Pub Date : 2018-01-01DOI: 10.1109/CISP-BMEI.2018.8633088
Ziyan Shi, Guolin Zhao, Qiaolin Hu
{"title":"The Research of Electronic Countermeasure Intelligence Correlation Analysis Based on Machine Learning","authors":"Ziyan Shi, Guolin Zhao, Qiaolin Hu","doi":"10.1109/CISP-BMEI.2018.8633088","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2018.8633088","url":null,"abstract":"","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"16 1","pages":"1-9"},"PeriodicalIF":0.0,"publicationDate":"2018-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76565125","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 : 2017-10-16DOI: 10.1109/CISP-BMEI.2017.8302066
Rong Dong, Juan Liao, Bo Li, Huiyu Zhou, D. Crookes
Measuring rainfall rates from videos is a novel research topic. Due to rain motion, reflection of light and background clutter, it is extremely challenging to obtain accurate measurements. In this paper, we propose a new technique for measuring rainfall rates from videos, which consists of the following technical steps: first, we detect raindrops in an image using gray-tone functions and direction of rain streaks; we then select the focused raindrops, based on two features: average color tensor response and average intensity difference. Afterwards, the size of the raindrops is estimated and a raindrop size distribution (RSD) curve is created according to the use of the RSD in meteorology. Finally, a rainfall rate is obtained by fitting the RSD curve with a Gamma distribution model. In the experiment section presented in this paper, the proposed algorithm is evaluated under different light, moderate and heavy rainy conditions. The measurement results of the proposed algorithm are consistent with those of a can-type rain gauge.
{"title":"Measurements of rainfall rates from videos","authors":"Rong Dong, Juan Liao, Bo Li, Huiyu Zhou, D. Crookes","doi":"10.1109/CISP-BMEI.2017.8302066","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302066","url":null,"abstract":"Measuring rainfall rates from videos is a novel research topic. Due to rain motion, reflection of light and background clutter, it is extremely challenging to obtain accurate measurements. In this paper, we propose a new technique for measuring rainfall rates from videos, which consists of the following technical steps: first, we detect raindrops in an image using gray-tone functions and direction of rain streaks; we then select the focused raindrops, based on two features: average color tensor response and average intensity difference. Afterwards, the size of the raindrops is estimated and a raindrop size distribution (RSD) curve is created according to the use of the RSD in meteorology. Finally, a rainfall rate is obtained by fitting the RSD curve with a Gamma distribution model. In the experiment section presented in this paper, the proposed algorithm is evaluated under different light, moderate and heavy rainy conditions. The measurement results of the proposed algorithm are consistent with those of a can-type rain gauge.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"40 1","pages":"1-9"},"PeriodicalIF":0.0,"publicationDate":"2017-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79109502","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8302322
F. Jabeen, L. Tao, Xinyue Wang
Mind Interactive Multimedia System (MIMS) is designed to allow people with severe disabilities (amyotrophic lateral sclerosis (ALS), traumatic brain injuries (TBI) and spinal cord injury (SCI)) to use computers with ease and accuracy through EEG signals. With primary focus towards special needs and user experience (UX) of patients, an interaction theory is presented, which makes a meaningful contribution in the area of Brain-Computer-Interaction (BCI). A BCI system (working with single cognitive signal) based on Emotiv EPOC headset (a commercial-type, 14-channel electroencephalography (EEG) signals reader) is also introduced, which meets the needs in communication, actual caregiving and recreation for patients. Evaluation results indicate that proposed MIMS (software and hardware) can provide convenient communication for ALS patients.
{"title":"Mind interactive multimedia system for disabled people","authors":"F. Jabeen, L. Tao, Xinyue Wang","doi":"10.1109/CISP-BMEI.2017.8302322","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302322","url":null,"abstract":"Mind Interactive Multimedia System (MIMS) is designed to allow people with severe disabilities (amyotrophic lateral sclerosis (ALS), traumatic brain injuries (TBI) and spinal cord injury (SCI)) to use computers with ease and accuracy through EEG signals. With primary focus towards special needs and user experience (UX) of patients, an interaction theory is presented, which makes a meaningful contribution in the area of Brain-Computer-Interaction (BCI). A BCI system (working with single cognitive signal) based on Emotiv EPOC headset (a commercial-type, 14-channel electroencephalography (EEG) signals reader) is also introduced, which meets the needs in communication, actual caregiving and recreation for patients. Evaluation results indicate that proposed MIMS (software and hardware) can provide convenient communication for ALS patients.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"19 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73459951","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 paper we present a study on historical documents denoising methods and make visual quality performance comparison through Deep Residual Learning, Alternating Direction Method of Multiplier and Anisotropic diffusion PDE. Experimental results demonstrate their denoising visual quality performance and we make a comparison to in different condition respectively.
