Pub Date : 2011-03-04DOI: 10.1109/CSPA.2011.5759923
M. Rozali, I. Yassin, A. Zabidi, W. Mansor, N. Tahir
This paper describes an application of the Orthogonal Least Squares (OLS) algorithm for feature selection of spoken letters. Traditionally used for system identification purposes, the OLS method was used to select important Mel-Frequency Cepstrum Coefficients (MFCC) for classification of two spoken letters - ‘A’ and ‘S’ using Multi-Layer Perceptron (MLP) neural network. We evaluated several network structures and parameters to determine the best performance in terms of accuracy and speed. The result found that OLS is an effective feature selection method, with the best classification performance of 85% with 6 hidden units.
{"title":"Application of Orthogonal Least Square (OLS) for selection of Mel Frequency Cepstrum Coefficients for classification of spoken letters using MLP classifier","authors":"M. Rozali, I. Yassin, A. Zabidi, W. Mansor, N. Tahir","doi":"10.1109/CSPA.2011.5759923","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759923","url":null,"abstract":"This paper describes an application of the Orthogonal Least Squares (OLS) algorithm for feature selection of spoken letters. Traditionally used for system identification purposes, the OLS method was used to select important Mel-Frequency Cepstrum Coefficients (MFCC) for classification of two spoken letters - ‘A’ and ‘S’ using Multi-Layer Perceptron (MLP) neural network. We evaluated several network structures and parameters to determine the best performance in terms of accuracy and speed. The result found that OLS is an effective feature selection method, with the best classification performance of 85% with 6 hidden units.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125675836","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759914
P. Karthikeyan, M. Murugappan, S. Yaacob
Assessing human stress in real-time is more difficult and challenging today. The present review deals about the measurement of stress in laboratory environment using different stress inducement stimuli by the help of physiological signals. Previous researchers have been used different stress inducement stimuli such as stroop colour word test (CWT), mental arithmetic test, public speaking task, cold pressor test, computer games and works used to induce the stress. Most of the researchers have been analyzed stress using questionnaire based approach and physiological signals. The several physiological signals like Electrocardiogram (ECG), Electromyogram (EMG), Galvanic Skin Response (GSR), Blood Pressure (BP), Skin Temperature (ST), Blood Volume Pulse (BVP), respiration rate (RIP) and Electroencephalogram (EEG) were briefly investigated to identify the stress. Different statistical methods like Analysis of variance (ANOVA), two-way ANOVA, Multivariate analysis of variance (MANOVA), t-test, paired t-tests and student t-tests have used to describe the correlation between stress inducement stimuli, subjective parameters (age, gender and etc.,) and physiological signals. This present works aims to find the most appropriate stress inducement stimuli, physiological signals and statistical method to efficiently asses the human stress.
{"title":"A review on stress inducement stimuli for assessing human stress using physiological signals","authors":"P. Karthikeyan, M. Murugappan, S. Yaacob","doi":"10.1109/CSPA.2011.5759914","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759914","url":null,"abstract":"Assessing human stress in real-time is more difficult and challenging today. The present review deals about the measurement of stress in laboratory environment using different stress inducement stimuli by the help of physiological signals. Previous researchers have been used different stress inducement stimuli such as stroop colour word test (CWT), mental arithmetic test, public speaking task, cold pressor test, computer games and works used to induce the stress. Most of the researchers have been analyzed stress using questionnaire based approach and physiological signals. The several physiological signals like Electrocardiogram (ECG), Electromyogram (EMG), Galvanic Skin Response (GSR), Blood Pressure (BP), Skin Temperature (ST), Blood Volume Pulse (BVP), respiration rate (RIP) and Electroencephalogram (EEG) were briefly investigated to identify the stress. Different statistical methods like Analysis of variance (ANOVA), two-way ANOVA, Multivariate analysis of variance (MANOVA), t-test, paired t-tests and student t-tests have used to describe the correlation between stress inducement stimuli, subjective parameters (age, gender and etc.,) and physiological signals. This present works aims to find the most appropriate stress inducement stimuli, physiological signals and statistical method to efficiently asses the human stress.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130082340","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759846
S. H. Nor Farah, M. Y. Norfatimah, M. L. Siti Noor Hajjar
Mahseer being the most important freshwater fish that contributes to the major fisheries activity of the rivers in the world. In Malaysia, Mahseer is an important indigenous freshwater fish because of their market value. However, the number of this freshwater fish declining from day to day due to the human activity and environment destruction. In this study, mitochondrial DNA (mtDNA) which is cytochrome b gene was used to examine the genetic variation between four populations of Mahseer Tor spp. In total, forty-four individuals Mahseer were sampled from kelah sanctuary (n=22), Keniam River (n=1), Sat River (n=7) and Sepia River (n= 14) at Pahang National Park. Neighbor-joining as well as maximum-parsimony methods were used to analyze the phylogenetic relationships between Mahseer tor spp. The continuation of study based on the findings should be taken in order to start proper management as well as conservation of this valuable fish.
