Support Vector Machine Classification Approach Research Articles

Support Vector Machine Algorithm for Mapping Land Cover Dynamics in Senegal, West Africa, Using Earth Observation Data

This paper addresses the problem of mapping land cover types in Senegal and recognition of vegetation systems in the Saloum River Delta on the satellite images. Multi-seasonal landscape dynamics were analyzed using Landsat 8-9 OLI/TIRS images from 2015 to 2023. Two image classification methods were compared, and their performance was evaluated in the GRASS GIS software (version 8.4.0, creator: GRASS Development Team, original location: Champaign, Illinois, USA, currently multinational project) by means of unsupervised classification using the k-means clustering algorithm and supervised classification using the Support Vector Machine (SVM) algorithm. The land cover types were identified using machine learning (ML)-based analysis of the spectral reflectance of the multispectral images. The results based on the processed multispectral images indicated a decrease in savannas, an increase in croplands and agricultural lands, a decline in forests, and changes to coastal wetlands, including mangroves with high biodiversity. The practical aim is to describe a novel method of creating land cover maps using RS data for each class and to improve accuracy. We accomplish this by calculating the areas occupied by 10 land cover classes within the target area for six consecutive years. Our results indicate that, in comparing the performance of the algorithms, the SVM classification approach increased the accuracy, with 98% of pixels being stable, which shows qualitative improvements in image classification. This paper contributes to the natural resource management and environmental monitoring of Senegal, West Africa, through advanced cartographic methods applied to remote sensing of Earth observation data.

Multi-Class Support Vector Machine Classification for Detecting Alopecia Areata and Scalp Diseases

Alopecia Areata is a health condition marked by the absence of hair in specific regions, such as the scalp, face, and parts of the body. It occurs due to an autoimmune reaction where the body's immune system erroneously targets the hair follicles, leading to irregular hair loss patterns. It can affect people of all ages, genders, and ethnicities, and it is estimated to affect about 2% of the population worldwide. Timely identification and precise diagnosis of this condition are crucial in order to implement effective treatment strategies. The most common type of alopecia is alopecia areata (AA), which is typically detected and diagnosed using medical image processing models. In this study, we describe a unique method for image processing that incorporates a multiclass support vector machine classification approach. Our proposed methodology aims to attain accurate detection and categorization of a wide range of scalp issues, encompassing Alopecia Areata and other related conditions. The proposed approach entails capturing images of individuals with alopecia disease, enhancing the quality of the images through preprocessing techniques, and extracting distinctive features from scalp images using a range of image processing methods. Next, the extracted features are fed into a Multi-class SVM classifier, and a machine learning model is trained to achieve precise classification of the various conditions associated with alopecia areata. The evaluation of the proposed method using hair and scalp image databases demonstrates that the Multi-class SVM model achieves an accuracy of 89.3%, outperforming other models in terms of classification accuracy.

MIXTURE FEATURE EXTRACTION BASED ON LOCAL BINARY PATTERN AND GREY-LEVEL CO-OCCURRENCE MATRIX TECHNIQUES FOR MOUTH EXPRESSION RECOGNITION

Some academics struggle to recognize facial emotions based on pattern recognition. In general, this recognition utilizes all facial features. However, this study was limited to identifying facial emotions in a single facial region. In this study, lips, one of the facial features that can reveal a person's expression, are utilized. Using a combination of local binary pattern feature extraction (LBP) and grey level co-occurrence matrix (GLCM) methods and a multiclass support vector machine classification approach for feature extraction in facial images. The concept begins with image segmentation to create an image of a mouth. Experiments were also conducted for various tests, and the outcomes of these experiments revealed a recognition performance of up to 95%. This result was obtained through experiments in which 10% to 40% of the data were evaluated. These findings are beneficial and can be applied to expression recognition in online learning media to monitor the audience's condition directly.

Can the Physical Parameters with the Support Vector Machine (SVM) Method Able to Classify Benign and Malignant Breast Cancer?

