Deep learning network construction method and system applicable to semantic segmentation

Abstract

The invention discloses a deep learning network construction method and system applicable to semantic segmentation. According to the invention, based on the deconvolution semantic segmentation, by considering the characteristic that a conditional random field is quite good for edge optimization, the conditional random field is explained to be a recursion network to be fused in a deconvolution network and end to end trainings are performed, so the parameter learning in the convolution network and the recursion network is allowed to act with each other and a better integration network is trained; through combined training of the deconvolution network and the conditional random field, quite accurate detail and shape information is obtained, so a problem of inaccuracy of image edge segmentation is solved; by use of the strategy of combining the multi-scale input and multi-scale pooling, a problem is solve that a big target is excessively segmented or segmentation of a small target is ignored generated by the single receptive field in the semantic segmentation; and by expanding the classic deconvolution network, by use of the united training of the conditional random field and the multi-feature information fusion, accuracy of the semantic segmentation is improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and more specifically relates to a deep learning network construction method and system suitable for semantic segmentation. Background technique [0002] With the explosive growth of network data, big data image processing and recognition has gradually become a popular direction, and deep learning technology has become an indispensable research tool for big data. Although the development time of deep learning is not long and the theoretical reserve is incomplete, the construction methods of deep network emerge in endlessly, and the application effect in the direction of computer vision is remarkable. The use of deep learning for visual perception is based on the visual mechanism of the human brain, and the multi-level network design is analogous to the hierarchical information processing visual system. The processing of the human visual system is divided into the following parts. The pix...

AI Technical Summary

Problems solved by technology

[0006] For the above defects or improvement needs of the prior art, the purpose of the present invention is to provide a deep learning network construction method and system suitable for semantic segmentation, thereby solving the problem of the existing convolutional neural network suitable for semantic segmentation. Technical issues with lower segmentation accuracy

Images