a function of QoD, and QoI and use a balanced weight technique to formulate the final model. R^2and adjusted-R^2values of 0.65 and 0.63 indicate a reasonable predictive power of the proposed scheme. Based upon the results appropriate recommendations are provided to the different smart-wearable vendors for improving their products, thereby ensuring a greater user-adoption.
{"title":"Quality of Experience Evaluation of Smart-Wearables: A Mathematical Modelling Approach","authors":"Debajyoti Pal, Tuul Triyason, Vijayakumar Varadarajan, Xiangmin Zhang","doi":"10.1109/ICDEW.2019.00-32","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00-32","url":null,"abstract":"A rapid growth in the smart-wearable industry is making it increasingly important to cater to the Quality of Experience (QoE) requirements of the end-users. In this work, we try to model the relationship between human experience and quality perception in relation to the smart-wearable segment. For this, the concepts of Quality of Data (QoD) and Quality of Information (QoI) are used. Step-counts and heart-rate measurement readings by the wearables are the parameters considered for evaluating the QoD, whereas perceived ease of use, perceived usefulness, and richness in information are the ones taken for evaluating the QoI. A subjective experiment comprising of 40 participants and 5 wearable devices is performed in a free-living condition in order to create the QoE model. We hypothesize QoE to be a function of QoD, and QoI and use a balanced weight technique to formulate the final model. R^2and adjusted-R^2values of 0.65 and 0.63 indicate a reasonable predictive power of the proposed scheme. Based upon the results appropriate recommendations are provided to the different smart-wearable vendors for improving their products, thereby ensuring a greater user-adoption.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123755017","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.00007
Xin Ju, Xiaofeng Zhang, W. K. Cheung
With the fast development of social networks, there exists a huge amount of users information as well as their social ties. Such information generally contains sensitive and correlated users' personal data. How to accurately analyze this large and correlated social graph data while protecting users' privacy has become a challenging research issue. In the literature, various research efforts have been made on this topic. Unfortunately, correlation based privacy protection techniques for social graph data have seldom been investigated. To the best of our knowledge, this paper is the first attempt to resolve this research issue. Particularly, this paper first protects users' raw data via local differential privacy technique and then a correlation based privacy protection approach is designed. Last, a K-means algorithm is applied on the perturbed local data and the clustering results are used to generate the synthetic graphs for further data analytics. Experiments are evaluated on two real-world data sets, i.e. Facebook dataset and Enron dataset, and the promising experimental results demonstrate that the proposed approach is superior to the state-of-the-art LDPGen and the baseline method, e.g. the DGG, with respect to the accuracy and utility evaluation criteria.
{"title":"Generating Synthetic Graphs for Large Sensitive and Correlated Social Networks","authors":"Xin Ju, Xiaofeng Zhang, W. K. Cheung","doi":"10.1109/ICDEW.2019.00007","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00007","url":null,"abstract":"With the fast development of social networks, there exists a huge amount of users information as well as their social ties. Such information generally contains sensitive and correlated users' personal data. How to accurately analyze this large and correlated social graph data while protecting users' privacy has become a challenging research issue. In the literature, various research efforts have been made on this topic. Unfortunately, correlation based privacy protection techniques for social graph data have seldom been investigated. To the best of our knowledge, this paper is the first attempt to resolve this research issue. Particularly, this paper first protects users' raw data via local differential privacy technique and then a correlation based privacy protection approach is designed. Last, a K-means algorithm is applied on the perturbed local data and the clustering results are used to generate the synthetic graphs for further data analytics. Experiments are evaluated on two real-world data sets, i.e. Facebook dataset and Enron dataset, and the promising experimental results demonstrate that the proposed approach is superior to the state-of-the-art LDPGen and the baseline method, e.g. the DGG, with respect to the accuracy and utility evaluation criteria.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"68 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126968724","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}
Cooking is a unique endeavor that forms the core of our cultural identity. Culinary systems across the world have evolved over a period of time in the backdrop of complex interplay of diverse sociocultural factors including geographic, climatic and genetic influences. Data-driven investigations can offer interesting insights into the structural and organizational principles of cuisines. Herein, we use a comprehensive repertoire of 158544 recipes from 25 geo-cultural regions across the world to investigate the statistical patterns in combinations of ingredients and their categories. Further, we develop computational models for the evolution of cuisines. Our analysis reveals copy-mutation as a plausible mechanism of culinary evolution. As the world copes with the challenges of diet-linked disorders, knowledge of the key determinants of culinary evolution can drive the creation of novel recipe generation algorithms aimed at dietary interventions for better nutrition and health.
