Multi-resolutional gabor filter in texture analysis
Pattern Recognition Letters
Statistical texture characterization from discrete wavelet representations
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
Computed tomography (CT) images have been widely used for liver disease diagnosis. The CT images have been used for the analysis as these images are more clear compared to other imaging techniques. This paper focuses on developing method for classifying liver tumor from CT images using texture analysis and neural network classifier. The Co-occurrence matrices are used to extract feature like contrast, correlation, entropy, homogeneity, energy. Then the probabilistic neural network is trained to classify liver tumors as malignant and benign. The proposed system was evaluated by several liver images. It produces accuracy of 93.75%. The performance of the proposed system is also evaluated by calculating specificity, sensitivity.