Feature extraction of ophthalmic image based on machine learning

The threshold range in the field of ophthalmic image does not match the pixels, which leads to the low detection ratio of feature points in ophthalmic image. Therefore, the research of ophthalmic image feature extraction method based on machine learning. Based on machine learning technology, a feature extraction method of ophthalmic image is designed. The feature blocks of ophthalmic image are divided, and the window of threshold selection is determined according to the comparison of pixel and threshold range. The shape parameters of ophthalmic image feature based on machine learning are calculated, and the feature extraction model of ophthalmic image is constructed. The experimental results show that the highest detection ratio of feature points is 0.75. Therefore, the eye image feature extraction method based on machine learning is better.