Image Reconstruction Using Pixel Wise Support Vector Machine SVM Classification.
Journal: International Journal of Scientific & Technology Research (Vol.4, No. 2)Publication Date: 2015-02-15
Authors : Mohammad Mahmudul Alam Mia; Rashedul Islam; Ferdous Wahid; Shovasis Kumar Biswas;
Page : 232-235
Keywords : Index Terms Neural networks; Classification; SVM; Kernel functions; Training set.;
Abstract
Abstract Image reconstruction using support vector machine SVM has been one of the major parts of image processing. The exactness of a supervised image classification is a function of the training data used in its generation. In this paper we studied support vector machine for classification aspects and reconstructed an image using support vector machine. Firstly value of the random pixels is used as the SVM classifier. Then the SVM classifier is trained by using those values of the random pixels. Finally the image is reconstructed after cross-validation with the trained SVM classifier. Matlab result shows that training with support vector machine produce better results and great computational efficiency with only a few minutes of runtime is necessary for training. Support vector machine have high classification accuracy and much faster convergence. Overall classification accuracy is 99.5. From our experiment It can be seen that classification accuracy mostly depends on the choice of the kernel function and best estimation of parameters for kernel is critical for a given image.
Other Latest Articles
- Image Reconstruction Using Multi Layer Perceptron MLP And Support Vector Machine SVM Classifier And Study Of Classification Accuracy
- URL Mining Using Agglomerative Clustering Algorithm
- The Social Development Dimension Of The Nursing Profession In Managing HIV Cases
- Analysis Of Japans Economy Based On 2014 From Macroeconomics Prospects
- Study Of Gear Teeth Distortions Due To Heat Treatment
Last modified: 2015-06-28 04:08:23