This paper presents a deep learning-based image segmentation method. The method proposes a region growing based image segmentation method by studying the feature selection, feature extraction and segmentation model design of image segmentation. The method first preprocesses the image, then uses the fully convolutional network to extract the image features, and then uses the region growing model to segment the image. Finally, the semantic segmentation result of the image is obtained. Experimental results show that the method can effectively realize the semantic segmentation of images and has high segmentation accuracy.
