An Intelligent Method for Predicting Skin Disease Using OM-Extreme Learning Machine

In terms of prevalence, skin disorders rank towards the top. Various organisms, including fungi, bacteria, allergens, viruses, and so on, can trigger skin illnesses. Improved speed and accuracy in the diagnosis of skin diseases has resulted from the development of lasers and Photonics-based medical technology. However, there is now a severe lack of affordable options for such diagnostics. Therefore, at the outset, image processing techniques aid in constructing an automated screening system for dermatology. To aid in the categorization of skin disorders, feature extraction is essential. Several methods exist where computer vision can aid in the diagnosis of skin disorders. Skin infections are frequent in Saudi Arabia because of the country's scorching, desert climate. This proposed method adds to the literature on the early diagnosis of skin diseases. To help diagnose skin conditions, to recommended using image processing. This technique uses digital images of the affected skin to determine the disease kind through the application of image analysis. The method to offer straightforward, quick, and can be done with just a camera and a computer.Preprocessing, feature extraction for segmentation, and model performance evaluation make up the suggested method. Preprocessing makes use of a histogram and an RGB to Grayscale conversion. Segmentation is performed with the help of adaptive thresholding. The GLCM, IQA, and complete reference methods are all utilized for feature extraction. Finally, OM-ELM is utilized to assess the model's performance. Comparing the suggested method to ELM and AE-ELM, it fares quite well.