Medical Image Segmentation Based On The Integration Of Learning Technology Research And Implementation

With the continuous development of medical imaging technology, the medicine is inseparable from the medical image processing technology, Disease diagnosis in medical imaging, scientific research, teaching, and played a significant role. Medical image segmentation medical image processing is a very important work, the field of image processing is a hot and difficult, due to the complexity of human body structure, the performance of medical imaging equipment, the interference of external electromagnetic radiation, medical images with fuzzy noise more features. In addition, different people in the organizational structure there are some small differences, these have brought a lot of medical image segmentation problems, and so far no algorithm capable of performing a variety of areas of image segmentation tasks.Ensemble learning is a new machine learning paradigm; it can significantly improve the generalization ability of learning systems through utilizing multiple learners to solve a problem. This paper attempts to research on digital image segmentation techniques base on the ensemble learning techniques. The main content of this dissertation is described as follows.(1)Describes the commonly used international medical image segmentation algorithm, and highlights the K means algorithm, FCM algorithm and region growing algorithm and application of these algorithms and the advantages and disadvantages.(2)A new algorithm of image segmentation based on Ensemble learning technology, namely W-MEANS is presented in the paper. This algorithm is an object removal method based on exemplar segmentation and can obtain improved performance over exemplar-based segmentation methods. The new segmentation algorithm accuracy and efficiency have a very significantly improvement, which is also proved integrated learning technology is a simple and effective method.(3)A new mixed model based on Ensemble learning technology is presented in the paper. The model is used in medical image segmentation, especially in the field of brain segmentation. Experimental results show that the model has better segmentation effect and precision. And this mode has the great practical value in the brain image segmentation .