Method and apparatus for image registration, and ultrasonic diagnosis apparatus

Extracting a first feature set of the first image and the second feature set by correlation using the learning transformation neural network between the feature extraction unit and the feature of extracting a second feature set from the second video data in the first feature set It provides a video aligning apparatus comprising a matching unit for matching a first image and the second image based on the conversion so as to correspond to obtain a set of conversion characteristics conversion unit and converting the first set of feature sets.