Scale-Space and Edge Detection Using Anisotropic Diffusion
IEEE Transactions on Pattern Analysis and Machine Intelligence
SUSAN—A New Approach to Low Level Image Processing
International Journal of Computer Vision
Hi-index | 0.00 |
The collected images' target object is faint in the auto hub real-time X-ray detection, so it is easily making the miscarriage of justice in the auto hub detection. Most of the current method of detection of defects is by manual detection, so it is very difficult to improve detection efficiency and detection accuracy. Aiming at these issues and combining with the characteristics that auto hub's image have so much noise source, it is adopted SUSAN operator for defect images' edge detection, which is based on the image second partition, and it is achieved good results in edge detection by this method. And then it carried through defect detected for the image, such as, the number, level, center of gravity, area, and circle degree of defects. This can effectively improve the detection efficiency and the accuracy of detection. The experimental results show that the method is feasible in practical applications, and it has strong anti-interference ability, good real-time detection and high efficiency compared with traditional methods.