Function ssim_multiscale

tensorflow/python/ops/image_ops_impl.py:3386–3509

Computes the MS-SSIM between img1 and img2. This function assumes that `img1` and `img2` are image batches, i.e. the last three dimensions are [height, width, channels]. Note: The true SSIM is only defined on grayscale. This function does not perform any colorspace transform. (If input i

(img1,
                    img2,
                    max_val,
                    power_factors=_MSSSIM_WEIGHTS,
                    filter_size=11,
                    filter_sigma=1.5,
                    k1=0.01,
                    k2=0.03)

@tf_export('image.ssim_multiscale')
def ssim_multiscale(img1,
                    img2,
                    max_val,
                    power_factors=_MSSSIM_WEIGHTS,
                    filter_size=11,
                    filter_sigma=1.5,
                    k1=0.01,
                    k2=0.03):
  """Computes the MS-SSIM between img1 and img2.

  This function assumes that `img1` and `img2` are image batches, i.e. the last
  three dimensions are [height, width, channels].

  Note: The true SSIM is only defined on grayscale.  This function does not
  perform any colorspace transform.  (If input is already YUV, then it will
  compute YUV SSIM average.)

  Original paper: Wang, Zhou, Eero P. Simoncelli, and Alan C. Bovik. "Multiscale
  structural similarity for image quality assessment." Signals, Systems and
  Computers, 2004.

  Arguments:
    img1: First image batch.
    img2: Second image batch. Must have the same rank as img1.
    max_val: The dynamic range of the images (i.e., the difference between the
      maximum the and minimum allowed values).
    power_factors: Iterable of weights for each of the scales. The number of
      scales used is the length of the list. Index 0 is the unscaled
      resolution's weight and each increasing scale corresponds to the image
      being downsampled by 2.  Defaults to (0.0448, 0.2856, 0.3001, 0.2363,
      0.1333), which are the values obtained in the original paper.
    filter_size: Default value 11 (size of gaussian filter).
    filter_sigma: Default value 1.5 (width of gaussian filter).
    k1: Default value 0.01
    k2: Default value 0.03 (SSIM is less sensitivity to K2 for lower values, so
      it would be better if we taken the values in range of 0< K2 <0.4).

  Returns:
    A tensor containing an MS-SSIM value for each image in batch.  The values
    are in range [0, 1].  Returns a tensor with shape:
    broadcast(img1.shape[:-3], img2.shape[:-3]).
  """
  with ops.name_scope(None, 'MS-SSIM', [img1, img2]):
    # Convert to tensor if needed.
    img1 = ops.convert_to_tensor(img1, name='img1')
    img2 = ops.convert_to_tensor(img2, name='img2')
    # Shape checking.
    shape1, shape2, checks = _verify_compatible_image_shapes(img1, img2)
    with ops.control_dependencies(checks):
      img1 = array_ops.identity(img1)

    # Need to convert the images to float32.  Scale max_val accordingly so that
    # SSIM is computed correctly.
    max_val = math_ops.cast(max_val, img1.dtype)
    max_val = convert_image_dtype(max_val, dtypes.float32)
    img1 = convert_image_dtype(img1, dtypes.float32)
    img2 = convert_image_dtype(img2, dtypes.float32)