Objective: Evaluating the diagnostic performance of SVM to classify benign and malignant by performing a meta-analysis.
 Methods: The data used for this study were secondary data. It consisted of 221 mammogram images (mean age 57.5 years) with 164 malignant and 57 benign, taken from a radiological database that has been examined by a radiologist with more than 20 years of experience. Also, histopathological record data that had been examined by an oncologist with more than 20 years of experience. Mammograms were taken from January 2022 to June 2022. In all, 221 mammograms consisting of 164 malignant and 57 benign were used as SVM method training, and 20 mammograms consisting of 10 malignant and 10 benign were used to test the performance of the SVM method. It was then evaluated using pathology results as the gold standard.
 Results: Benign had a significantly lower deviation (an average of 29.2661230 ± 10.14916673) than malignant (an average of 33.1841234 ± 11.70238757). The SVM method performance value obtained the values ​​of TP, FP, TN, FN, accuracy, sensitivity, Specificity, and Precision, respectively 7,7, 3, 3, 50%, 70%, 30%, and 50%.
 Conclusion: A proper performance to distinguish benign and malignant can be obtained using the physical deviation parameters with the SVM classification approach. However, these findings should be proven in larger datasets with different mammographic scanners. Our meta-analysis shows that the physical parameters and SVM have high sensitivity but low specificity. Of the nine physical parameters in the mammogram, only the parameter deviation was significant to distinguish between benign and malignant. The SVM method proved to be able to differentiate between benign and malignant.

Brain structure-function coupling provides signatures for task decoding and individual fingerprinting

Brain signatures of functional activity have shown promising results in both decoding brain states, meaning distinguishing between different tasks, and fingerprinting, that is identifying individuals within a large group. Importantly, these brain signatures do not account for the underlying brain anatomy on which brain function takes place. Structure-function coupling based on graph signal processing (GSP) has recently revealed a meaningful spatial gradient from unimodal to transmodal regions, on average in healthy subjects during resting-state. Here, we explore the specificity of structure-function coupling to distinct brain states (tasks) and to individual subjects. We used multimodal magnetic resonance imaging of 100 unrelated healthy subjects from the Human Connectome Project both during rest and seven different tasks and adopted a support vector machine classification approach for both decoding and fingerprinting, with various cross-validation settings. We found that structure-function coupling measures allow accurate classifications for both task decoding and fingerprinting. In particular, key information for fingerprinting is found in the more liberal portion of functional signals, with contributions strikingly localized to the fronto-parietal network. Moreover, the liberal portion of functional signals showed a strong correlation with cognitive traits, assessed with partial least square analysis, corroborating its relevance for fingerprinting. By introducing a new perspective on GSP-based signal filtering and FC decomposition, these results show that brain structure-function coupling provides a new class of signatures of cognition and individual brain organization at rest and during tasks. Further, they provide insights on clarifying the role of low and high spatial frequencies of the structural connectome, leading to new understanding of where key structure-function information for characterizing individuals can be found across the structural connectome graph spectrum.

Classification of gear pitting damage using vibration measurements

Abstract In future aero engines, a planetary gearbox is to be integrated between fan and turbine to increase the efficiency and bypass ratio. This gearbox has to be monitored during operation to detect possible gearbox faults such as gear wear or gear pitting at an early stage. This paper presents a method consisting of vibration measurement, sensor-dependent feature extraction and support-vector machine (SVM)-based classification of pitting for gear condition monitoring. Several gears were loaded with a constant torque on a standardized back-to-back test rig to provoke pitting, and the pitting amount was captured during the tests with a camera. Features are extracted from accelerometers and an acoustic emission sensor, and based on the results of the visually recorded pitting surface, SVM classification is applied to identify the pitting defect. In this contribution, two different SVM classification approaches are investigated. One approach uses a Two-Class SVM, where tests from one gearset are used for SVM training and another approach utilizes a One-Class SVM based on outlier detection. Both methods show that single tooth pitting defects with a relative pitting area of less than 1 % can be effectively identified, whereas the One-Class SVM method showed a higher pitting detection accuracy.

