Due to the large amount of information loss generated by traditional SegNet model during the sampling process,it cau-ses the accuracy of image semantic segmentation low.Therefore,a new encoder-decoder network structure with fusion residual con-nection is proposed.The multi-residual connection strategy is introduced to fully retain a large number of detailed information con-tained in multi-scale images,and reduce the information loss caused by sampling.In order to further accelerate the convergence effi-ciency of network training and improve the imbalance problem of samples,a cross-entropy loss function with balance factor is de-signed,and the imbalance phenomenon of positive and negative samples is emphatically optimized to train the model more efficient.Experimental results show that this method solves the problems of information loss and inaccurate segmentation in semantic segmenta-tion,and compared with SegNet model,the fine labeling mean intersection over union(mIoU)index of the network on Cityscapes dataset is increased by about 13％.
关键词
语义分割/残差连接/交叉熵损失函数/SegNet模型/深度学习
Key words
semantic segmentation/residual connection/cross entropy loss function/SegNet model/deep learning
维普
万方数据