Pub Date : 2019-10-31DOI: 10.5121/sipij.2019.10501
B. Sadou, A. Lahoulou, T. Bouden, Anderson R. Avila, T. Falk, Z. Akhtar
This paper focuses on no-reference image quality assessment(NR-IQA)metrics. In the literature, a wide range of algorithms are proposed to automatically estimate the perceived quality of visual data. However, most of them are not able to effectively quantify the various degradations and artifacts that the image may undergo. Thus, merging of diverse metrics operating in different information domains is hoped to yield better performances, which is the main theme of the proposed work. In particular, the metric proposed in this paper is based on three well-known NR-IQA objective metrics that depend on natural scene statistical attributes from three different domains to extract a vector of image features. Then, Singular Value Decomposition (SVD) based dominant eigenvectors method is used to select the most relevant image quality attributes. These latter are used as input to Relevance Vector Machine (RVM) to derive the overall quality index. Validation experiments are divided into two groups; in the first group, learning process (training and test phases) is applied on one single image quality database whereas in the second group of simulations, training and test phases are separated on two distinct datasets. Obtained results demonstrate that the proposed metric performs very well in terms of correlation, monotonicity and accuracy in both the two scenarios.
{"title":"Free- Reference Image Quality Assessment Framework Using Metrics Fusion and Dimensionality Reduction","authors":"B. Sadou, A. Lahoulou, T. Bouden, Anderson R. Avila, T. Falk, Z. Akhtar","doi":"10.5121/sipij.2019.10501","DOIUrl":"https://doi.org/10.5121/sipij.2019.10501","url":null,"abstract":"This paper focuses on no-reference image quality assessment(NR-IQA)metrics. In the literature, a wide range of algorithms are proposed to automatically estimate the perceived quality of visual data. However, most of them are not able to effectively quantify the various degradations and artifacts that the image may undergo. Thus, merging of diverse metrics operating in different information domains is hoped to yield better performances, which is the main theme of the proposed work. In particular, the metric proposed in this paper is based on three well-known NR-IQA objective metrics that depend on natural scene statistical attributes from three different domains to extract a vector of image features. Then, Singular Value Decomposition (SVD) based dominant eigenvectors method is used to select the most relevant image quality attributes. These latter are used as input to Relevance Vector Machine (RVM) to derive the overall quality index. Validation experiments are divided into two groups; in the first group, learning process (training and test phases) is applied on one single image quality database whereas in the second group of simulations, training and test phases are separated on two distinct datasets. Obtained results demonstrate that the proposed metric performs very well in terms of correlation, monotonicity and accuracy in both the two scenarios.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"20 1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85678654","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 : 2019-10-31DOI: 10.5121/sipij.2019.10504
Itaru Kaneko, Y. Yoshida, E. Yuda
The use of Holter Electrocardiograph (Holter ECG) is rapidly spreading. It is a wearable electrocardiograph that records 24-hour electrocardiograms in a built-in flash memory, making it possible to detect atrial fibrillation (Atrial Fibrillation, AF) through all-day activities. It is also useful for screening for diseases other than atrial fibrillation and for improving health. It is said that more useful information can be obtained by combining electrocardiograph with the analysis of physical activity. For that purpose, the Holter electrocardiograph is equipped with heart rate sensor and acceleration sensors. If acceleration data is analysed, we can estimate activities in daily life, such as getting up, eating, walking, using transportation, and sitting. In combination with such activity status, electrocardiographic data can be expected to be more useful. In this study, we investigate the estimation of physical activity. For the better analysis, we evaluated activity estimation using machine learning as well as several different feature extractions. In this report, we will show several different feature extraction methods and result of human body analysis using machine learning.