{"title":"Computational Models for the Evolution of World Cuisines","authors":"Rudraksh Tuwani, Nutan Sahoo, Navjot Singh, Ganesh Bagler","doi":"10.1109/ICDEW.2019.00-30","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00-30","url":null,"abstract":"Cooking is a unique endeavor that forms the core of our cultural identity. Culinary systems across the world have evolved over a period of time in the backdrop of complex interplay of diverse sociocultural factors including geographic, climatic and genetic influences. Data-driven investigations can offer interesting insights into the structural and organizational principles of cuisines. Herein, we use a comprehensive repertoire of 158544 recipes from 25 geo-cultural regions across the world to investigate the statistical patterns in combinations of ingredients and their categories. Further, we develop computational models for the evolution of cuisines. Our analysis reveals copy-mutation as a plausible mechanism of culinary evolution. As the world copes with the challenges of diet-linked disorders, knowledge of the key determinants of culinary evolution can drive the creation of novel recipe generation algorithms aimed at dietary interventions for better nutrition and health.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121906038","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.000-6
Xin Jin, Longbin Lai
Subgraph Matching is a fundamental problem in graph analysis, and is widely used in many application scenarios in biology, chemistry and social network. Given a data graph and a query graph, subgraph matching aims to compute all subgraphs of the data graph that are isomorphic to the query graph. The problem is computationally expensive as the core operation it depends on, known as subgraph isomorphism, is NP-complete. In recent years, graph is increasing extensively and it is hard to compute subgraph matching on massive graph data using existing serial algorithm. Meanwhile, there exist distributed solutions, but they are mostly limited to the case where the graphs are unlabelled. In response to this gap, we study the subgraph matching problem in the multi-core environment. From the algorithm level, we propose a multi-core parallel subgraph matching algorithm called MPMatch. From the research level, we explore the concurrent allocation of subgraph matching search space to approach load balancing. We conduct extensive empirical studies on real and synthetic graphs to demonstrate that our techniques improve the performance of serial subgraph matching algorithm via parallelization and well-developed load balancing schema.
{"title":"MPMatch: A Multi-core Parallel Subgraph Matching Algorithm","authors":"Xin Jin, Longbin Lai","doi":"10.1109/ICDEW.2019.000-6","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.000-6","url":null,"abstract":"Subgraph Matching is a fundamental problem in graph analysis, and is widely used in many application scenarios in biology, chemistry and social network. Given a data graph and a query graph, subgraph matching aims to compute all subgraphs of the data graph that are isomorphic to the query graph. The problem is computationally expensive as the core operation it depends on, known as subgraph isomorphism, is NP-complete. In recent years, graph is increasing extensively and it is hard to compute subgraph matching on massive graph data using existing serial algorithm. Meanwhile, there exist distributed solutions, but they are mostly limited to the case where the graphs are unlabelled. In response to this gap, we study the subgraph matching problem in the multi-core environment. From the algorithm level, we propose a multi-core parallel subgraph matching algorithm called MPMatch. From the research level, we explore the concurrent allocation of subgraph matching search space to approach load balancing. We conduct extensive empirical studies on real and synthetic graphs to demonstrate that our techniques improve the performance of serial subgraph matching algorithm via parallelization and well-developed load balancing schema.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"255 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132895420","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.000-1
Monica Senapati, L. Njilla, P. Rao
We propose a method for scalable first-order rule learning on large-scale Twitter data. By learning rules, probabilistic inference queries can be executed to reason over the data to ascertain its veracity. Our method employs a divide-and-conquer approach, graph-based modeling, and data parallel processing during rule learning using a commodity cluster to overcome the problem of slow structure learning on large-scale Twitter data. The first-order predicates (constructed on the posts) are first partitioned in a balanced way by pivoting around users to reduce the chance of missing relevant rules. By constructing a weighted graph and applying graph partitioning, balanced partitions of the ground predicates can be created. Each partition is then processed using an existing structure learning approach to get the set of rules for that partition. We report a preliminary evaluation of our method to show that it offers a promising solution for scalable first-order rule learning on Twitter data.
