Pub Date : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120625
E. Hasan, R. Ibrahim, Kishore Bingi, S. Hassan, Syed Faizan-ul-Haq Gilani
Nonlinear behaviour of the systems happens to be a common problem in industrial processes. They cause a large amount of time, resources and efforts to be utilized in order to deal with them. A Major hurdle in Nonlinear Industrial Processes is system modeling. Due to this reason, several methods and techniques have been designed and developed in order to improve the overall control performance in industrial process control. Model based controllers have been developed and implemented on various applications with promising results. Their main benefit is they can identify and tune unknown system parameters in real-time. This paper focuses on real-time controller development and its implementation on Gas Pressure Process Plant using MPC. MPC is considered to be one of the robust and effective controllers due to impressive control performance in different applications previously. MPC makes use of a model for system identification and based upon that, it can dynamically send next control move for the system. This research work incorporates State-Space Model for unknown system-parameter identification. The identified parameters will be utilized by MPC for control law development. The proposed methodology is validated by real-time experimental results on the aforementioned system.
{"title":"Real-time model predictive control for nonlinear gas pressure process plant","authors":"E. Hasan, R. Ibrahim, Kishore Bingi, S. Hassan, Syed Faizan-ul-Haq Gilani","doi":"10.1109/ICSIPA.2017.8120625","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120625","url":null,"abstract":"Nonlinear behaviour of the systems happens to be a common problem in industrial processes. They cause a large amount of time, resources and efforts to be utilized in order to deal with them. A Major hurdle in Nonlinear Industrial Processes is system modeling. Due to this reason, several methods and techniques have been designed and developed in order to improve the overall control performance in industrial process control. Model based controllers have been developed and implemented on various applications with promising results. Their main benefit is they can identify and tune unknown system parameters in real-time. This paper focuses on real-time controller development and its implementation on Gas Pressure Process Plant using MPC. MPC is considered to be one of the robust and effective controllers due to impressive control performance in different applications previously. MPC makes use of a model for system identification and based upon that, it can dynamically send next control move for the system. This research work incorporates State-Space Model for unknown system-parameter identification. The identified parameters will be utilized by MPC for control law development. The proposed methodology is validated by real-time experimental results on the aforementioned system.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120951366","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120587
Afiqah Abu Samah, M. F. A. Fauzi, Sarina Mansor
Breast cancer leads the list of cancer that act on women worldwide. It starts when cells in the breast begin to build up beyond control. These cells normally create a tumour that can usually be seen on an x-ray or felt as a lump. Analysing and grading the tumour will take up much of a pathologist time. Pathologists have been largely diagnosing disease the same way for the past years, by manually reviewing images under a microscope. Thus, to help the pathologists improve accuracy and significantly change the way breast cancer been diagnosed, this paper presents an automated classification program. BreakHis dataset was used which build of 7909 breast tumor images gathered from 82 patients. This system is developed in order to categorize the cancer cells into two classes of cancer which are benign and malignant. The classification system compared different types of feature extractors using k-nearest neighbours classifier to efficiently observe the performance of the classification system. An extensive set of experiments showed that the overall accuracy rates range from 83% to 86%.
