Improving the process of recognition the treated teeth in the Panoramic images based on the optimal features selection: تحسين عملية التعرف على الأسنان المعالجة في الصور البانورامية بالاعتماد على الاختيار الأمثل للسمات

    Due to the significant development in the field of machine learning and patterns recognitions, the area of image processing has an important role in this context, especially in the field of medical images of various kinds. In this research, we have been developed powerful, simple, cost-effective and more accurate interpretation algorithm for recognition treated teeth In the X-ray images. There are many difficulties in determining the objects such as it is difficult to interpret the radiographic image because there are very subtle differences in X-rays, poor image quality representation and the splitting of all the teeth in the image of radiographic imaging. In this research, comprehensive methodology was proposed that enables the identification of the teeth that have been treated by the optimal features selection. Where the digital image was processed and then extracted statistical features of it using second order statistical and gray level co-occurrence matrix GLCM. Then, the optimal features were chosen, which express the pattern to be recognized, be categorized then to classify the extracted features. The results obtained showed great accuracy in the results obtained, where the features of homogeneity, contrast and correlation were chosen as expressive features of pulp canal therapy with standard deviations, 0.647%, 1.602% and 1.925% respectively, as well as the reconstructed dental crown with standard deviations of the aforementioned features", 1.07%, 2.80% and 0.57%, respectively, because they gave the lowest values of the standard deviation and thus the lowest percentage of error and therefore can be adopted as expressive features of the treated tooth.