{"title":"A Method for Scalable First-Order Rule Learning on Twitter Data","authors":"Monica Senapati, L. Njilla, P. Rao","doi":"10.1109/ICDEW.2019.000-1","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.000-1","url":null,"abstract":"We propose a method for scalable first-order rule learning on large-scale Twitter data. By learning rules, probabilistic inference queries can be executed to reason over the data to ascertain its veracity. Our method employs a divide-and-conquer approach, graph-based modeling, and data parallel processing during rule learning using a commodity cluster to overcome the problem of slow structure learning on large-scale Twitter data. The first-order predicates (constructed on the posts) are first partitioned in a balanced way by pivoting around users to reduce the chance of missing relevant rules. By constructing a weighted graph and applying graph partitioning, balanced partitions of the ground predicates can be created. Each partition is then processed using an existing structure learning approach to get the set of rules for that partition. We report a preliminary evaluation of our method to show that it offers a promising solution for scalable first-order rule learning on Twitter data.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114261435","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.00-10
Yu Mao, Haiqing Du, Yong Liu
Video person re-identification is a crucial task due to its applications in visual surveillance and human-computer interaction. The purpose of these kinds of algorithms are to search for the corresponding pedestrian image from a large number of cross-device surveillance videos with a given pedestrian image as a probe. In recent years, more and more scholars have begun to regard this problem as a special type of image retrieval. Existing works mainly focus on extracting representative features from the whole image and integrate those features in a sequence through temporal modeling. However, these approaches rarely consider harnessing local visual cues to enhance the power of image-level feature learning. In this paper, we propose a novel neural network which incorporate human semantic parsing to improve imag-elevel representations. Specifically, the human semantic parsing network is able to segment a human image into multiple parts with fine-grained semantics, and the following attentive feature pooling layer can select most significant body parts to enhance the power of feature representations. The carefully designed experiments on two public datasets show the effectiveness of each components of the proposed deep network, improving state-of-the-art video person sequence retrieval on: iLIDS-VID [1] by ∼13% and PRID-2011 by ∼7% in rank-1.
{"title":"Semantic Parsing and Attentive Feature-Temporal Pooling Network for Video-Based Person Image Retrieval","authors":"Yu Mao, Haiqing Du, Yong Liu","doi":"10.1109/ICDEW.2019.00-10","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00-10","url":null,"abstract":"Video person re-identification is a crucial task due to its applications in visual surveillance and human-computer interaction. The purpose of these kinds of algorithms are to search for the corresponding pedestrian image from a large number of cross-device surveillance videos with a given pedestrian image as a probe. In recent years, more and more scholars have begun to regard this problem as a special type of image retrieval. Existing works mainly focus on extracting representative features from the whole image and integrate those features in a sequence through temporal modeling. However, these approaches rarely consider harnessing local visual cues to enhance the power of image-level feature learning. In this paper, we propose a novel neural network which incorporate human semantic parsing to improve imag-elevel representations. Specifically, the human semantic parsing network is able to segment a human image into multiple parts with fine-grained semantics, and the following attentive feature pooling layer can select most significant body parts to enhance the power of feature representations. The carefully designed experiments on two public datasets show the effectiveness of each components of the proposed deep network, improving state-of-the-art video person sequence retrieval on: iLIDS-VID [1] by ∼13% and PRID-2011 by ∼7% in rank-1.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121137542","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.00059
Yunbo Rao, Wei Liu, J. Pu, Zheng Wang, Qifei Wang
In this paper, we focus on the problem of image feature extraction and similarity measure using region division search. Specifically, we proposed a novel image region division to roughly mimic the location distribution of image color and deal with the color histogram failing to describe spatial information. Furthermore, an image descriptor combining local color histogram and Gabor texture features with reduced feature dimensions are developed for optimizing our region division search method. Moreover, an extended Canberra distance is proposed for images similarity measure to increase the faulttolerant ability of the whole large-scale image search. Extensive experiments on several benchmark image retrieval databases validate the superiority of the proposed approaches.
