Sonography, also known as ultrasound, has helped in identifying and treating infertile patients. Ultrasound imaging of the ovary's follicles reveals the type of cyst, the diversity of follicles, and the size of the follicles' response to hormonal imbalance. Image segmentation enriches the image's region of interest with additional data and accurately identifies the object and its background. However, because segmentation on ultrasound images is difficult due to noise, follicle identification can be streamlined and made more effective by combining photograph preprocessing with morphological operations. In order to classify PCOS Ovaries and Normal Ovaries, the machine learning algorithms Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Logistic Regression, and the proposed method are used. Physical identification is used to compare classification results. The proposed algorithm yields an accuracy of 98%.