{"title":"A study of historical documents denoising","authors":"Guoming Chen, Qiang Chen, Xiongyong Zhu, Yiqun Chen","doi":"10.1109/CISP-BMEI.2017.8301947","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8301947","url":null,"abstract":"In this paper we present a study on historical documents denoising methods and make visual quality performance comparison through Deep Residual Learning, Alternating Direction Method of Multiplier and Anisotropic diffusion PDE. Experimental results demonstrate their denoising visual quality performance and we make a comparison to in different condition respectively.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"107 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74641360","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8301983
Alireza Mojtabavi, P. Farnia, A. Ahmadian, M. Alimohamadi, Ahmad Pourrashidi, H. S. Rad, J. Alirezaie
Accurate segmentation and characterization of abnormalities in brain tumor are challenging task, especially in the case of GBM tumors, where the ambiguities presented in the boundaries of these tumors necessitates using efficient segmentation method. Level set methods have proven to be a flexible and powerful tool for image segmentation because of being shape-driven method with a properly defined speed function to grow or shrink the boundaries to segment complex objects of interest, precisely. In this study a combined level set algorithm consists of both region and boundary terms for GBM segmentation is proposed. The modified speed function incorporates threshold based level set and the Laplacian filter to highlight the fine details for performing an accurate extraction of the tumor region using multiple seed points selected by the user. An evaluation was performed on a dataset containing 6 patients with GBM by using three measures Dice, false positive error (FPE) and false negative error (FNE). Manual segmentation of GBM is considered as gold standard. Compared to traditional method, the mean of FPE and FNE are improved by 53.5% and 53.1%, respectively. The mean of Dice coefficients between our results and gold standard measurement reached to 0.88. As the results proved, the proposed combined method improves the accuracy of GBM segmentation by 16% compared to conventional level set method with threshold based speed function. Our method is also robust to change of parameters.
{"title":"Segmentation of GBM in MRI images using an efficient speed function based on level set method","authors":"Alireza Mojtabavi, P. Farnia, A. Ahmadian, M. Alimohamadi, Ahmad Pourrashidi, H. S. Rad, J. Alirezaie","doi":"10.1109/CISP-BMEI.2017.8301983","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8301983","url":null,"abstract":"Accurate segmentation and characterization of abnormalities in brain tumor are challenging task, especially in the case of GBM tumors, where the ambiguities presented in the boundaries of these tumors necessitates using efficient segmentation method. Level set methods have proven to be a flexible and powerful tool for image segmentation because of being shape-driven method with a properly defined speed function to grow or shrink the boundaries to segment complex objects of interest, precisely. In this study a combined level set algorithm consists of both region and boundary terms for GBM segmentation is proposed. The modified speed function incorporates threshold based level set and the Laplacian filter to highlight the fine details for performing an accurate extraction of the tumor region using multiple seed points selected by the user. An evaluation was performed on a dataset containing 6 patients with GBM by using three measures Dice, false positive error (FPE) and false negative error (FNE). Manual segmentation of GBM is considered as gold standard. Compared to traditional method, the mean of FPE and FNE are improved by 53.5% and 53.1%, respectively. The mean of Dice coefficients between our results and gold standard measurement reached to 0.88. As the results proved, the proposed combined method improves the accuracy of GBM segmentation by 16% compared to conventional level set method with threshold based speed function. Our method is also robust to change of parameters.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"96 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74866332","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8302078
Xiaofang Xu, Joao Amaro, Sam Caulfield, A. Forembski, G. Falcão, D. Moloney
2D Convolutional Neural Networks (CNNs) have enjoyed a surge in popularity over the last few years, mainly because they outperform traditional algorithms/methods in a myriad of computer vision (and other fields) tasks. On the other hand, the problem becomes more complex when dealing with 3D volumes. Lack of readily available training data, memory and computational requirements are just some of the factors hindering the progress of 3D CNNs. We propose a synthetic 3D voxelized point-clouds generation method containing object and scene in this paper. Furthermore, an efficient 3D volumetric representation called VOLA is applied. VOLA (Volumetric Accelerator) is a sexaquaternary (power-of-four subdivision) tree-based representation which aims to save significant memory for volumetric data. After training the model, it was deployed onto Movidius Neural Compute Stick which is a USB, containing a low-power processing unit as well as dedicated CNN hardware blocks. The trained model on NCS takes only ∼ 90 frames per second to perform inference on each 3D volume, with an average power consumption of 1.2W.
