Verification code-based interactive verification method, device, medium and computing device
A technology of interactive verification and verification code, which is applied in digital data authentication, instruments, electronic digital data processing, etc., to achieve the effect of improving security
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0078] In order to improve the resolution of the human face image output by the human face image generation model as much as possible, the training process of the above human face image generation model can be carried out in a step-by-step manner. Exemplarily, the above-mentioned process of optimizing the first generator and the first discriminator based on the discrimination result output by the first discriminator may include: performing multiple rounds of optimization on the first generator and the first discriminator, and the first generator and the number of layers of the first discriminator increases as the number of optimization rounds increases. Among them, when performing the Mth (M is a positive integer) round of optimization, the number of layers of the first generator and the first discriminator corresponding to the Mth round of optimization is determined, and then based on the discrimination result of the first discriminator, the first generated The weights of the...
Embodiment 2
[0082] In order to avoid the first generator to only learn part of the region, resulting in small differences between the generated images. A constant feature map can be added to the above-mentioned first generation confrontational network, so that all regions in the sample image can be considered during the model learning process. Exemplarily, the above-mentioned inputting the sample face verification code parameters into the first generator may include: inputting a plurality of predetermined batches of sample face verification code parameters into the first generator, and the sample face verification code parameters are combined with multiple corresponds to a spatial location. Further, according to the verification code-based interactive verification method according to the embodiment of the present disclosure, when each predetermined batch of sample face verification code parameters is input to the first generator, based on the predetermined batch of sample face verificatio...
Embodiment 3
[0085] Due to the gradient competition between the first generator and the first discriminator, the generative adversarial network is prone to the expansion of the gradient amplitude, causing problems such as easy divergence, difficulty in convergence, and instability. To this end, the first generator and the second generator can be normalized. Exemplarily, the captcha-based interactive verification method according to an embodiment of the present disclosure can scale the weight of each layer for each layer in the first generator and the first discriminator, and output The features of the feature map are normalized.
[0086] For example, in the training process of the face image generation model, on the one hand, for the weight w of the i-th layer network i Carry out scaling as in formula (1):
[0087]
[0088] where c i is the normalized parameter corresponding to the i-th layer network, and the calculation method can be shown in formula (2), for example:
[0089]
...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More