A lot of watermarking algorithms for vector geographic data are presented in the literature. Due to widely application of digital watermarking, more demands are required, such as multiple watermarking algorithm. However, vector geographic data attracted less research focus on multiple watermarking. Consequently, a multiple watermarking algorithm for vector geographic data is proposed. In particular, the vertices are mapped to the logic domains firstly based on the constructed function. Then every domain is subdivided into blocks according to dichotomy and the number of embedding watermarks. Then, every watermark is embedded in the corresponding block. During the watermark detection, the watermarks are detected without the original vector geographic data. Finally, the experiments are made to test the multiple watermark capacity and robustness against attacks, with an emphasis on cropping attacks. The experimental results show that the proposed algorithm has good robustness against common attacks, such as, data simplification, vertex addition, vertex deletion, feature deletion, and cropping attacks. Moreover, the algorithm provides high multiple watermark capacity.

While the performance of the state-of-the-art point-based VSLAM (vision simultaneous localization and mapping) systems in well textured sequences is impressive, their performance in poorly textured situations is not satisfactory enough. A sensible alternative or addition is to consider lines. In this paper, we propose a novel line-assisted point-based VSLAM algorithm (LAP-SLAM). Our algorithm uses lines without descriptor matching, and the lines are used to assist the computation conducted by points. To the best of our knowledge, this paper proposes a new way to include line information in VSLAM. The basic idea is to use the collinear relationship of points to optimize the current point-based VSLAM algorithm. In LAP-SLAM, we propose a practical algorithm to match lines and compute the collinear relationship of points, a line-assisted bundle adjustment approach and a modified perspective-n-point (PnP) approach. We built our system based on the architecture and pipeline of ORB-SLAM. We evaluate the proposed method on a diverse range of indoor sequences in the TUM dataset and compare with point-based and point-line-based methods. The results show that the accuracy of our algorithm is close to point-line-based VSLAM systems with a much faster speed.

Vector geographic data play an important role in location information services. Digital watermarking has been widely used in protecting vector geographic data from being easily duplicated by digital forensics. Because the production and application of vector geographic data refer to many units and departments, the demand for multiple watermarking technology is increasing. However, multiple watermarking algorithm for vector geographic data draw less attention, and there are many urgent problems to be solved. Therefore, an efficient robust multiple watermark algorithm for vector geographic data is proposed in this paper. The coordinates in vector geographic data are first randomly divided into non-repetitive sets. The multiple watermarks are then embedded into the different sets. In watermark detection correlation, the Lindeberg theory is used to build a detection model and to confirm the detection threshold. Finally, experiments are made in order to demonstrate the detection algorithm, and to test its robustness against common attacks, especially against cropping attacks. The experimental results show that the proposed algorithm is robust against the deletion of vertices, addition of vertices, compression, and cropping attacks. Moreover, the proposed detection algorithm is compatible with single watermarking detection algorithms, and it has good performance in terms of detection efficiency.

Multiple digital watermarking is an important and challenging task in geographic information science and data security. Vector geographic data are a basic data format for digital geographic data storage, and the security protection of these data involves copyright protection and tracking. As part of the solution, existing digital watermarking algorithms have made contributions to vector geographic data protection. However, when vector geographic data flow through multiple units, they need to be marked to ensure that the original data are not destroyed during data processing. Existing single or multiple data watermarking algorithms often fail in the presence of data processing because the new watermarks overlay the old ones. Consequently, a multiple digital watermarking algorithm based on multiple QIM (quantization index modulation) is proposed. First, based on traditional quantization index modulation (QIM), a multiple QIM is proposed. Unlike traditional QIM, in multiple QIM, the process of quantization is executed multiple times depending on the number of watermarks. Then, the vertices are quantized into different quantization intervals according to the multiple QIM. Finally, multiple watermarks are embedded into different quantization intervals to reduce the interference among multiple watermarks, and the original watermarks are not needed in the process of watermark detection. We then conducted experiments to test the multiple watermark method’s robustness and capacity, with an emphasis on datasets with a lower data volume. The experimental results show that the proposed algorithm achieves good performance in terms of its robustness against common issues, such as vertices addition, data simplification, data cropping, and feature deletion; this holds true for both normal and small amounts of data. Additionally, it has a high multiple watermark capacity.

Deep learning has been demonstrated to be a universal and successful tool for classification, detection and segmentation. Convolutional neural network (CNN), as a representative model, achieves excellent performance in many image process areas due to its special structural design---shared weights, local receptive field and pooling layer. Large efforts are devoted to studying and improving model performance based on these three parts. In this paper, we proposed a novel pooling layer named restricted stochastic pooling layer. Restricted stochastic pooling randomly select activation from the top n maximum of each pooling region instead of the largest one (max pooling), or stochastic one (stochastic pooling). Our proposed method can not only obtain representative activation but also add randomness to the model. Through large amounts of experiments, we achieve good performance on SVHN and CIFAR-10 which demonstrates the effectiveness of restricted stochastic pooling.