{"title":"Improvements of the Analysis of Human Activity Using Acceleration Record of Electrocardiographs","authors":"Itaru Kaneko, Y. Yoshida, E. Yuda","doi":"10.5121/sipij.2019.10504","DOIUrl":"https://doi.org/10.5121/sipij.2019.10504","url":null,"abstract":"The use of Holter Electrocardiograph (Holter ECG) is rapidly spreading. It is a wearable electrocardiograph that records 24-hour electrocardiograms in a built-in flash memory, making it possible to detect atrial fibrillation (Atrial Fibrillation, AF) through all-day activities. It is also useful for screening for diseases other than atrial fibrillation and for improving health. It is said that more useful information can be obtained by combining electrocardiograph with the analysis of physical activity. For that purpose, the Holter electrocardiograph is equipped with heart rate sensor and acceleration sensors. If acceleration data is analysed, we can estimate activities in daily life, such as getting up, eating, walking, using transportation, and sitting. In combination with such activity status, electrocardiographic data can be expected to be more useful. In this study, we investigate the estimation of physical activity. For the better analysis, we evaluated activity estimation using machine learning as well as several different feature extractions. In this report, we will show several different feature extraction methods and result of human body analysis using machine learning.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"12 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89920634","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 : 2019-10-31DOI: 10.5121/sipij.2019.10502
Mahdi Naghibi, R. Anvari, A. Forghani, B. Minaei
It has been proven that deeper convolutional neural networks (CNN) can result in better accuracy in many problems, but this accuracy comes with a high computational cost. Also, input instances have not the same difficulty. As a solution for accuracy vs. computational cost dilemma, we introduce a new test-cost-sensitive method for convolutional neural networks. This method trains a CNN with a set of auxiliary outputs and expert branches in some middle layers of the network. The expert branches decide to use a shallower part of the network or going deeper to the end, based on the difficulty of input instance. The expert branches learn to determine: is the current network prediction is wrong and if the given instance passed to deeper layers of the network it will generate right output; If not, then the expert branches stop the computation process. The experimental results on standard dataset CIFAR-10 show that the proposed method can train models with lower test-cost and competitive accuracy in comparison with basic models.
{"title":"Test-cost-sensitive Convolutional Neural Networks with Expert Branches","authors":"Mahdi Naghibi, R. Anvari, A. Forghani, B. Minaei","doi":"10.5121/sipij.2019.10502","DOIUrl":"https://doi.org/10.5121/sipij.2019.10502","url":null,"abstract":"It has been proven that deeper convolutional neural networks (CNN) can result in better accuracy in many problems, but this accuracy comes with a high computational cost. Also, input instances have not the same difficulty. As a solution for accuracy vs. computational cost dilemma, we introduce a new test-cost-sensitive method for convolutional neural networks. This method trains a CNN with a set of auxiliary outputs and expert branches in some middle layers of the network. The expert branches decide to use a shallower part of the network or going deeper to the end, based on the difficulty of input instance. The expert branches learn to determine: is the current network prediction is wrong and if the given instance passed to deeper layers of the network it will generate right output; If not, then the expert branches stop the computation process. The experimental results on standard dataset CIFAR-10 show that the proposed method can train models with lower test-cost and competitive accuracy in comparison with basic models.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"72 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80584660","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 : 2019-10-31DOI: 10.5121/sipij.2019.10503
Omar Y. Adwan
In this paper a robust watermarking method operating in the wavelet domain for grayscale digital images is developed. The method first computes the differences between the watermark and the HH1 sub-band of the cover image values and then embed these differences in one of the frequency sub-bands. The results show that embedding the watermark in the LH1 sub-band gave the best results. The results were evaluated using the RMSE and the PSNR of both the original and the watermarked image. Although the watermark was recovered perfectly in the ideal case, the addition of Gaussian noise, or compression of the image using JPEG with quality less than 100 destroys the embedded watermark. Different experiments were carried out to test the performance of the proposed method and good results were obtained.
{"title":"Robust Image Watermarking Method using Wavelet Transform","authors":"Omar Y. Adwan","doi":"10.5121/sipij.2019.10503","DOIUrl":"https://doi.org/10.5121/sipij.2019.10503","url":null,"abstract":"In this paper a robust watermarking method operating in the wavelet domain for grayscale digital images is developed. The method first computes the differences between the watermark and the HH1 sub-band of the cover image values and then embed these differences in one of the frequency sub-bands. The results show that embedding the watermark in the LH1 sub-band gave the best results. The results were evaluated using the RMSE and the PSNR of both the original and the watermarked image. Although the watermark was recovered perfectly in the ideal case, the addition of Gaussian noise, or compression of the image using JPEG with quality less than 100 destroys the embedded watermark. Different experiments were carried out to test the performance of the proposed method and good results were obtained.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79483146","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 : 2019-08-31DOI: 10.5121/sipij.2019.10401
X. Mi, Xiaobing Wang, Xinyi He, F. Dai
The JONSWAP spectrum sea surface is mainly determined by parameters such as the wind speed, the fetch length and the peak enhancement factor. In view of the study of electromagnetic scattering from JONSWAP spectrum sea surface, we need to determine the above parameters. In this paper, we use the double summation model to generate the multi-directional irregular rough JONSWAP sea surface and analyze the distribution concentration parameter and the peak enhancement factor’s influence on the rough sea surface model, then using physical optics method to analysis the JONSWAP spectrum sea surface’s average backward scattering coefficient change with the different distribution concentration parameters and the peak enhancement factors, the simulation results show that the peak enhancement factor influence on the ocean surface of the average backward scattering coefficient is less than 1 dB, but the distribution concentration parameter influence on the JONSWAP surface of the average backward scattering coefficient is more than 5 dB. Therefore, when we study the electromagnetic scattering of the JONSWAP spectral sea surface, the peak enhancement factor can be taken as the mean value but the distribution concentration parameter have to be determined by the wave growth state.