{"title":"Convolutional neural network on neural compute stick for voxelized point-clouds classification","authors":"Xiaofang Xu, Joao Amaro, Sam Caulfield, A. Forembski, G. Falcão, D. Moloney","doi":"10.1109/CISP-BMEI.2017.8302078","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302078","url":null,"abstract":"2D Convolutional Neural Networks (CNNs) have enjoyed a surge in popularity over the last few years, mainly because they outperform traditional algorithms/methods in a myriad of computer vision (and other fields) tasks. On the other hand, the problem becomes more complex when dealing with 3D volumes. Lack of readily available training data, memory and computational requirements are just some of the factors hindering the progress of 3D CNNs. We propose a synthetic 3D voxelized point-clouds generation method containing object and scene in this paper. Furthermore, an efficient 3D volumetric representation called VOLA is applied. VOLA (Volumetric Accelerator) is a sexaquaternary (power-of-four subdivision) tree-based representation which aims to save significant memory for volumetric data. After training the model, it was deployed onto Movidius Neural Compute Stick which is a USB, containing a low-power processing unit as well as dedicated CNN hardware blocks. The trained model on NCS takes only ∼ 90 frames per second to perform inference on each 3D volume, with an average power consumption of 1.2W.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"895 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75457016","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8301909
Zhenghua Shu, Guodong Liu, Zhihua Xie, Z. Ren
In radar tracking guidance, intelligent video surveillance, robot vision system, the parameters of position and velocity and steering state often need to get the target of interest, based on the motion characteristics of the target and further to control it. The filtering method is used to estimate the desired state parameters based on the functional relationship between the measured values and the state variables. This method is also called target tracking technique. At present, there are many target tracking technologies for different systems, but there is a big gap between the robustness and real-time requirements of the actual system. In order to solve the problem of large computation and bad real-time performance of Particle Filters, a real-time target tracking algorithm based on nonlinear mean shift and Particle Filters is proposed. The distribution of particles is closer to the actual posterior distribution by selecting the important probability density function. Furthermore, the nonlinear mean shift algorithm is integrated into the Particle Filters, so that the particles are further clustered into the real distribution. Finally, the algorithm is applied in the traffic video surveillance, and the effective tracking of the target motorcycle and vehicle is realized.
{"title":"Real time target tracking based on nonlinear mean shift and particle filters","authors":"Zhenghua Shu, Guodong Liu, Zhihua Xie, Z. Ren","doi":"10.1109/CISP-BMEI.2017.8301909","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8301909","url":null,"abstract":"In radar tracking guidance, intelligent video surveillance, robot vision system, the parameters of position and velocity and steering state often need to get the target of interest, based on the motion characteristics of the target and further to control it. The filtering method is used to estimate the desired state parameters based on the functional relationship between the measured values and the state variables. This method is also called target tracking technique. At present, there are many target tracking technologies for different systems, but there is a big gap between the robustness and real-time requirements of the actual system. In order to solve the problem of large computation and bad real-time performance of Particle Filters, a real-time target tracking algorithm based on nonlinear mean shift and Particle Filters is proposed. The distribution of particles is closer to the actual posterior distribution by selecting the important probability density function. Furthermore, the nonlinear mean shift algorithm is integrated into the Particle Filters, so that the particles are further clustered into the real distribution. Finally, the algorithm is applied in the traffic video surveillance, and the effective tracking of the target motorcycle and vehicle is realized.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"8 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75563104","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8302028
S. Liang, Xuehui Hou, X. Sui, M. Wang
The fraction of absorbed photosynthetically active radiation (FPAR) is directly related to the primary productivity of photosynthesis, and is widely used to estimate ecosystem primary production. The forest canopy can be divided into photosynthetically active vegetation and non-photosynthetic vegetation according to their photosynthetical function. In this study, the scattering by arbitrary inclined leaves (SAIL) model was used to partition the PAR absorbed by canopy components into two parts: PAR absorbed by PAV and PAR by NPV. The characteristics of green FPAR (the fraction of PAR absorbed by PAV) and the relationships between green FPAR and spectral vegetation indices (NDVI, EVI, EVI2, SAVI) were analyzed. The results showed that the green FPAR varied with the canopy structure. In broadleaf deciduous forests with high coverage, the green FPAR was close to the total FPAR, and the contribution of NPV to the total FPAR was very low. Plant area index had more important impacts on the green FPAR than the proportion of PAV in the canopy and optical properties of PAV. The green FPAR had significant relationships with four spectral vegetation indices, but the correlation coefficient between green FPAR and EVI was the largest. Therefore, compared with other three vegetation indices, EVI may be more suitable to estimate forest green FPAR.