Monitoring urban expansion using SVM classification approach in Khenifra city (Morocco)

Urban expansion is a major process in developing countries such as Morocco. Since their appearance in the 1970s, digital remote-sensing space sensors have been used to provide images with spatial and spectral resolutions that are adequate to comprehend the phenomenon of urban expansion. Become real tools for cities management and planning, through the many applications. Using Landsat images, we studied Khenifra’s (Moroccan city) urban area evolution and extension to quantify its impact on current landscape. A supervised classification using support vector machines (SVM) was employed to extract the urban state in three periods 1991, 2000 and 2017. The results reveal a clear progression of total occupied urban space from 12% in 2000 until 36% in 2017. This expansion is unevenly distributed in space, and at slightly different rates depending on the period considered. This phenomenon of significant urban expansion is largely related to population growth and the rural exodus.

Tree Species Classification in Mixed Deciduous Forests Using Very High Spatial Resolution Satellite Imagery and Machine Learning Methods

Spatially explicit information on tree species composition is important for both the forest management and conservation sectors. In combination with machine learning algorithms, very high-resolution satellite imagery may provide an effective solution to reduce the need for labor-intensive and time-consuming field-based surveys. In this study, we evaluated the possibility of using multispectral WorldView-3 (WV-3) satellite imagery for the classification of three main tree species (Quercus robur L., Carpinus betulus L., and Alnus glutinosa (L.) Geartn.) in a lowland, mixed deciduous forest in central Croatia. The pixel-based supervised classification was performed using two machine learning algorithms: random forest (RF) and support vector machine (SVM). Additionally, the contribution of gray level cooccurrence matrix (GLCM) texture features from WV-3 imagery in tree species classification was evaluated. Principal component analysis confirmed GLCM variance to be the most significant texture feature. Of the 373 visually interpreted reference polygons, 237 were used as training polygons and 136 were used as validation polygons. The validation results show relatively high overall accuracy (85%) for tree species classification based solely on WV-3 spectral characteristics and the RF classification approach. As expected, an improvement in classification accuracy was achieved by a combination of spectral and textural features. With the additional use of GLCM variance, the overall accuracy improved by 10% and 7% for RF and SVM classification approaches, respectively.

Automatic feature scaling and selection for support vector machine classification with functional data

FunctionalData Analysis (FDA) has become a very important field in recent years due to its wide range of applications. However, there are several real-life applications in which hybrid functional data appear, i.e., data with functional and static covariates. The classification of such hybrid functional data is a challenging problem that can be handled with the Support Vector Machine (SVM). Moreover, the selection of the most informative features may yield to drastic improvements in the classification rates. In this paper, an embedded feature selection approach for SVM classification is proposed, in which the isotropic Gaussian kernel is modified by associating a bandwidth to each feature. The bandwidths are jointly optimized with the SVM parameters, yielding an alternating optimization approach. The effectiveness of our methodology was tested on benchmark data sets. Indeed, the proposed method achieved the best average performance when compared to 17 other feature selection and SVM classification approaches. A comprehensive sensitivity analysis of the parameters related to our proposal was also included, confirming its robustness.

Machine learning based assessment of auditory threshold perception in human beings

This study presents a novel approach to assess the perception of auditory Absolute threshold (ATTh) in healthy individuals exposed to noise and solvents in their occupational environment using machine learning approaches. 396 subjects with no known history of auditory pathology were chosen from three groups, namely, employees from Chemical Industries (CI), Fabrication Industries (FI), and professional Basketball Players (BP), with each category having 132 subjects. Absolute Threshold Test (ATT) was developed using MATLAB and the experiment was conducted in a silent, noise-free environment. ATTh was obtained twice, during the commencement and conclusion of the employees’ workshift in CI and FI. For BP, ATTh was obtained before and after their basketball training sessions and was used as features for binary SVM classification approach, in which the RBF kernel-based technique was found to provide maximum accuracy as compared to linear and quadratic approach. For three-class classification, MLP neural network with Levenberg–Marquardt training function in the hidden layer and Mean Square Error function in the output layer was found to be optimal along with k-Nearest Neighbor (kNN) and Support Vector Machine (SVM) approach using Radial Basis Kernel Function (RBF), in which, an accuracy of 81.06% was observed in kNN approach and 92.4% using MLP neural network approach, whereas SVM yielded an accuracy of 93.94% in the classification of the subjects into CI, FI and BP, showing that the SVM outperformed kNN and MLP neural network for healthy subjects based on their occupational exposure/professional sports training. Such machine learning approaches could further be probed into, to improve the accuracy of classification. Also, such techniques can help in real-time classification of subjects based on their occupational exposure so as to predict and prevent plausible permanent hearing dysfunction due to occupational exposure as well as to aid in sports rehabilitation and training programs to assess the auditory perceptive abilities of the individuals.

