Fusing features from different feature descriptors or different convolutional layers can improve the of understanding of scene and enhance the classification accuracy. In this paper, we propose a hierarchical deep texture feature fusion network, abbreviated as HDTFF-Net, aiming to improve the classification accuracy of high-resolution remote sensing scene classification (RSSC). The proposed HDTFF-Net can effectively combine the shallow texture information from manual features and the deep texture information by convolutional neural networks (CNNs). First, for deeply excavating the multi-scale and multi-directional shallow texture features in images, an improved Wavelet feature extraction module and a Gabor feature extraction module are designed by fully fusing the structural features into the backbone neural network. Then, to make the output texture features more discriminative and interpretative, we incorporate the above texture feature extraction modules into traditional convolutional neural networks, and design two improved deep networks, namely Wave-CNN and Gabor-CNN. Finally, according to the Dempster- Shafer (D-S) evidence theory, the designed Wave-CNN and Gabor-CNN are fused with the traditional CNN (Tra-CNN) by a decision-level fusion strategy, which can effectively capture the deep texture features by different feature descriptors and improve the classification performance. Experiments on high-resolution remote sensing images demonstrate the effectiveness of the proposed HDTFF-Net, and verify that it can greatly improve the classification performance.

Semantic segmentation of high-resolution remote sensing images holds paramount importance in the field of remote sensing. To better excavate and fully fuse the features in high-resolution remote sensing images, this paper introduces a novel Global and Local Feature Fusion Network, abbreviated as GLF-Net, by incorporating the extensive contextual information and refined fine-grained features. The proposed GLF-Net, devised as an encoder–decoder network, employs the powerful ResNet50 as its baseline model. It incorporates two pivotal components within the encoder phase: a Covariance Attention Module (CAM) and a Local Fine-Grained Extraction Module (LFM). And an additional wavelet self-attention module (WST) is integrated into the decoder stage. The CAM effectively extracts the features of different scales from various stages of the ResNet and then encodes them with graph convolutions. In this way, the proposed GLF-Net model can well capture the global contextual information with both universality and consistency. Additionally, the local feature extraction module refines the feature map by encoding the semantic and spatial information, thereby capturing the local fine-grained features in images. Furthermore, the WST maximizes the synergy between the high-frequency and the low-frequency information, facilitating the fusion of global and local features for better performance in semantic segmentation. The effectiveness of the proposed GLF-Net model is validated through experiments conducted on the ISPRS Potsdam and Vaihingen datasets. The results verify that it can greatly improve segmentation accuracy.

Aiming at the problems of long synchronization time and instability issues in the existing 3D information hiding algorithms, a novel 3D projection sampling strategy for 3D point cloud data is proposed. Projecting all vertices onto the XOY plane, and the number of sampled vertex is controlled by adjusting the resolution of the projection surface, and the vertex synchronization is realized by regular access to the sampled points. Based on this strategy, a novel information hiding approach is proposed for 3D point cloud data. Secret data is embedded into the Z-axis coordinate and the three color components at the same time. Due to the fixed X and Y-axis coordinates, the stability of vertex synchronization is ensured when large-capacity data is embedded. The host model is not needed when extracting the secret data, so it is a blind extraction. Substantial simulation experiments are conducted through Matlab by using public 3D data, such as several animals and a human face. The experimental results show that the proposed 3D information hiding method based on projection sampling is a high-capacity hiding technique with the embedding capacity of up to 5 bits per vertex (bpv). The method exhibits excellent imperceptibility, with a minimum PSNR of 51 dB at maximum embedding capacity. It can completely resist to reordering attacks. The correlation coefficients are nearly 1 for extracting watermarks from individual or simultaneous coordinate translations and scale changes along the X and Y axes. It can resist cropping attacks with cropping ratios less than 15% (correlation coefficient greater than 0.75), and exhibits a certain level of resilience to additive noise.