Scene semantic segmentation method based on deep learning

Abstract

The invention discloses a scene semantic segmentation method based on deep learning. The method comprises a training stage and a testing stage; the training stage comprises the steps: employing Resnet101 for pre-training on a COCO data set to obtain a pre-training model, and then loading the pre-training model into a constructed convolutional neural network to extract a low-level feature image; performing high-level feature extraction and feature fusion on the low-level feature image through a feature enhancement network, an adaptive deformable cavity space convolution pooling pyramid networkand a feature attention network in sequence, and finally outputting a semantic segmentation Mask graph through an up-sampling operation, and obtaining a convolutional neural network semantic segmentation model weight; the test stage comprises the steps: inputting a PASCAL VOC 2012 or Cityscapes test data set into the weight of the convolutional neural network semantic segmentation model, and obtaining a predicted semantic segmentation Mask graph. According to the method, the boundary contour precision of the target image and the accuracy of scene semantic segmentation can be improved.

Description

technical field [0001] The invention relates to a computer vision technology, relates to the field of image semantic segmentation, and in particular relates to a scene semantic segmentation method based on deep learning. Background technique [0002] Hinton et al. proposed the basic concept of deep learning in 2006, and gradually applied it to the field of computer vision, such as image, sound and text, which accelerated the pace of solving complex tasks in the field of computer vision and improved the accuracy of various tasks. . [0003] In the image classification task, through the development of the following years, a series of classic networks such as AlexNet, VGG, GoogLeNet, Resnet, Inception, etc. are proposed. These networks are still active in the current convolutional neural network. For example, Resnet and Inception will be applied In tasks such as image segmentation and target detection, the features of the image are extracted in the form of a backbone network, ...

AI Technical Summary

Problems solved by technology

[0006] Existing research methods can achieve image segmentation of target categories, but there are also some shortcomings. These shortcomings are mainly reflected in the fact that the target outline cannot be divided and recognized in detail. Therefore, in order to perform accurate semantic segmentation on the target scene, it is necessary to consider Different scene categories will have different deformation capabilities. How to better model these images with multiple deformation capabilities is the key to improving the accuracy of deep learning networks for scene segmentation.

Images