Assessment of Land Cover Changes Using Taguchi-Based Optimized SVM Classification Approach

Land use and land cover mapping is of great importance in many research areas. The main objective of this study was to assess land cover changes using support vector machine (SVM) classification approach. The most important challenge in this case is to determine the optimum structure of classification method. The optimization Taguchi method was performed to optimize the structure of SVM. Results showed that Taguchi method can be effectively used to cope with this problem. In fact, using Taguchi method to optimize SVM parameters can significantly decrease the number of classification tests. In this study, land use/cover maps of the Yazd–Ardakan plain were produced in three dates of 1990, 2002 and 2015. Maximum and minimum accuracy of the classification was 0.9 and 0.85 for 2015 and 1990 imagery, respectively. This may be related to the differences in the quality of the imagery. The land use/cover changes were then assessed based on the best classification results. It was shown that maximum change in the land use/cover was taken place in residential areas.

Motion artifact detection in colonoscopy images

Abstract Computer-aided detection is an integral part of medical image evaluation process because examination of each image takes a long time and generally experts’ do not have enough time for the elimination of images with motion artifact (blurred images). Computer-aided detection is required for both increasing accuracy rate and saving experts’ time. Large intestine does not have straight structure thus camera of the colonoscopy should be moved continuously to examine inside of the large intestine and this movement causes motion artifact on colonoscopy images. In this study, images were selected from open-source colonoscopy videos and obtained at Kayseri Training and Research Hospital. Totally 100 images were analyzed half of which were clear. Firstly, a modified version of histogram equalization was applied in the pre-processing step to all images in our dataset, and then, used Laplacian, wavelet transform (WT), and discrete cosine transform-based (DCT) approaches to extract features for the discrimination of images with no artifact (clear) and images with motion artifact. The Laplacian-based feature extraction method was used for the first time in the literature on colonoscopy images. The comparison between Laplacian-based features and previously used methods such as WT and DCT has been performed. In the classification phase of our study, support vector machines (SVM), linear discriminant analysis (LDA), and k nearest neighbors (k-NN) were used as the classifiers. The results showed that Laplacian-based features were more successful in the detection of images with motion artifact when compared to popular methods used in the literature. As a result, a combination of features extracted using already existing approaches (WT and DCT) and the Laplacian-based methods reached 85% accuracy levels with SVM classification approach

Performance evaluation of support vector machine classification approaches in data mining

At present, knowledge extraction from the given data set plays a significant role in all the fields in our society. Feature selection process used to choose a few relevant features to achieve better classification performance. The existing feature selection algorithms consider the job as a single objective problem. Selecting attributes is prepared by the combination of attribute evaluation and search method using the WEKA Machine Learning Tool. The proposed method is performed in three phases. In the first step, support vector classifiers are implemented with four different kernel methods such as linear function, Polynomial function, Radial basis function and sigmoid functions to classify data items. In the second step, classifier subset evaluation is applied to feature selection, along with the SVM classification for optimizing feature vectors and this obtains the maximum accuracy. In the third step, introducing new kernel approach which generates the maximum accuracy in classification compared to the other four kernel methods. From the experimental analysis, SVM with the proposed kernel approach has produced maximum accuracy over other kernel methods.