{"title":"Large-Scale Image Search using Region Division","authors":"Yunbo Rao, Wei Liu, J. Pu, Zheng Wang, Qifei Wang","doi":"10.1109/ICDEW.2019.00059","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00059","url":null,"abstract":"In this paper, we focus on the problem of image feature extraction and similarity measure using region division search. Specifically, we proposed a novel image region division to roughly mimic the location distribution of image color and deal with the color histogram failing to describe spatial information. Furthermore, an image descriptor combining local color histogram and Gabor texture features with reduced feature dimensions are developed for optimizing our region division search method. Moreover, an extended Canberra distance is proposed for images similarity measure to increase the faulttolerant ability of the whole large-scale image search. Extensive experiments on several benchmark image retrieval databases validate the superiority of the proposed approaches.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122255932","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 : 2019-04-01DOI: 10.1109/ICDEW.2019.00010
Chunmiao Li, Yang Cao, Zhenjiang Hu, Masatoshi Yoshikawa
Early diagnosis and resection of colorectal polyps can effectively reduce the incidence and mortality rate. Colorectal cancer is a common gastrointestinal malignancy, ranking one of the three major malignancies around the world. With the improvement of living standards and dietary habits related problems, the incidence and mortality of colorectal cancer are showing an upward trend. Colorectal cancer is mostly from adenoma polyp malignant change, so early detection has important clinical significance. Although colonoscopy conducted by doctors is considered the most effective way in detecting polyps, uncertainty such as fatigue can affect the results. To solve this problem, we propose a fully convolutional densenet method to achieve the automatic detection and segmentation of colorectal polyps by computer. In this paper, we apply densenet to full convolutional network in segmentation of colorectal polyp, under the condition that not requiring post-processing and pre-training situation, we compare the number of parameters in different layers and assess accuracy and IOU respectively in segmentation of colorectal polyps. The results show that accuracy is improved as the layer increases gradually. When the layer number is 78(N=78), accuracy reaches 97.1% and the average IOU is 83.4%. In addition, we make a comparison with the state-of-the-art polyp segmentation method, the results reveal our method make a considerable improvement.
{"title":"Fully Convolutional DenseNets for Polyp Segmentation in Colonoscopy","authors":"Chunmiao Li, Yang Cao, Zhenjiang Hu, Masatoshi Yoshikawa","doi":"10.1109/ICDEW.2019.00010","DOIUrl":"https://doi.org/10.1109/ICDEW.2019.00010","url":null,"abstract":"Early diagnosis and resection of colorectal polyps can effectively reduce the incidence and mortality rate. Colorectal cancer is a common gastrointestinal malignancy, ranking one of the three major malignancies around the world. With the improvement of living standards and dietary habits related problems, the incidence and mortality of colorectal cancer are showing an upward trend. Colorectal cancer is mostly from adenoma polyp malignant change, so early detection has important clinical significance. Although colonoscopy conducted by doctors is considered the most effective way in detecting polyps, uncertainty such as fatigue can affect the results. To solve this problem, we propose a fully convolutional densenet method to achieve the automatic detection and segmentation of colorectal polyps by computer. In this paper, we apply densenet to full convolutional network in segmentation of colorectal polyp, under the condition that not requiring post-processing and pre-training situation, we compare the number of parameters in different layers and assess accuracy and IOU respectively in segmentation of colorectal polyps. The results show that accuracy is improved as the layer increases gradually. When the layer number is 78(N=78), accuracy reaches 97.1% and the average IOU is 83.4%. In addition, we make a comparison with the state-of-the-art polyp segmentation method, the results reveal our method make a considerable improvement.","PeriodicalId":186190,"journal":{"name":"2019 IEEE 35th International Conference on Data Engineering Workshops (ICDEW)","volume":"38 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116412356","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：