{"title":"The Study on Electromagnetic Scattering Characteristics of Jonswap Spectrum Sea Surface","authors":"X. Mi, Xiaobing Wang, Xinyi He, F. Dai","doi":"10.5121/sipij.2019.10401","DOIUrl":"https://doi.org/10.5121/sipij.2019.10401","url":null,"abstract":"The JONSWAP spectrum sea surface is mainly determined by parameters such as the wind speed, the fetch length and the peak enhancement factor. In view of the study of electromagnetic scattering from JONSWAP spectrum sea surface, we need to determine the above parameters. In this paper, we use the double summation model to generate the multi-directional irregular rough JONSWAP sea surface and analyze the distribution concentration parameter and the peak enhancement factor’s influence on the rough sea surface model, then using physical optics method to analysis the JONSWAP spectrum sea surface’s average backward scattering coefficient change with the different distribution concentration parameters and the peak enhancement factors, the simulation results show that the peak enhancement factor influence on the ocean surface of the average backward scattering coefficient is less than 1 dB, but the distribution concentration parameter influence on the JONSWAP surface of the average backward scattering coefficient is more than 5 dB. Therefore, when we study the electromagnetic scattering of the JONSWAP spectral sea surface, the peak enhancement factor can be taken as the mean value but the distribution concentration parameter have to be determined by the wave growth state.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87143935","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 : 2019-08-31DOI: 10.5121/sipij.2019.10402
Nidhal Azawi, J. Gauch
Colonoscopy is a procedure that has been used widely to detect the abnormality in a colon. Colonoscopy images suffer from a lot of problems that make it hard for the doctor to investigate/ understand a colon patient. Unfortunately, with the current technology, three is no way for doctors to know if the whole colon surface has been investigated or not. We have developed a method that utilizes RANSAC-based image registration to align sequences of any length in the colonoscopy video and restores each frame of the video using information from these aligned images. We proposed two methods. First method used the deep neural net for the classification of informative and non-informative image. The classification result was used as a preprocessing for alignment method. Also, we proposed a visualization structure for the classification results. The second method used the alignment to decide/classify the bad and good alignment by using two factors. The first factor is the accumulated error and the second factor contain three checking steps that check the pair error alignment beside the geometry transform status. The second method was able to align long sequences.
{"title":"Ransac Based Motion Compensated Restoration for Colonoscopy Images","authors":"Nidhal Azawi, J. Gauch","doi":"10.5121/sipij.2019.10402","DOIUrl":"https://doi.org/10.5121/sipij.2019.10402","url":null,"abstract":"Colonoscopy is a procedure that has been used widely to detect the abnormality in a colon. Colonoscopy images suffer from a lot of problems that make it hard for the doctor to investigate/ understand a colon patient. Unfortunately, with the current technology, three is no way for doctors to know if the whole colon surface has been investigated or not. We have developed a method that utilizes RANSAC-based image registration to align sequences of any length in the colonoscopy video and restores each frame of the video using information from these aligned images. We proposed two methods. First method used the deep neural net for the classification of informative and non-informative image. The classification result was used as a preprocessing for alignment method. Also, we proposed a visualization structure for the classification results. The second method used the alignment to decide/classify the bad and good alignment by using two factors. The first factor is the accumulated error and the second factor contain three checking steps that check the pair error alignment beside the geometry transform status. The second method was able to align long sequences.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"154 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74498956","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 : 2019-06-30DOI: 10.5121/SIPIJ.2019.10301
Y. Yoshida, E. Yuda, Kento Yamamoto, Yutaka Miura, J. Hayano
Machine-Learning Estimation of Body Posture and Physical Activity by Wearable Acceleration and Heartbeat Sensors
基于可穿戴加速和心跳传感器的身体姿势和身体活动的机器学习估计
{"title":"Machine-Learning Estimation of Body Posture and Physical Activity by Wearable Acceleration and Heartbeat Sensors","authors":"Y. Yoshida, E. Yuda, Kento Yamamoto, Yutaka Miura, J. Hayano","doi":"10.5121/SIPIJ.2019.10301","DOIUrl":"https://doi.org/10.5121/SIPIJ.2019.10301","url":null,"abstract":"Machine-Learning Estimation of Body Posture and Physical Activity by Wearable Acceleration and Heartbeat Sensors","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"25 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77454567","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 : 2019-06-29DOI: 10.5121/SIPIJ.2019.10305