Optimal customer relationship management in telecalling industry by using data mining and business intelligence

Decision making system in telecommunication industries plays a more important role where it is required to find customer churn. Customer churn prediction requires finding out and analyzing the information about the business data intelligence techniques which can be done efficiently by adapting the business intelligence techniques. Business intelligence provides tools to predict and analyze the historical, current and predictive views of business operations. However, this would be more complex task with high volume of data which are gathered from million of telephone users for the time being. It can be handled effectively by introducing the data mining techniques which select the most useful information from the gathered data set from which decision making can be done efficiently. In this research method, telecommunication industry is considered in which customer churn prediction application is focused. The main goal of this research method is to introduce the data mining technique which can select the most useful information from the telecommunication industry dataset. This is done by introducing the Hybrid Genetic Algorithm with Particle Swarm Optimization (HGAPSO) method which can select the most useful information. In this research, the hybrid HGAPSO combines the advantages of PSO and GA optimally. From the selected information, decision making about the customer churn prediction can be done accurately. Finally decision making is done by predicting the customer behaviour using Support Vector Machine classification approach. The performance metrics are considered such as precision, recall, f-measure, accuracy, True Positive Rate (TPR), False Positive Rate (FPR), time complexity and ROC. Experimental results demonstrated that HGAPSO provides highly scalable which is used for prediction examination in the business intelligence.

TESTING THE POTENTIAL OF VEGETATION INDICES FOR LAND USE/COVER CLASSIFICATION USING HIGH RESOLUTION DATA

Abstract. Accurate and reliable land use/land cover (LULC) information obtained by remote sensing technology is necessary in many applications such as environmental monitoring, agricultural management, urban planning, hydrological applications, soil management, vegetation condition study and suitability analysis. But this information still remains a challenge especially in heterogeneous landscapes covering urban and rural areas due to spectrally similar LULC features. In parallel with technological developments, supplementary data such as satellite-derived spectral indices have begun to be used as additional bands in classification to produce data with high accuracy. The aim of this research is to test the potential of spectral vegetation indices combination with supervised classification methods and to extract reliable LULC information from SPOT 7 multispectral imagery. The Normalized Difference Vegetation Index (NDVI), the Ratio Vegetation Index (RATIO), the Soil Adjusted Vegetation Index (SAVI) were the three vegetation indices used in this study. The classical maximum likelihood classifier (MLC) and support vector machine (SVM) algorithm were applied to classify SPOT 7 image. Catalca is selected region located in the north west of the Istanbul in Turkey, which has complex landscape covering artificial surface, forest and natural area, agricultural field, quarry/mining area, pasture/scrubland and water body. Accuracy assessment of all classified images was performed through overall accuracy and kappa coefficient. The results indicated that the incorporation of these three different vegetation indices decrease the classification accuracy for the MLC and SVM classification. In addition, the maximum likelihood classification slightly outperformed the support vector machine classification approach in both overall accuracy and kappa statistics.

DETERMINATION OF LAND COVER/LAND USE USING SPOT 7 DATA WITH SUPERVISED CLASSIFICATION METHODS

Abstract. Land use/ land cover (LULC) classification is a key research field in remote sensing. With recent developments of high-spatial-resolution sensors, Earth-observation technology offers a viable solution for land use/land cover identification and management in the rural part of the cities. There is a strong need to produce accurate, reliable, and up-to-date land use/land cover maps for sustainable monitoring and management. In this study, SPOT 7 imagery was used to test the potential of the data for land cover/land use mapping. Catalca is selected region located in the north west of the Istanbul in Turkey, which is mostly covered with agricultural fields and forest lands. The potentials of two classification algorithms maximum likelihood, and support vector machine, were tested, and accuracy assessment of the land cover maps was performed through error matrix and Kappa statistics. The results indicated that both of the selected classifiers were highly useful (over 83% accuracy) in the mapping of land use/cover in the study region. The support vector machine classification approach slightly outperformed the maximum likelihood classification in both overall accuracy and Kappa statistics.

Human cell structure-driven model construction for predicting protein subcellular location from biological images.

