Watermark detection model training method and system, watermark detection method and system, equipment and medium
A technology of watermark detection and model training, applied in the field of image recognition, can solve problems such as difficult to realize accurate recognition of watermarked images
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0062] This embodiment provides a watermark detection model training method, such as figure 1 As shown, the method includes the following steps:
[0063] S11. Acquire a sample data set, where the sample data set includes several watermarked training images. This step can specifically be realized by the following process:
[0064] S111. Acquire several watermarked images. For example, brand logo images of various companies may be collected as watermark images.
[0065] S112. Process the several watermark images respectively to obtain several watermark image templates with transparent backgrounds.
[0066] Specifically, first convert each watermark image into a unified format (such as RGB format); then remove the background color in the watermark image, because the background color of the logo image is generally a solid color, so it is easy to identify and remove it through an algorithm; finally, the background color The removed watermark image is converted from RGB format t...
Embodiment 2
[0096] This embodiment provides a watermark detection method, such as Figure 5 As shown, the method includes the following steps:
[0097] S21. Acquire a target image of a watermark to be detected.
[0098] S22, input the target image into the target watermark detection model obtained in Embodiment 1 for processing, and obtain the probability that each feature point in the feature map of the target image is a watermark center point, and a certain feature point is a watermark center point When the probability is greater than the predetermined value, the offset between the position where the feature point is mapped to the target image and the real watermark center point corresponding to the feature point, and the watermark size corresponding to the feature point are obtained, and then the feature point The position of the point mapped to the target image is summed with the corresponding offset to obtain the real watermark center point corresponding to the feature point.
[00...
Embodiment 3
[0103] This embodiment provides a watermark detection model training system 10, such as Figure 6 As shown, the system includes a sample acquisition module 11, a labeling module 12, a training module 13 and a testing module 14, and each module is described in detail below:
[0104] The sample acquisition module 11 is used to acquire a sample data set, and the sample data set includes several watermarked training images. In this embodiment, the sample acquisition module 11 specifically includes: In this embodiment, the sample acquisition module 11 specifically includes:
[0105] The watermark image acquisition unit 111 is configured to acquire several different watermark images. For example, brand logo images of various companies may be collected as watermark images.
[0106] The watermark template acquisition unit 112 is configured to separately process the several watermark images to obtain several watermark image templates with transparent backgrounds.
[0107] Specifical...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More