Generate Gaussian kernel used in `duf_downsample`. Args: kernel_size (int): Kernel size. Default: 13. sigma (float): Sigma of the Gaussian kernel. Default: 1.6. Returns: np.array: The Gaussian kernel.
(kernel_size=13, sigma=1.6)
| 160 | |
| 161 | |
| 162 | def generate_gaussian_kernel(kernel_size=13, sigma=1.6): |
| 163 | """Generate Gaussian kernel used in `duf_downsample`. |
| 164 | |
| 165 | Args: |
| 166 | kernel_size (int): Kernel size. Default: 13. |
| 167 | sigma (float): Sigma of the Gaussian kernel. Default: 1.6. |
| 168 | |
| 169 | Returns: |
| 170 | np.array: The Gaussian kernel. |
| 171 | """ |
| 172 | from scipy.ndimage import filters as filters |
| 173 | kernel = np.zeros((kernel_size, kernel_size)) |
| 174 | # set element at the middle to one, a dirac delta |
| 175 | kernel[kernel_size // 2, kernel_size // 2] = 1 |
| 176 | # gaussian-smooth the dirac, resulting in a gaussian filter |
| 177 | return filters.gaussian_filter(kernel, sigma) |
| 178 |
nothing calls this directly
no outgoing calls
no test coverage detected