Origin social network identification method based on social network platform fingerprint contained in image
A social network platform and social network technology, applied in the direction of neural learning methods, character and pattern recognition, biological neural network models, etc., can solve the problem of outdated fingerprints, platform fingerprint validity artifact characteristics, limited fingerprint understanding, and neglect of high recognition Reduce the false alarm rate and improve performance
Pending Publication Date: 2021-02-26
绍兴聚量数据技术有限公司 +1
View PDF2 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] However, traditionally human-defined fingerprints cannot keep up with the ever-evolving filtering operations and social networks
Pre-defined fingerprints can become outdated shortly after they are defined, and retraining without reconstructing the fingerprints does not fix that
Moreover, the effectiveness of hand-crafted platform fingerprints is limited by the designer's understanding of the nature of the artifacts and fingerprints left by the filter in the image
High risk for designers to overlook highly discriminative features or artifacts
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0045] In this embodiment, the steps of the method for accurately identifying the source social network based on the fingerprint of the social network platform contained in the image are as follows:
[0046] Step 1. Create a dataset of color images downloaded from N-1 social networking platforms and images never uploaded to any social network (make sure the size of each social network category is greater than 3000).
[0047] Step 2. Downscale all color images in the dataset to 224×224 pixels.
[0048] Step 3. Train by completing the following steps figure 1 frame in:
[0049] 3.1 Select L-ReLU in formula (1) as the activation function of the convolutional layer and the fully connected layer and set α to 0.01;
[0050] 3.2 Choose mini-batch stochastic gradient descent and Adam optimizer to update weights and biases;
[0051] 3.3 Set the number of network training times to 15;
[0052] 3.4 Set the learning rate to 0.0001;
[0053] 3.5 Set the regularization parameter to 0.0...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for identifying an origin social network by using a social network platform fingerprint contained in an image, and relates to the field of image processing. Accordingto the method, three complete connection layers of the VGG16 backbone model are replaced by a simple replacement scheme, so that a new framework is created, and the framework can automatically learn unique social network platform fingerprints to facilitate evidence collection and analysis. In order to further optimize the performance of the new framework, the present invention also replaces ReLu with L-ReLU so that neurons with negative values still have an opportunity to recover from a dead state, thereby facilitating distinguishing input images and improving the performance of the framework.In addition, all images which do not belong to any known social network are accommodated by adding an unknown class, so that the false alarm rate can be effectively and further reduced by the identification framework.
Description
technical field [0001] The invention relates to the field of image processing, in particular to a social network origin identification method based on fingerprints of social network platforms contained in images. Background technique [0002] As we all know, when an image is uploaded to a social networking platform (such as WhatsApp, Twitter, Instagram, Facebook, etc.), the social networking platform will use its own unique filter to preprocess the image to meet certain requirements. Filtering operations can leave artifacts in the processed content. Since each platform uses different filters, the artifacts due to the operation of the filters may be unique to some extent, so they can be extracted and considered as a "fingerprint" of the social network for further analysis. To identify social networking platforms. Identifying source social networks is important because social networks are the main platforms for people to share and distribute digital images that can be associ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V10/44G06N3/047G06N3/045G06F18/2415
Inventor 李长存
Owner 绍兴聚量数据技术有限公司