{"title":"Classification of benign and malignant tumors in histopathology images","authors":"Afiqah Abu Samah, M. F. A. Fauzi, Sarina Mansor","doi":"10.1109/ICSIPA.2017.8120587","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120587","url":null,"abstract":"Breast cancer leads the list of cancer that act on women worldwide. It starts when cells in the breast begin to build up beyond control. These cells normally create a tumour that can usually be seen on an x-ray or felt as a lump. Analysing and grading the tumour will take up much of a pathologist time. Pathologists have been largely diagnosing disease the same way for the past years, by manually reviewing images under a microscope. Thus, to help the pathologists improve accuracy and significantly change the way breast cancer been diagnosed, this paper presents an automated classification program. BreakHis dataset was used which build of 7909 breast tumor images gathered from 82 patients. This system is developed in order to categorize the cancer cells into two classes of cancer which are benign and malignant. The classification system compared different types of feature extractors using k-nearest neighbours classifier to efficiently observe the performance of the classification system. An extensive set of experiments showed that the overall accuracy rates range from 83% to 86%.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116585274","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120580
J. Shah, Hassaan Haider, K. Kadir, Sheroz Khan
Compressed Sensing is a novel sampling technique that can be used to faithfully recover sparse signals from fewer measurements than those proposed by the Nyquist theorem. A simple and intuitive measure of sparsity in a signal is ℓ0-norm. However, the ℓ0-norm function does not satisfy all the axiomatic properties of a true mathematical norm. The discrete and discontinuous nature of ℓ0-norm poses many challenges in its applications to recover sparse signals from their subsampled measurements. This paper presents, a novel mathematical function that can be used to closely approximate the ℓ0-norm. The proposed function is smooth and differentiable that allows gradient based algorithms to be used in the reconstruction of sparse signals. We use the proposed approximation along with steepest ascent method to develop a complete sparse signal recovery algorithm for the compressed sensing framework. Experimental results have shown that the proposed recovery algorithm outperforms the conventional SL0 method in terms of reconstruction accuracy such as Mean Square Error (MSE) and Signal-to-Noise Ratio (SNR).
{"title":"Sparse signal reconstruction of compressively sampled signals using smoothed ℓ0-norm","authors":"J. Shah, Hassaan Haider, K. Kadir, Sheroz Khan","doi":"10.1109/ICSIPA.2017.8120580","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120580","url":null,"abstract":"Compressed Sensing is a novel sampling technique that can be used to faithfully recover sparse signals from fewer measurements than those proposed by the Nyquist theorem. A simple and intuitive measure of sparsity in a signal is ℓ0-norm. However, the ℓ0-norm function does not satisfy all the axiomatic properties of a true mathematical norm. The discrete and discontinuous nature of ℓ0-norm poses many challenges in its applications to recover sparse signals from their subsampled measurements. This paper presents, a novel mathematical function that can be used to closely approximate the ℓ0-norm. The proposed function is smooth and differentiable that allows gradient based algorithms to be used in the reconstruction of sparse signals. We use the proposed approximation along with steepest ascent method to develop a complete sparse signal recovery algorithm for the compressed sensing framework. Experimental results have shown that the proposed recovery algorithm outperforms the conventional SL0 method in terms of reconstruction accuracy such as Mean Square Error (MSE) and Signal-to-Noise Ratio (SNR).","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116196817","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120600
Tanvira Ismail, L. J. Singh
Accurate dialect identification technique helps in improving the speech recognition systems that exist in most of the present day electronic devices and is also expected to help in providing new services in the field of e-health and telemedicine which is especially important for older and homebound people. The accuracy of a dialect identification system is highly dependent on its speech corpora. Therefore, in this paper, we describe how speech corpora have been developed for Goalparia dialect and languages it is similar to i.e. Assamese and Bengali. Finally, identification of Goalparia dialect, Assamese and Bengali languages have been done using the developed speech corpora in order to evaluate it.
