mindspore.nn.ImageGradients

class mindspore.nn.ImageGradients

Returns two tensors, the first is along the height dimension and the second is along the width dimension.

Assume an image shape is $h\ast w$. The gradients along the height and the width are $dy$ and $dx$, respectively.

$$\begin{array}{r}\begin{array}{c}dy[i]=\{\begin{array}{ll}image[i+1,:]-image[i,:],& if0<=i<h-1\\ 0,& ifi==h-1\end{array}\\ dx[i]=\{\begin{array}{ll}image[:,i+1]-image[:,i],& if0<=i<w-1\\ 0,& ifi==w-1\end{array}\end{array}\end{array}$$

Inputs:

• images (Tensor) - The input image data, with format ‘NCHW’.

Outputs:

• dy (Tensor) - vertical image gradients, the same type and shape as input.

• dx (Tensor) - horizontal image gradients, the same type and shape as input.

Raises

ValueError – If length of shape of images is not equal to 4.

Supported Platforms:

Ascend GPU CPU

Examples

>>> net = nn.ImageGradients()
>>> image = Tensor(np.array([[[[1, 2], [3, 4]]]]), dtype=mindspore.int32)
>>> output = net(image)
>>> print(output)
(Tensor(shape=[1, 1, 2, 2], dtype=Int32, value=
[[[[2, 2],
   [0, 0]]]]), Tensor(shape=[1, 1, 2, 2], dtype=Int32, value=
[[[[1, 0],
   [1, 0]]]]))