{"title":"Genetic variation of Mahseer Tor spp. in Pahang National Park using cytochrome b mtDNA gene based on phylogenetic analysis - a research framework","authors":"S. H. Nor Farah, M. Y. Norfatimah, M. L. Siti Noor Hajjar","doi":"10.1109/CSPA.2011.5759846","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759846","url":null,"abstract":"Mahseer being the most important freshwater fish that contributes to the major fisheries activity of the rivers in the world. In Malaysia, Mahseer is an important indigenous freshwater fish because of their market value. However, the number of this freshwater fish declining from day to day due to the human activity and environment destruction. In this study, mitochondrial DNA (mtDNA) which is cytochrome b gene was used to examine the genetic variation between four populations of Mahseer Tor spp. In total, forty-four individuals Mahseer were sampled from kelah sanctuary (n=22), Keniam River (n=1), Sat River (n=7) and Sepia River (n= 14) at Pahang National Park. Neighbor-joining as well as maximum-parsimony methods were used to analyze the phylogenetic relationships between Mahseer tor spp. The continuation of study based on the findings should be taken in order to start proper management as well as conservation of this valuable fish.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128225389","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759899
Norshafinaz Mohd Disa, I. Maarof, Z. Abd Latif, A. Samad
Light Detection and Ranging (LiDAR) is a technology that has been used for years with the variety of applications including the production of digital terrain models (DTMs), and high-accuracy mapping. LiDAR offers a very detailed collection of 3-D point clouds of the earth surface which can be used in generating orthophotos. Traditional orthophoto production based on the DTM has to accept that buildings and other objects above ground are not correctly placed in the orthophoto. These deficiencies can be overcome by simply taking the digital surface model (DSM) into account. In this paper, the literature reviews on the quality of true orthophoto production by fusing the digital aerial photos and LiDAR DSM done by former researchers are investigated. The results of the true orthophoto will be assessed by comparing the output and achievement gained by the researchers on the same field of study.
{"title":"LiDAR : A review on generating digital true orthophoto","authors":"Norshafinaz Mohd Disa, I. Maarof, Z. Abd Latif, A. Samad","doi":"10.1109/CSPA.2011.5759899","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759899","url":null,"abstract":"Light Detection and Ranging (LiDAR) is a technology that has been used for years with the variety of applications including the production of digital terrain models (DTMs), and high-accuracy mapping. LiDAR offers a very detailed collection of 3-D point clouds of the earth surface which can be used in generating orthophotos. Traditional orthophoto production based on the DTM has to accept that buildings and other objects above ground are not correctly placed in the orthophoto. These deficiencies can be overcome by simply taking the digital surface model (DSM) into account. In this paper, the literature reviews on the quality of true orthophoto production by fusing the digital aerial photos and LiDAR DSM done by former researchers are investigated. The results of the true orthophoto will be assessed by comparing the output and achievement gained by the researchers on the same field of study.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129136567","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759834
S. Shahidan, N. Nor, N. M. Bunnori
This paper is present a review on the evaluation of concrete structure damage by utilizing the moment tensor analysis (MTA) of acoustic emission (AE) source technique. In general moment tensor analysis is concerning the quantitative information on kinematics of cracks according to the AE source. This concept of AE analysis has been developed and mostly applied in reinforced concrete structure. Furthermore, the formulation of the evaluation of MTA is divided to a three different parts which are; namely kinematics crack, crack classification and crack volume. All these kinds of formulation have been established and proved by the previous researches. This paper also provides a brief overview of research work and several research papers on these topics were cited. Finally, this paper concluded with a discussion for future research area.
