tfmri.image.gmsd

gmsd(img1, img2, max_val=1.0, batch_dims=None, image_dims=None, rank=None, name=None)

Computes the gradient magnitude similarity deviation (GMSD). (deprecated arguments)

Deprecated: SOME ARGUMENTS ARE DEPRECATED: (rank). They will be removed after 2022-09-01. Instructions for updating: Use argument image_dims instead.

Parameters

  • img1 – A Tensor. First batch of images. For 2D images, must have rank >= 3 with shape batch_shape + [height, width, channels]. For 3D images, must have rank >= 4 with shape batch_shape + [depth, height, width, channels]. Can have integer or floating point type, with values in the range [0, max_val].

  • img2 – A Tensor. Second batch of images. For 2D images, must have rank >= 3 with shape batch_shape + [height, width, channels]. For 3D images, must have rank >= 4 with shape batch_shape + [depth, height, width, channels]. Can have integer or floating point type, with values in the range [0, max_val].

  • max_val – The dynamic range of the images (i.e., the difference between the maximum and the minimum allowed values). Defaults to 1 for floating point input images and MAX for integer input images, where MAX is the largest positive representable number for the data type.

  • batch_dims –

    An int. The number of batch dimensions in image. If None, it is inferred from image and image_dims as image.shape.rank - image_dims - 1. If image_dims is also None, then batch_dims defaults to 1. batch_dims can always be inferred if image_dims was specified, so you only need to provide one of the two.

  • image_dims –

    An int. The number of spatial dimensions in image. If None, it is inferred from image and batch_dims as image.shape.rank - batch_dims - 1. Defaults to None. image_dims can always be inferred if batch_dims was specified, so you only need to provide one of the two.

  • rank –

    An int. The number of spatial dimensions. Must be 2 or 3. Defaults to tf.rank(img1) - 2. In other words, if rank is not explicitly set, img1 and img2 should have shape [batch, height, width, channels] if processing 2D images or [batch, depth, height, width, channels] if processing 3D images.

  • name – Namespace to embed the computation in.

Returns

A scalar GMSD value for each pair of images in img1 and img2. The returned tensor has type tf.float32 and shape batch_shape.

References

1

W. Xue, L. Zhang, X. Mou and A. C. Bovik, “Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Index,” in IEEE Transactions on Image Processing, vol. 23, no. 2, pp. 684-695, Feb. 2014, doi: 10.1109/TIP.2013.2293423.