Tatiani Pivem, Felipe de Oliveira de Araujo, Laura de Oliveira de Araujo, Gustavo Spontoni de Oliveira
Application of A Computer Vision Method for Soiling Recognition in Photovoltaic Modules for Autonomous Cleaning Robots
计算机视觉方法在自主清洁机器人光伏组件污垢识别中的应用
{"title":"Application of A Computer Vision Method for Soiling Recognition in Photovoltaic Modules for Autonomous Cleaning Robots","authors":"Tatiani Pivem, Felipe de Oliveira de Araujo, Laura de Oliveira de Araujo, Gustavo Spontoni de Oliveira","doi":"10.5121/SIPIJ.2019.10305","DOIUrl":"https://doi.org/10.5121/SIPIJ.2019.10305","url":null,"abstract":"Application of A Computer Vision Method for Soiling Recognition in Photovoltaic Modules for Autonomous Cleaning Robots","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"49 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85081465","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 : 2019-06-29DOI: 10.5121/SIPIJ.2019.10302
V. Semenov, P. Omelchenko, O. Kruhlyk
In this paper we propose the method for the detection of Carrier-in-Carrier signals using QPSK modulations. The method is based on the calculation of fourth-order cumulants. In accordance with the methodology based on the Receiver Operating Characteristic (ROC) curve, a threshold value for the decision rule is established. It was found that the proposed method provides the correct detection of the sum of QPSK signals for a wide range of signal-to-noise ratios and also for the different bandwidths of mixed signals. The obtained results indicate the high efficiency of the proposed detection method. The advantage of the proposed detection method over the “radiuses” method is also shown.
{"title":"Method for the Detection of Mixed QPSK Signals Based on the Calculation of Fourth-Order Cumulants","authors":"V. Semenov, P. Omelchenko, O. Kruhlyk","doi":"10.5121/SIPIJ.2019.10302","DOIUrl":"https://doi.org/10.5121/SIPIJ.2019.10302","url":null,"abstract":"In this paper we propose the method for the detection of Carrier-in-Carrier signals using QPSK modulations. The method is based on the calculation of fourth-order cumulants. In accordance with the methodology based on the Receiver Operating Characteristic (ROC) curve, a threshold value for the decision rule is established. It was found that the proposed method provides the correct detection of the sum of QPSK signals for a wide range of signal-to-noise ratios and also for the different bandwidths of mixed signals. The obtained results indicate the high efficiency of the proposed detection method. The advantage of the proposed detection method over the “radiuses” method is also shown.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"15 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72979997","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 : 2019-06-29DOI: 10.5121/SIPIJ.2019.10304
S. Ibrahem, Y. M. A. El-Latif, Naglaa M. Reda
Automatic leaf recognition via image processing has been greatly important for a number of professionals, such as botanical taxonomic, environmental protectors, and foresters. Learn an over-complete leaf dictionary is an essential step for leaf image recognition. Big leaf images dimensions and training images number is facing of fast and complete data leaves dictionary. In this work an efficient approach applies to construct over-complete data leaves dictionary to set of big images diminutions based on sparse representation. In the proposed method a new cropped-contour method has used to crop the training image. The experiments are testing using correlation between the sparse representation and data dictionary and with focus on the computing time.
{"title":"A Novel Data Dictionary Learning for Leaf Recognition","authors":"S. Ibrahem, Y. M. A. El-Latif, Naglaa M. Reda","doi":"10.5121/SIPIJ.2019.10304","DOIUrl":"https://doi.org/10.5121/SIPIJ.2019.10304","url":null,"abstract":"Automatic leaf recognition via image processing has been greatly important for a number of professionals, such as botanical taxonomic, environmental protectors, and foresters. Learn an over-complete leaf dictionary is an essential step for leaf image recognition. Big leaf images dimensions and training images number is facing of fast and complete data leaves dictionary. In this work an efficient approach applies to construct over-complete data leaves dictionary to set of big images diminutions based on sparse representation. In the proposed method a new cropped-contour method has used to crop the training image. The experiments are testing using correlation between the sparse representation and data dictionary and with focus on the computing time.","PeriodicalId":90726,"journal":{"name":"Signal and image processing : an international journal","volume":"82 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83968523","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：