{"title":"Overview of moment tensor analysis of acoustic emission signal in evaluation concrete structure","authors":"S. Shahidan, N. Nor, N. M. Bunnori","doi":"10.1109/CSPA.2011.5759834","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759834","url":null,"abstract":"This paper is present a review on the evaluation of concrete structure damage by utilizing the moment tensor analysis (MTA) of acoustic emission (AE) source technique. In general moment tensor analysis is concerning the quantitative information on kinematics of cracks according to the AE source. This concept of AE analysis has been developed and mostly applied in reinforced concrete structure. Furthermore, the formulation of the evaluation of MTA is divided to a three different parts which are; namely kinematics crack, crack classification and crack volume. All these kinds of formulation have been established and proved by the previous researches. This paper also provides a brief overview of research work and several research papers on these topics were cited. Finally, this paper concluded with a discussion for future research area.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129330627","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759919
D. A. Awang Iskandar, R. Baini, A. Y. Wee, Shapiee Abdul Rahman, A. H. Fauzi
Pepper is a key export of the state of Sarawak (Malaysian Borneo); it produces 98% of Malaysia's pepper. At present, processed pepper berries are graded manually. This process is time consuming and error prone as it is very much dependent on the experience of the pepper grader. To overcome these weaknesses, we propose a Pepper Grading System which employs image processing and machine learning approaches using image features and moisture content data of the pepper berries. For instance, from initial tests, a high correlation between the grade of pepper berries to the colour features has been detected. Using existing machine learning algorithms in WEKA, we have obtained a 100% accuracy in categorising the pepper berries into the correct grades. In addition, moisture content and colourometer readings provide another 2 other parameters which may complement the image features in accurately classifying the berries into the right grades.
{"title":"iPepper: Intelligent pepper grading and quality assurance system","authors":"D. A. Awang Iskandar, R. Baini, A. Y. Wee, Shapiee Abdul Rahman, A. H. Fauzi","doi":"10.1109/CSPA.2011.5759919","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759919","url":null,"abstract":"Pepper is a key export of the state of Sarawak (Malaysian Borneo); it produces 98% of Malaysia's pepper. At present, processed pepper berries are graded manually. This process is time consuming and error prone as it is very much dependent on the experience of the pepper grader. To overcome these weaknesses, we propose a Pepper Grading System which employs image processing and machine learning approaches using image features and moisture content data of the pepper berries. For instance, from initial tests, a high correlation between the grade of pepper berries to the colour features has been detected. Using existing machine learning algorithms in WEKA, we have obtained a 100% accuracy in categorising the pepper berries into the correct grades. In addition, moisture content and colourometer readings provide another 2 other parameters which may complement the image features in accurately classifying the berries into the right grades.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132786898","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759841
Abdelwahab M. Bubtiena, Ahmed Elshafie, Othman Jafaar
Artificial Neural networks ANNs are dynamic systems which have the ability not only to capture the relationship between input and output parameters of complex systems but also highly effective when there is no any mathematical formula or model for the system. Therefore, they are very potential and appropriate for design of systems whose functions cannot be expressed explicitly in the form of mathematical model. If significant variables are known, without knowing the exact relationships, ANN is suitable to perform a kind of function fitting by using multiple parameters on the existing information and predict the possible relationships in the near future. This is the case in the water distribution network design or operation problems wherein the input (pipe diameters, lengths, age, soil, etc…)-output (reliability of the network) relationship is given by the set of nonlinear continuity equations, path head loss equations and the head-discharge relationship. This paper introduces a methodology of establishing ANN of modeling the pipe breaks from which rehabilitation strategies (proactive maintenance strategy), prioritization of rehabilitation implementation, finding the optimum time for rehabilitation of the pipe and determining the parameters that most affect the likelihood of pipe breaks, can be determined for predicting the number of breaks for each individual pipe in the water distribution system of Benghazi city (WDSB). Because this work is a part of a research has not completed yet, this paper presents only the modeling technique using ANN to achieve the main objective which is; expected number of pipe breaks.
