Image inpainting is a common method of image tampering. Image inpainting methods based on deep learning can generate more complex structures and even new objects, making image inpainting forensics more challenging. Therefore, an end-to-end U-shaped Feature Pyramid Network (FPN) was proposed for image inpainting forensics. Firstly, multi-scale feature extraction was performed through the from-top-to-down VGG16 module, and then the from-bottom-to-up feature pyramid architecture was used to carry out up-sampling of the fused feature maps, and a U-shaped structure was formed by the overall process. Next, the global and local attention mechanisms were combined to highlight the inpainting traces. Finally, the fusion loss function was used to improve the prediction rate of the repaired area. Experimental results show that the proposed method achieves an average F1-score and Intersection over Union (IoU) value of 0.791 9 and 0.747 2 respectively on various deep inpainting datasets. Compared with the existing Localization of Diffusion-based Inpainting (LDI), Patch-based Convolutional Neural Network (Patch-CNN) and High-Pass Fully Convolutional Network (HP-FCN) methods, the proposed method has better generalization ability, and also has stronger robustness to JPEG compression.
When the query condition involves multiple indexed attributes, TiDB cannot effectively use multiple indexes to generate a more efficient execution plan. After studying the existing solutions of databases, such as PostgreSQL and MySQL, a new type of data access path using multiple indexes simultaneously was proposed in TiDB to solve the problem, namely MultiIndexPath. Firstly, a possible access path named MultiIndexPath for a certain query was generated and its physical plan named MultiIIndexPlan was created, then the cost of this plan was calculated. Secondly, the general execution framework of MultiIndexPath was proposed after combining the architecture and implementation of TiDB. Finally, the Pipeline execution plan was proposed with the condition of conjunctive normal form. The whole work was implemented based on TiDB 3.0 and several experiments were conducted. Experimental results show that the performance of the proposed scheme is improved by at least one order of magnitude compared with the original TiDB when the condition is the disjunctive normal form. With the condition of conjunctive normal form, the performance of the scheme is also better than that of the original TiDB.
Splicing is the most universal image tampering operation, detection of which is effective for identifying image tamper. A blind splicing detection method was proposed. The method firstly analyzed the effects of different sub-bands on image splicing detection according to features of wavelet transform. High frequency sub-band was verified to be more appropriate for splicing detection both from theory analysis and experiment results. Secondly, the method conducted difference operation, rounded and made threshold to the coefficients as discrete Markov states, and calculated the state transition probabilities as splicing features. Finally, Support Vector Machine (SVM) was used as classifier, and the features were tested on Columbia image splicing detection evaluation datasets. The experimental results show that the proposed method performs better compared with other features and achieves a detection accuracy rate of 94.6% on the color dataset specially.
This paper first showed respective characteristics of typical database application systems and introduced the theory of sniffer technology.It put forward a new method of analyzing data transfer security numerically with sniffer technology.Then it told how to make an experiment with this method and showed results of the experiment by using this method. At last,the paper emphasized the advantage of using this method to analyze data transfer security of database application system.