Image Semantic Segmentation Method Based on Deep Convolutional Neural Network and Superpixel

Abstract

The invention discloses an image semantic segmentation method based on a deep convolutional neural network and superpixels, which overcomes the problem in the prior art that the accuracy of the existing semantic segmentation method still needs to be improved. The invention contains the following steps: 1: Train a deep convolutional network classification model from image to category label on the image classification data set; 2: Add a deconvolution layer to the deep convolutional neural network classification model, and perform semantic segmentation on the image Perform fine-tuning training on the data set to realize the mapping from image to image semantic segmentation results; 3: Input the test image into the deep convolutional neural network semantic segmentation model to obtain the semantic labels of each pixel, and at the same time send the test image to superpixel segmentation Algorithm to obtain several superpixel regions; 4: Fusion superpixels and semantic labels to obtain the final improved semantic segmentation results. This method improves the accuracy of existing image semantic segmentation methods, which is of great significance in image recognition and applications.

Description

technical field [0001] The invention relates to an image semantic segmentation method, in particular to an image semantic segmentation method based on a deep convolutional neural network and superpixels. Background technique [0002] With the continuous decline in the classification error rate of convolutional neural networks on public data sets, researchers began to pay attention to image pixel-level segmentation, that is, image semantic segmentation. Semantic feature is a higher-level feature that marks each pixel in the image as a corresponding category. The general method is to realize semantic segmentation by discriminating image regions. Carreira et al., Farabet et al., Girshick et al. divided the image into several regions by means of superpixels, and extracted regional features through a deep convolutional neural network to classify them to achieve semantic segmentation of the entire image. In 2015, Long et al. proposed an end-to-end semantic segmentation model base...

AI Technical Summary

Problems solved by technology

[0004] The present invention overcomes the problem that the accuracy of the existing semantic segmentation method still needs to be improved in the prior art, and provides an image semantic segmentation method based on a deep convolutional neural network and superpixels with better use effect

Images