{"title":"Application of Artificial Neural networks in modeling water networks","authors":"Abdelwahab M. Bubtiena, Ahmed Elshafie, Othman Jafaar","doi":"10.1109/CSPA.2011.5759841","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759841","url":null,"abstract":"Artificial Neural networks ANNs are dynamic systems which have the ability not only to capture the relationship between input and output parameters of complex systems but also highly effective when there is no any mathematical formula or model for the system. Therefore, they are very potential and appropriate for design of systems whose functions cannot be expressed explicitly in the form of mathematical model. If significant variables are known, without knowing the exact relationships, ANN is suitable to perform a kind of function fitting by using multiple parameters on the existing information and predict the possible relationships in the near future. This is the case in the water distribution network design or operation problems wherein the input (pipe diameters, lengths, age, soil, etc…)-output (reliability of the network) relationship is given by the set of nonlinear continuity equations, path head loss equations and the head-discharge relationship. This paper introduces a methodology of establishing ANN of modeling the pipe breaks from which rehabilitation strategies (proactive maintenance strategy), prioritization of rehabilitation implementation, finding the optimum time for rehabilitation of the pipe and determining the parameters that most affect the likelihood of pipe breaks, can be determined for predicting the number of breaks for each individual pipe in the water distribution system of Benghazi city (WDSB). Because this work is a part of a research has not completed yet, this paper presents only the modeling technique using ANN to achieve the main objective which is; expected number of pipe breaks.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124620328","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759870
N. M. Thamrin, M. A. Haron, F. Ruslan
Medical imaging has become essential in health industry. It becomes crucial during on-line or off-line data transmission between healthcare institutions. Larger image requires more spaces to be saved and more time to be loaded. Thus, in this paper, an artificial neural network is chosen to quantize the image into smaller number of colour palettes to reduce its size. A modified Kohonen Self-Organizing Maps algorithm is applied for hardware implementation. The Euclidean calculation in typical Kohonen algorithm is replaced with Manhattan Distance calculation to accelerate the computation in hardware implementation. In this research, the KSOM Processing Element CoProcessor hardware implementation, it consists of two main modules namely Datapath Unit (DPU) module and Control Unit (CU) module. The coprocessor is tested with one RGB colour input and three initial weights or desired palettes with three iterations. From the simulation testing, it took 480 nanoseconds to complete three iterations.
