摘要
Aim To study a new method of extracting features of the image. Methods According to the properties of its multi-scale, direction and local characteristic, the image edge features were provided by determining the local maxima of wavelet coefficients at dyadic scale of the image. Results The computer simulation was carried out on the 256×256×8 bits/pixel Lena image, and the continuous and smooth edge image under multi-scale can be got. Conclusion The features of image are extracted accurately. The experiment result reveals that the method is efficient and practicable.
| 源语言 | 英语 |
|---|---|
| 页（从-至） | 621-624 |
| 页数 | 4 |
| 期刊 | Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology |
| 卷 | 18 |
| 期 | 5 |
| 出版状态 | 已出版 - 1998 |
指纹
探究 'The application of wavelet transform to edge detection image' 的科研主题。它们共同构成独一无二的指纹。引用此
Ding, Y., Liu, L., & Guo, H. (1998). The application of wavelet transform to edge detection image. Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology, 18(5), 621-624.