{"title":"Deciduous broadleaf forests green FPAR and its relationship with spectral vegetation indices","authors":"S. Liang, Xuehui Hou, X. Sui, M. Wang","doi":"10.1109/CISP-BMEI.2017.8302028","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302028","url":null,"abstract":"The fraction of absorbed photosynthetically active radiation (FPAR) is directly related to the primary productivity of photosynthesis, and is widely used to estimate ecosystem primary production. The forest canopy can be divided into photosynthetically active vegetation and non-photosynthetic vegetation according to their photosynthetical function. In this study, the scattering by arbitrary inclined leaves (SAIL) model was used to partition the PAR absorbed by canopy components into two parts: PAR absorbed by PAV and PAR by NPV. The characteristics of green FPAR (the fraction of PAR absorbed by PAV) and the relationships between green FPAR and spectral vegetation indices (NDVI, EVI, EVI2, SAVI) were analyzed. The results showed that the green FPAR varied with the canopy structure. In broadleaf deciduous forests with high coverage, the green FPAR was close to the total FPAR, and the contribution of NPV to the total FPAR was very low. Plant area index had more important impacts on the green FPAR than the proportion of PAV in the canopy and optical properties of PAV. The green FPAR had significant relationships with four spectral vegetation indices, but the correlation coefficient between green FPAR and EVI was the largest. Therefore, compared with other three vegetation indices, EVI may be more suitable to estimate forest green FPAR.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"30 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74041626","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8302018
D. Mitrea, S. Nedevschi, P. Mitrea, M. Platon-Lupsor, R. Badea
Cirrhosis is an important disease, as it can precede liver cancer, and also can lead to death by itself. Detecting the severity grades of cirrhosis is a major issue in this context. The best nowadays standard for this purpose is the biopsy, however this procedure is invasive, dangerous for the patient. Also, there is no objective study in order to establish which the cirrhosis grades are. Our research purpose is to discover the cirrhosis grades using computerized methods and to perform non-invasive, computer assisted and automatic diagnosis of the disease evolution phases. Concerning the employed features, we adopted the texture-based methods, able to emphasize those characteristics of the tissue that cannot be detected by the eye of the medical expert. In this paper, we emphasized the role of the CETMCM Matrix concerning the detection of the cirrhosis severity grades. The method was validated by supervised classification, providing a recognition rate above 95%.
{"title":"The role of the cooccurrence matrix based on complex extended microstructures in discovering the cirrhosis severity grades within US images","authors":"D. Mitrea, S. Nedevschi, P. Mitrea, M. Platon-Lupsor, R. Badea","doi":"10.1109/CISP-BMEI.2017.8302018","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302018","url":null,"abstract":"Cirrhosis is an important disease, as it can precede liver cancer, and also can lead to death by itself. Detecting the severity grades of cirrhosis is a major issue in this context. The best nowadays standard for this purpose is the biopsy, however this procedure is invasive, dangerous for the patient. Also, there is no objective study in order to establish which the cirrhosis grades are. Our research purpose is to discover the cirrhosis grades using computerized methods and to perform non-invasive, computer assisted and automatic diagnosis of the disease evolution phases. Concerning the employed features, we adopted the texture-based methods, able to emphasize those characteristics of the tissue that cannot be detected by the eye of the medical expert. In this paper, we emphasized the role of the CETMCM Matrix concerning the detection of the cirrhosis severity grades. The method was validated by supervised classification, providing a recognition rate above 95%.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"17 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74771714","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 : 2017-10-01DOI: 10.1109/CISP-BMEI.2017.8302045
Shan Ding, Li Xin
Diabetic Retinopathy (DR) is a serious diabetic complication which may lead to new-onset blindness or visual injury. As the smallest lesions and the earliest sign that can be observed, the screening and localization of MAs is especially important for the diabetes diagnose of early lesions. In this paper, a combination of algorithms is proposed to detect MAs accurately. In the proposed algorithm, a primary candidate set will be detected by using the convolution dynamic multiparameter template (CDMPT) matching scheme and then uses a Random Forest to obtain true MA classification from the candidate set. In this work, the proposed algorithm is tested on a public dataset. The experimental results validate the effectiveness of the new algorithm.
{"title":"An improved algorithm based on convolution dynamic multi-parameter template for microaneurysms detection","authors":"Shan Ding, Li Xin","doi":"10.1109/CISP-BMEI.2017.8302045","DOIUrl":"https://doi.org/10.1109/CISP-BMEI.2017.8302045","url":null,"abstract":"Diabetic Retinopathy (DR) is a serious diabetic complication which may lead to new-onset blindness or visual injury. As the smallest lesions and the earliest sign that can be observed, the screening and localization of MAs is especially important for the diabetes diagnose of early lesions. In this paper, a combination of algorithms is proposed to detect MAs accurately. In the proposed algorithm, a primary candidate set will be detected by using the convolution dynamic multiparameter template (CDMPT) matching scheme and then uses a Random Forest to obtain true MA classification from the candidate set. In this work, the proposed algorithm is tested on a public dataset. The experimental results validate the effectiveness of the new algorithm.","PeriodicalId":6474,"journal":{"name":"2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"194 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75138348","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：