The systematic study of subcellular location pattern is very important for fully characterizing the human proteome. Nowadays, with the great advances in automated microscopic imaging, accurate bioimage-based classification methods to predict protein subcellular locations are highly desired. All existing models were constructed on the independent parallel hypothesis, where the cellular component classes are positioned independently in a multi-class classification engine. The important structural information of cellular compartments is missed. To deal with this problem for developing more accurate models, we proposed a novel cell structure-driven classifier construction approach (SC-PSorter) by employing the prior biological structural information in the learning model. Specifically, the structural relationship among the cellular components is reflected by a new codeword matrix under the error correcting output coding framework. Then, we construct multiple SC-PSorter-based classifiers corresponding to the columns of the error correcting output coding codeword matrix using a multi-kernel support vector machine classification approach. Finally, we perform the classifier ensemble by combining those multiple SC-PSorter-based classifiers via majority voting. We evaluate our method on a collection of 1636 immunohistochemistry images from the Human Protein Atlas database. The experimental results show that our method achieves an overall accuracy of 89.0%, which is 6.4% higher than the state-of-the-art method. The dataset and code can be downloaded from https://github.com/shaoweinuaa/. dqzhang@nuaa.edu.cn Supplementary data are available at Bioinformatics online.

Bio-inspired classifier for road extraction from remote sensing imagery

An adaptive approach for road extraction inspired by the mechanism of primary visual cortex (V1) is proposed. The motivation is originated by the characteristics in the receptive field from V1. It has been proved that human or primate visual systems can distinguish useful cues from real scenes effortlessly while traditional computer vision techniques cannot accomplish this task easily. This idea motivates us to design a bio-inspired model for road extraction from remote sensing imagery. The proposed approach is an improved support vector machine (SVM) based on the pooling of feature vectors, using an improved Gaussian radial basis function (RBF) kernel with tuning on synaptic gains. The synaptic gains comprise the feature vectors through an iterative optimization process representing the strength and width of Gaussian RBF kernel. The synaptic gains integrate the excitation and inhibition stimuli based on internal connections from V1. The summation of synaptic gains contributes to pooling of feature vectors. The experimental results verify the correlation between the synaptic gain and classification rules, and then show better performance in comparison with hidden Markov model, SVM, and fuzzy classification approaches. Our contribution is an automatic approach to road extraction without prelabeling and postprocessing work. Another apparent advantage is that our method is robust for images taken even under complex weather conditions such as snowy and foggy weather. © 2014 SPIE.

A Comparative Analysis of EO-1 Hyperion, Quickbird and Landsat TM Imagery for Fuel Type Mapping of a Typical Mediterranean Landscape

Forest fires constitute a natural disturbance factor and an agent of environmental change with local to global impacts on Earth’s processes and functions. Accurate knowledge of forest fuel extent and properties can be an effective component for assessing the impacts of possible future wildfires on ecosystem services. Our study aims to evaluate and compare the spectral and spatial information inherent in the EO-1 Hyperion, Quickbird and Landsat TM imagery. The analysis was based on a support vector machine classification approach in order to discriminate and map Mediterranean fuel types. The fuel classification scheme followed a site-specific fuel model within the study area, which is suitable for fire behavior prediction and spatial simulation. The overall accuracy of the Quickbird-based fuel type mapping was higher than 74% with a quantity disagreement of 9% and an allocation disagreement of 17%. Both classifications from the Hyperion and Landsat TM fuel type maps presented approximately 70% overall accuracy and 16% allocation disagreement. The McNemar’s test indicated that the overall accuracy differences between the three produced fuel type maps were not significant (p < 0.05). Based on both overall and individual higher accuracies obtained with the use of the Quickbird image, this study suggests that the high spatial resolution might be more decisive than the high spectral resolution in Mediterranean fuel type mapping.

Fast Support Vector Machine Classification for Large Data Sets

AbstractNormal support vector machine (SVM) algorithms are not suitable for classification of large data sets because of high training complexity. This paper introduces a novel two-stage SVM classification approach for large data sets. Fast clustering techniques are introduced to select the training data from the original data set for the first stage SVM, and a de-clustering technique is then proposed to recover the training data for the second stage SVM. The proposed two-stage SVM classifier has distinctive advantages on dealing with huge data sets such as those in bioinformatics. Finally, we apply the proposed method on several benchmark problems. Experimental results demonstrate that our approach has good classification accuracy while the training is significantly faster than other SVM classifiers.