Normalization and Calculation of Energy, Entropy, and Moment of Inertia Using Gray-Level Co-occurrence Matrix

This process involves normalizing the co-occurrence matrix using the gray-level co-occurrence matrix (GLCM) and computing four key texture parameters: energy, entropy, moment of inertia, and correlation. The GLCM is a statistical method for characterizing texture features by calculating the frequency of co-occurrence between different pixel pairs, thereby capturing spatial relationships among image pixels. When we normalize the co-occurrence matrix using GLCM, we can effectively quantify the relationships between different pixels and extract their distinctive characteristics. By calculating these four texture parameters, we gain deeper insights into the texture features within an image, which can be applied to various image analysis and processing tasks. In code implementation, this typically involves using functions like `graycomatrix` in MATLAB to generate the GLCM, followed by normalization and computation of texture features using custom algorithms or built-in functions such as `graycoprops`. Key steps in the implementation include: 1. Generating the GLCM for specific pixel distances and angles. 2. Normalizing the GLCM by dividing each element by the sum of all elements to obtain probability values. 3. Calculating energy as the sum of squared elements of the normalized GLCM, representing uniformity. 4. Computing entropy as the negative sum of each element multiplied by its logarithm, indicating randomness. 5. Determining the moment of inertia (contrast) by measuring the intensity variations between pixels and their neighbors. 6. Evaluating correlation to assess linear dependencies between pixel pairs. These parameters are crucial for applications like texture classification, defect detection, and medical image analysis.