ABSTRACT

---

Abstract
The importance of global features in the analysis of tissue images cannot be overemphasized especially in texture image classification and retrieval. This paper develops different techniques for detection, classification and analysis of diseases pattern in medical images. The research work studies the structure of biomedical images; and extracts the similarity features characterized by the Holder exponent for pattern classification. Global features from multi-fractal descriptors have been extracted and combined with local features from fractal descriptors to generate new descriptors for efficient analysis of images. The experimental approaches are validated using different extracted features during the classification process to determine the appropriate image features that could yield the maximum detection accuracy. The experimental results showed that the descriptors extracted from the combined features considerably improved the performance of the models. The results achieved in this paper have greatly demonstrated the importance of global features in the analysis and classification of pattern diseases.
Keywords: global features; feature extraction; biomedical images; local features; machine learning and multi-fractal analysis

---