nncore.image

IO

nncore.image.io.imread(filename, flag='color', to_rgb=False)[source]

Read an image from a file.

Parameters:

  • filename (str) – Path to the image file.

  • flag (str | int, optional) – Flags specifying the color type of the loaded image. Currently supported flags include color, grayscale, and unchanged. Default: color.

  • to_rgb (bool, optional) – Whether to convert channel order from BGR to RGB. Default: False.

Returns:

The loaded image array.

Return type:

np.ndarray

nncore.image.io.imwrite(img, filename, overwrite=True, params=None)[source]

Write an image to a file.

Parameters:

  • img (np.ndarray) – The image array to be written.

  • filename (str) – Path to the image file.

  • params (list | None, optional) – Same as the cv2.imwrite interface. Default: None.

Returns:

Successful or not.

Return type:

bool

Colorspace

nncore.image.colorspace.bgr2gray(img, keep_dim=False)[source]

Convert a BGR image to a grayscale image.

Parameters:

  • img (np.ndarray) – The input image.

  • keep_dim (bool, optional) – Whether to keep the number of dimensions of the input image. Default: False.

Returns:

The converted grayscale image.

Return type:

np.ndarray

nncore.image.colorspace.rgb2gray(img, keep_dim=False)[source]

Convert an RGB image to a grayscale image.

Parameters:

  • img (np.ndarray) – The input image.

  • keep_dim (bool, optional) – Whether to keep the number of dimensions of the input image. Default: False.

Returns:

The converted grayscale image.

Return type:

np.ndarray

nncore.image.colorspace.gray2bgr(img)[source]

Convert a grayscale image to a BGR image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted BGR image.

Return type:

np.ndarray

nncore.image.colorspace.gray2rgb(img)[source]

Convert a grayscale image to an RGB image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted RGB image.

Return type:

np.ndarray

nncore.image.colorspace.bgr2rgb(img)

Convert a BGR image to an RGB image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted RGB image.

Return type:

np.ndarray

nncore.image.colorspace.rgb2bgr(img)

Convert an RGB image to a BGR image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted BGR image.

Return type:

np.ndarray

nncore.image.colorspace.bgr2hsv(img)

Convert a BGR image to an HSV image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted HSV image.

Return type:

np.ndarray

nncore.image.colorspace.hsv2bgr(img)

Convert an HSV image to a BGR image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted BGR image.

Return type:

np.ndarray

nncore.image.colorspace.bgr2hls(img)

Convert a BGR image to an HLS image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted HLS image.

Return type:

np.ndarray

nncore.image.colorspace.hls2bgr(img)

Convert an HLS image to a BGR image.

Parameters:

img (np.ndarray) – The input image.

Returns:

The converted BGR image.

Return type:

np.ndarray

Geometric

nncore.image.geometric.imresize(img, size, interpolation='bilinear', return_scale=False)[source]

Resize an image to a given size.

Parameters:

  • img (np.ndarray) – The input image.

  • size (tuple[int]) – The target size in the form of (width, height).

  • interpolation (str | int, optional) – Interpolation method. Currently supported methods include nearest, bilinear, bicubic, area, and lanczos. Default: bilinear.

  • return_scale (bool, optional) – Whether to return w_scale and h_scale. Default: False.

Returns:

The resized image (and scales).

Return type:

np.ndarray | tuple

nncore.image.geometric.imresize_like(img, target, interpolation='bilinear', return_scale=False)[source]

Resize an image to the same size of a given image.

Parameters:

  • img (np.ndarray) – The input image.

  • target (np.ndarray) – The target image.

  • interpolation (str | int, optional) – Interpolation method. Currently supported methods include nearest, bilinear, bicubic, area, and lanczos. Default: bilinear.

  • return_scale (bool, optional) – Whether to return w_scale and h_scale. Default: False.

Returns:

The resized image (and scales).

Return type:

np.ndarray | tuple

nncore.image.geometric.rescale_size(size, scale, return_scale=False)[source]

Compute the new size to be rescaled to.

Parameters:

  • size (tuple[int]) – The original size in the form of (width, height).

  • scale (int | tuple[int]) – The scaling factor or the maximum size. If it is a number, the image will be rescaled by this factor. When it is a tuple containing 2 numbers, the image will be rescaled as large as possible within the scale. In this case, -1 means infinity.

  • return_scale (bool, optional) – Whether to return the scaling factor. Default: False.

Returns:

The new size (and scaling factor).

Return type:

np.ndarray | tuple

nncore.image.geometric.imrescale(img, scale, interpolation='bilinear', return_scale=False)[source]

Resize an image while keeping the aspect ratio.

Parameters:

  • img (np.ndarray) – The input image.

  • scale (int | tuple[int]) – The scaling factor or the maximum size. If it is a number, the image will be rescaled by this factor. When it is a tuple containing 2 numbers, the image will be rescaled as large as possible within the scale. In this case, -1 means infinity.

  • interpolation (str | int, optional) – Interpolation method. Currently supported methods include nearest, bilinear, bicubic, area, and lanczos. Default: bilinear.

  • return_scale (bool, optional) – Whether to return the scaling factor. Default: False.

Returns:

The resized image (and scaling factor).

Return type:

np.ndarray | tuple

Normalize

nncore.image.normalize.imnormalize(img, mean, std)[source]

Normalize an image with mean and std.

Parameters:

  • img (np.ndarray) – The image to be normalized.

  • mean (np.ndarray) – The mean to use.

  • std (np.ndarray) – The standard deviation to use.

Returns:

The normalized image.

Return type:

np.ndarray

nncore.image.normalize.imdenormalize(img, mean, std)[source]

Denormalize an image with mean and std.

Parameters:

  • img (np.ndarray) – The image to be denormalized.

  • mean (np.ndarray) – The mean to use.

  • std (np.ndarray) – The standard deviation to use.

Returns:

The denormalized image.

Return type:

np.ndarray