{"title":"Development of speech corpora for Goalparia dialect and similar languages","authors":"Tanvira Ismail, L. J. Singh","doi":"10.1109/ICSIPA.2017.8120600","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120600","url":null,"abstract":"Accurate dialect identification technique helps in improving the speech recognition systems that exist in most of the present day electronic devices and is also expected to help in providing new services in the field of e-health and telemedicine which is especially important for older and homebound people. The accuracy of a dialect identification system is highly dependent on its speech corpora. Therefore, in this paper, we describe how speech corpora have been developed for Goalparia dialect and languages it is similar to i.e. Assamese and Bengali. Finally, identification of Goalparia dialect, Assamese and Bengali languages have been done using the developed speech corpora in order to evaluate it.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117080896","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120591
Umma Hany, L. Akter
In this paper, we propose distance and path loss based weighted centroid localization (WCL) algorithms for video capsule endoscope (VCE) using the received signal strength indicator (RSSI). We evaluate the performance of both algorithms considering real channel characteristics of human body. One of the major challenge in RSSI based VCE localization is the shadow fading and multi-path propagation effects of non-homogeneous medium of human body for which the measured RSSI is highly random resulting in high localization error. Again, due to the complex environment of experiment, accurate estimation of the channel parameters is quite difficult. We evaluate the performance of both algorithms in presence of randomness in path loss and estimation errors in channel parameters. To address the randomness issue, we estimate the smoothed path loss using moving averaging filter. Then, we introduce 10–50% errors in channel parameters to analyze the performance of both algorithms. We develop a simulation tool using MATLAB to visualize the results and to compare the performance. We observe significant improvement in performance by applying moving averaging method of smoothed path loss estimation using both algorithms. We also observe that the accuracy of distance based WCL decreases significantly in presence of errors in channel parameters. Whereas path loss based WCL is robust to the errors in channel parameters as it estimates the positions by using the estimated path loss directly without prior precise knowledge of channel parameters.
{"title":"Performance of distance based and path loss based weighted centroid localization algorithms for video capsule endoscope","authors":"Umma Hany, L. Akter","doi":"10.1109/ICSIPA.2017.8120591","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120591","url":null,"abstract":"In this paper, we propose distance and path loss based weighted centroid localization (WCL) algorithms for video capsule endoscope (VCE) using the received signal strength indicator (RSSI). We evaluate the performance of both algorithms considering real channel characteristics of human body. One of the major challenge in RSSI based VCE localization is the shadow fading and multi-path propagation effects of non-homogeneous medium of human body for which the measured RSSI is highly random resulting in high localization error. Again, due to the complex environment of experiment, accurate estimation of the channel parameters is quite difficult. We evaluate the performance of both algorithms in presence of randomness in path loss and estimation errors in channel parameters. To address the randomness issue, we estimate the smoothed path loss using moving averaging filter. Then, we introduce 10–50% errors in channel parameters to analyze the performance of both algorithms. We develop a simulation tool using MATLAB to visualize the results and to compare the performance. We observe significant improvement in performance by applying moving averaging method of smoothed path loss estimation using both algorithms. We also observe that the accuracy of distance based WCL decreases significantly in presence of errors in channel parameters. Whereas path loss based WCL is robust to the errors in channel parameters as it estimates the positions by using the estimated path loss directly without prior precise knowledge of channel parameters.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125287123","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120599
R. Kannan, Nur Izzati Abdul Samad, M. Romlie, N. M. Nor, L. Kumar
Electric vehicles and hybrid electric vehicles are seen as the future of the automotive industry with its aim to replace the conventional combustion engine vehicle. The conventional Multi-Input Multi-Output topology used in the electric and hybrid electric vehicle applications. The weaknesses of this topology are the complexity of circuit which increases the size of the converter and overall cost. In this research, the novel idea would be to implement the Single-Input Multi-Output DC-DC converter topology in an electric vehicle. The proposed idea will be able to overcome the downsides of the conventional method of the DC-DC converter used in electric vehicle and thus benefiting users. The limitations of this research would be the implementation of the system in a real electric vehicle. The circuit designed will be simulated, fabricated and evaluated.
