A method and apparatus for image segmentation using residual convolution based deep learning network
The present invention relates to image segmentation method and apparatus using a residual convolution-based deep learning network. The image segmentation method using a residual convolution-based deep learning network, according to an embodiment of the present invention, comprises the steps of: (a)...
Saved in:
Main Authors | , |
---|---|
Format | Patent |
Language | English Korean |
Published |
23.09.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The present invention relates to image segmentation method and apparatus using a residual convolution-based deep learning network. The image segmentation method using a residual convolution-based deep learning network, according to an embodiment of the present invention, comprises the steps of: (a) obtaining input information; (b) inputting, into a first fire module comprising a squeeze layer composed of a first convolution filter and an expand layer composed of a second convolution filter and a third convolution filter, the input information; (c) concatenating output information of the first fire module and the input information obtained through residual connection; and (d) performing encoding by max pooling a result value calculated by the concatenation. The present invention can perform smooth learning of a deep learning network architecture through a residual unit.
본 발명은 잔차 컨볼루션 기반 딥러닝 네트워크를 이용한 영상 분할 방법 및 장치에 관한 것이다. 본 발명의 일 실시예에 따른 잔차 컨볼루션 기반 딥러닝 네트워크를 이용한 영상 분할 방법은, (a) 입력 정보를 획득하는 단계; (b) 상기 입력 정보를 제1 컨볼루션 필터로 구성된 스퀴즈 레이어(squeeze layer)와 제2 컨볼루션 필터 및 제3 컨볼루션 필터로 구성된 확장 레이어(expand layer)를 포함하는 제1 파이어(fire) 모듈에 입력하는 단계; (c) 상기 제1 파이어 모듈의 출력 정보와 잔차 연결(residual connection)을 통해 획득된 상기 입력 정보를 연결(concatenate)하는 단계; 및 (d) 상기 연결하여 산출된 결과값을 맥스 풀링(max pooling)하여 인코딩을 수행하는 단계;를 포함할 수 있다. |
---|---|
Bibliography: | Application Number: KR20210034278 |