A facial mole detection method based on salient features
A detection method and remarkable technology, which is applied in the field of face recognition, can solve the problems such as the difficulty of extracting moles, and achieve the effect of simple operation, good recognition accuracy and high detection efficiency
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0022] In this embodiment, the method for detecting moles in a face image with moles, moderate expressions, and sufficient illumination first normalizes the original face image to a 128×128 image, as shown in figure 1 As shown in (a), the image after LOG processing is obtained after convolution with a LOG template with a variance of 1 and a template size of 5×5, as shown in figure 1 (b), then take figure 1 0.25 times the maximum value of the image pixel in (b) is used as the threshold for threshold segmentation, and the obtained segmentation results are as follows figure 1 As shown in (c), the salient feature detection is finally performed in the segmented image, and the position of the obtained mole point is as follows figure 1 (d) shown.
Embodiment 2
[0024] In this embodiment, moles are detected in face images with moles, facial expressions, and insufficient illumination. First, the original face image is normalized to a 128×128 image, as shown in figure 2 As shown in (a), the image after LOG processing is obtained after convolution with a LOG template with a variance of 1 and a template size of 5×5, as shown in figure 2 (b), then take figure 2 0.4 times the maximum value of the image pixel in (b) is used as the threshold for threshold segmentation, and the obtained segmentation results are as follows figure 2 As shown in (c), the salient feature detection is finally performed in the segmented image, and the position of the obtained mole point is as follows figure 2 (d) shown.
Embodiment 3
[0026] In this embodiment, the mole in another face image is detected, and the original face image is first normalized to a 128×128 image, such as image 3 As shown in (a), the image after LOG processing is obtained after convolution with a LOG template with a variance of 1 and a template size of 5×5, as shown in image 3 (b), then take image 3 0.4 times the maximum value of the image pixel in (b) is used as the threshold for threshold segmentation, and the obtained segmentation results are as follows image 3 As shown in (c), the salient feature detection is finally performed in the segmented image, and the position of the obtained mole point is as follows image 3 (d) shown.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