{"title":"Design and execution of single input multiple output DC-DC converter","authors":"R. Kannan, Nur Izzati Abdul Samad, M. Romlie, N. M. Nor, L. Kumar","doi":"10.1109/ICSIPA.2017.8120599","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120599","url":null,"abstract":"Electric vehicles and hybrid electric vehicles are seen as the future of the automotive industry with its aim to replace the conventional combustion engine vehicle. The conventional Multi-Input Multi-Output topology used in the electric and hybrid electric vehicle applications. The weaknesses of this topology are the complexity of circuit which increases the size of the converter and overall cost. In this research, the novel idea would be to implement the Single-Input Multi-Output DC-DC converter topology in an electric vehicle. The proposed idea will be able to overcome the downsides of the conventional method of the DC-DC converter used in electric vehicle and thus benefiting users. The limitations of this research would be the implementation of the system in a real electric vehicle. The circuit designed will be simulated, fabricated and evaluated.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128505749","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120629
W. Ung, T. Tang, F. Mériaudeau, Esther Gunaseli M. Ebenezer
Cognitive rehabilitation has been proposed as an alternative treatment for Alzheimer's disease (AD) as it helps to preserve brain functionality. However, gains of cognitive training or rehabilitation may be eliminated due to cognitive overload and mental fatigue. This paper reports the development of a functional near-infrared spectroscopy (fNIRS) — brain-computer interface (BCI) that can adjust task difficulty adaptively. The aim is to have participants trained at their optimal level of difficulty and workload to maximize their gains. One patient with mild AD and one healthy control were recruited to test the functionality of proposed fNIRS-BCI system. The fNIRS-BCI system is able to process fNIRS signals in real time and adjust task difficulty accordingly. The healthy control was able to proceed to higher task levels, as compared to the mild AD patient. The fNIRS-BCI system has the potential as a tool to examine the efficacy of cognitive rehabilitation as an alternative treatment for AD.
{"title":"Dynamic optimization of mental workload in fNIRS-BCI system for cognitive rehabilitation","authors":"W. Ung, T. Tang, F. Mériaudeau, Esther Gunaseli M. Ebenezer","doi":"10.1109/ICSIPA.2017.8120629","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120629","url":null,"abstract":"Cognitive rehabilitation has been proposed as an alternative treatment for Alzheimer's disease (AD) as it helps to preserve brain functionality. However, gains of cognitive training or rehabilitation may be eliminated due to cognitive overload and mental fatigue. This paper reports the development of a functional near-infrared spectroscopy (fNIRS) — brain-computer interface (BCI) that can adjust task difficulty adaptively. The aim is to have participants trained at their optimal level of difficulty and workload to maximize their gains. One patient with mild AD and one healthy control were recruited to test the functionality of proposed fNIRS-BCI system. The fNIRS-BCI system is able to process fNIRS signals in real time and adjust task difficulty accordingly. The healthy control was able to proceed to higher task levels, as compared to the mild AD patient. The fNIRS-BCI system has the potential as a tool to examine the efficacy of cognitive rehabilitation as an alternative treatment for AD.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125806563","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120630
R. Zafar, A. Malik, Aliyu Nuhu Shuaibu, M. J. U. Rehman, S. Dass
In recent years convolutional neural network have obtained more popularity because of its progressive performance for different applications especially for object recognition. In neuroimaging, data varies from person to person and condition to condition so it is always a challenging job to model the brain data. Any analysis in neuroimaging is also dependent on the quality of data and currently, functional magnetic resonance imaging is considered as the best among all techniques. It is most reliable and popular modality to measure the brain activity patterns. In fMRI, region of interest is a common method of analysis in which data is taken from a specific brain region based on the structural or functional information. In this study, convolutional neural network is applied to the significant voxels obtained through the t-contrast of the design matrix during the ROI analysis. Data is taken against two conditions and 1000 significant voxels with highest absolute values are taken for each condition for further analysis. During the proposed method, analysis is performed using convolutional neural network along with ROI analysis. Support vector machine is used in the classification of both methods; ROI and proposed methods. In conclusion, it is shown that the features extracted through convolutional neural network can provide better significant results compared to the other one.
