References in periodicals archive ?
The GLRLM features include short run emphasis (SRE), long run emphasis (LRE), run length nonuniformity (RLN), gray level nonuniformity (GLN), run percentage (RP), low gray-level run emphasis (LGRE), high gray-level run emphasis (HGRE), short run low gray-level emphasis (SRLGE), short run high gray-level emphasis (SRHGE), long run low gray-level emphasis (LRLGE), and long run high gray-level emphasis (LRHGE) .
Mixed features are a combined set of all features, that is, intensity histogram features, GLCM features, GLRLM features, and invariant moment features.
Results of the training data show that selected GLRLM features yield an accuracy of 90%, sensitivity of 86.7%, specificity of 93.3%, false positive rate computed of 6.7%, false negative rate computed of 13.3%, and misclassification rate of 10%.
The results of the testing data show that GLRLM histogram features yield an accuracy of 95%, sensitivity of 95%, specificity of 95%, false positive rate computed of 5%, false negative rate computed of 5%, and misclassification rate of 5%.
The results showed that the selected mixed features yielded an accuracy of around 91.67% on the training set as compared to GLRLM features and GLCM features, which yielded an accuracy of around 90% and 86.7%, respectively, on the training set.
For testing the network against test data, GLRLM features gave better result.
Five sets of features such as GLCM, intensity histogram, GLRLM, invariant moments, and mixed features were extracted.
Acronyms browser ?
Full browser ?