基于改进SEGNET模型的图像语义分割  

Semantic Image Segmentation Based on Improved SEGNET Model

作　　者：罗嗣卿[1] 张志超 岳琪[1] LUO Siqing;ZHANG Zhichao;YUE Qi(School of Information and Computer Engineering,Northeast Forestry University,Harbin 150040,China)

机构地区：[1]东北林业大学信息与计算机工程学院,哈尔滨150040

出　　处：《计算机工程》2021年第4期256-261,共6页Computer Engineering

基　　金：国家自然科学基金青年项目“基于用户标签和主体兴趣的社会媒体信息推荐研究”(61806049)。

摘　　要：使用原始SEGNET模型对图像进行语义分割时,未对图像中相邻像素点间的关系进行考虑,导致同一目标中像素点类别预测结果不一致。通过在SEGNET结构中加入一条自上而下的通道,使得SEGNET包含的多尺度语义信息更加丰富,从而提升对每个像素点的类别预测精度,在模型中加入生成对抗网络以充分考虑空间中相邻像素点间关系。实验结果表明,该模型的语义分割效果相比原始SEGNET模型显著提升,且可有效解决SEGNET测试中出现的分类错误问题。When applied to semantic image segmentation,the original SEGNET model does not account for the relationship between adjacent pixels in the image,resulting in inconsistent prediction results of pixel categories in the same target.By adding a top-down channel in the SEGNET structure,the multi-scale semantic information of the SEGNET model is enriched,and the accuracy of category prediction for each pixel is improved.The generative adversarial network is added to the model to ensure that the model can consider the relationship between adjacent pixels in space.The experimental results show that the semantic segmentation effect of the improved SEGNET model is significantly improved compared with the original SEGNET model.It can effectively avoid the classification errors in the SEGNET test.

关 键 词：SEGNET模型 生成对抗网络 多尺度语义信息 相邻像素类别关系 特征融合 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]