{"title":"Classification of fMRI data using support vector machine and convolutional neural network","authors":"R. Zafar, A. Malik, Aliyu Nuhu Shuaibu, M. J. U. Rehman, S. Dass","doi":"10.1109/ICSIPA.2017.8120630","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120630","url":null,"abstract":"In recent years convolutional neural network have obtained more popularity because of its progressive performance for different applications especially for object recognition. In neuroimaging, data varies from person to person and condition to condition so it is always a challenging job to model the brain data. Any analysis in neuroimaging is also dependent on the quality of data and currently, functional magnetic resonance imaging is considered as the best among all techniques. It is most reliable and popular modality to measure the brain activity patterns. In fMRI, region of interest is a common method of analysis in which data is taken from a specific brain region based on the structural or functional information. In this study, convolutional neural network is applied to the significant voxels obtained through the t-contrast of the design matrix during the ROI analysis. Data is taken against two conditions and 1000 significant voxels with highest absolute values are taken for each condition for further analysis. During the proposed method, analysis is performed using convolutional neural network along with ROI analysis. Support vector machine is used in the classification of both methods; ROI and proposed methods. In conclusion, it is shown that the features extracted through convolutional neural network can provide better significant results compared to the other one.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"106 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133136146","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120660
S. J. S. Gardezi, M. Awais, I. Faye, F. Mériaudeau
This paper presents a method for classification of normal and abnormal tissues in mammograms using a deep learning approach. VGG-16 CNN deep learning architecture with convolutional filter of (3×3) is implemented on mammograms ROIs from the IRMA dataset. The deep feature matrix is computed from first fully connected layer. The results are evaluated using 10 fold cross validation on SVM, binary trees, simple logistics and KNN (with k=1, 3, 5) classifiers. The method produced 100% classification accuracies with AUC 1.0.
{"title":"Mammogram classification using deep learning features","authors":"S. J. S. Gardezi, M. Awais, I. Faye, F. Mériaudeau","doi":"10.1109/ICSIPA.2017.8120660","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120660","url":null,"abstract":"This paper presents a method for classification of normal and abnormal tissues in mammograms using a deep learning approach. VGG-16 CNN deep learning architecture with convolutional filter of (3×3) is implemented on mammograms ROIs from the IRMA dataset. The deep feature matrix is computed from first fully connected layer. The results are evaluated using 10 fold cross validation on SVM, binary trees, simple logistics and KNN (with k=1, 3, 5) classifiers. The method produced 100% classification accuracies with AUC 1.0.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121210210","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 : 2017-09-01DOI: 10.1109/ICSIPA.2017.8120574
Rija Hasan, S. Monir
In this paper an efficient approach of fruit maturity classification based on apparent color of the specimen is implemented by the aid of fuzzy inference system (FIS). Heuristically acquired hue and its corresponding saturation and lightness are the attributes of choice, which are utilized to classify the sample into three classes; Raw, Ripe, and Overripe. The membership functions and fuzzy rules required by the Mamdani FIS are estimated by the approach of classification tree. The experimentation is performed upon 200 guava samples. The fuzzy system is trained upon 60% of the dataset, yielding 93.4% classification accuracy.
{"title":"Fruit maturity estimation based on fuzzy classification","authors":"Rija Hasan, S. Monir","doi":"10.1109/ICSIPA.2017.8120574","DOIUrl":"https://doi.org/10.1109/ICSIPA.2017.8120574","url":null,"abstract":"In this paper an efficient approach of fruit maturity classification based on apparent color of the specimen is implemented by the aid of fuzzy inference system (FIS). Heuristically acquired hue and its corresponding saturation and lightness are the attributes of choice, which are utilized to classify the sample into three classes; Raw, Ripe, and Overripe. The membership functions and fuzzy rules required by the Mamdani FIS are estimated by the approach of classification tree. The experimentation is performed upon 200 guava samples. The fuzzy system is trained upon 60% of the dataset, yielding 93.4% classification accuracy.","PeriodicalId":268112,"journal":{"name":"2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128488604","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：