{"title":"A field programmable gate array implementation for biomedical system-on-chip (SoC)","authors":"N. M. Thamrin, M. A. Haron, F. Ruslan","doi":"10.1109/CSPA.2011.5759870","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759870","url":null,"abstract":"Medical imaging has become essential in health industry. It becomes crucial during on-line or off-line data transmission between healthcare institutions. Larger image requires more spaces to be saved and more time to be loaded. Thus, in this paper, an artificial neural network is chosen to quantize the image into smaller number of colour palettes to reduce its size. A modified Kohonen Self-Organizing Maps algorithm is applied for hardware implementation. The Euclidean calculation in typical Kohonen algorithm is replaced with Manhattan Distance calculation to accelerate the computation in hardware implementation. In this research, the KSOM Processing Element CoProcessor hardware implementation, it consists of two main modules namely Datapath Unit (DPU) module and Control Unit (CU) module. The coprocessor is tested with one RGB colour input and three initial weights or desired palettes with three iterations. From the simulation testing, it took 480 nanoseconds to complete three iterations.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121505271","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759905
J. Ghasemi, M. R. Karami mollaei
In this paper, a new and quick method is introduced for speech enhancement. The base or basic of this method is due to filtering the singular value which is obtained from SVD. The efficiency of the proposed methods is its strong capability and the speed of in reduction the noise effect and also does not have the typical “musical tone”, which is usually present in other noise reduction methods. The signals to be experimented are combined with an additive white Gaussian noise for a variety of Signal to Noise Ratio (SNR). These methods will be evaluated in SNR optimization and mean opinion score, which led to remarkable results.
{"title":"A new approach based on SVD for speech enhancement","authors":"J. Ghasemi, M. R. Karami mollaei","doi":"10.1109/CSPA.2011.5759905","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759905","url":null,"abstract":"In this paper, a new and quick method is introduced for speech enhancement. The base or basic of this method is due to filtering the singular value which is obtained from SVD. The efficiency of the proposed methods is its strong capability and the speed of in reduction the noise effect and also does not have the typical “musical tone”, which is usually present in other noise reduction methods. The signals to be experimented are combined with an additive white Gaussian noise for a variety of Signal to Noise Ratio (SNR). These methods will be evaluated in SNR optimization and mean opinion score, which led to remarkable results.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121892857","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 : 2011-03-04DOI: 10.1109/CSPA.2011.5759884
K. Chia, H. Abdul Rahim, R. Abdul Rahim
Generally, non-linear predictive models should be superior to linear predictive models. The objective of this study is to compare the performance of soluble solid content (SSC) prediction via Artificial Neural Network with Principal Components (PCs-ANN) and Principal Component Regression (PCR) in Visible and Shortwave Near Infrared (VIS-SWNIR) (400 – 1000 nm) spectrum. The spectra of 116 Fuji Apple samples were separated into calibration set of 84 apple samples and testing set of 32 apple samples randomly. Firstly, multiplicative scattering correction (MSC) was used to pre-process the spectra. Secondly, Principal Component Regression (PCR) was used to obtain the optimal number of principal components (PCs). Thirdly, the optimal PCs were used as the inputs of both multiple linear regression (MLR) and Artificial Neural Network (ANN) models. The results from this study showed that the predictive performance was improved significantly when PCs-ANN with two neurons was used compared to the PCR.
{"title":"A comparison of Principal Component Regression and Artificial Neural Network in fruits quality prediction","authors":"K. Chia, H. Abdul Rahim, R. Abdul Rahim","doi":"10.1109/CSPA.2011.5759884","DOIUrl":"https://doi.org/10.1109/CSPA.2011.5759884","url":null,"abstract":"Generally, non-linear predictive models should be superior to linear predictive models. The objective of this study is to compare the performance of soluble solid content (SSC) prediction via Artificial Neural Network with Principal Components (PCs-ANN) and Principal Component Regression (PCR) in Visible and Shortwave Near Infrared (VIS-SWNIR) (400 – 1000 nm) spectrum. The spectra of 116 Fuji Apple samples were separated into calibration set of 84 apple samples and testing set of 32 apple samples randomly. Firstly, multiplicative scattering correction (MSC) was used to pre-process the spectra. Secondly, Principal Component Regression (PCR) was used to obtain the optimal number of principal components (PCs). Thirdly, the optimal PCs were used as the inputs of both multiple linear regression (MLR) and Artificial Neural Network (ANN) models. The results from this study showed that the predictive performance was improved significantly when PCs-ANN with two neurons was used compared to the PCR.","PeriodicalId":282179,"journal":{"name":"2011 IEEE 7th International Colloquium on Signal Processing and its Applications","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122547760","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：