Abstract: To tackle the problem of object imperfection and region abruption in existing deep salient object detection algorithms, a multiscale context enhanced fully convolutional network(MCE-FCN) based on non-local deep features(NLDF) is proposed. The network includes four modules which are responsible for multi-level features extraction, multiscale context feature enhancement, contrast feature extraction and local-global fusion. Firstly, multi-level local features are extracted from VGG16 model. Secondly, multiscale context information is exploited to enhance the local features. Then, combined loss function is designed to extract contrast features. Finally, the saliency map is predicted in the local-global fusion way. The proposed algorithm is compared with other algorithms in ECSSD, HKU-IS and DUT-OMRON datasets. Experimental results show that the proposed method is more robust for natural images with complex scenes and has more effective suppression on background noise, and the result salient object regions are more consistent and complete.