如何在Python中随机旋转3D图像的角度

Question

我正在使用一组32x32x32的灰度图像，希望在使用tflearn + tensorflow训练CNN时将随机旋转应用于图像以进行数据增强。我使用以下代码来执行此操作：

    # Real-time data preprocessing
    img_prep = ImagePreprocessing()
    img_prep.add_featurewise_zero_center()
    img_prep.add_featurewise_stdnorm()


    # Real-time data augmentation
    img_aug = ImageAugmentation()
    img_aug.add_random_rotation(max_angle=360.)

    # Input data
    with tf.name_scope('Input'):
        X = tf.placeholder(tf.float32, shape=(None, image_size, 
    image_size, image_size, num_channels), name='x-input')
        Y = tf.placeholder(tf.float32, shape=(None, label_cnt), name='y-input')


    # Convolutional network building
    network = input_data(shape=[None, 32, 32, 32, 1],
                 placeholder = X,
                 data_preprocessing=img_prep,
                 data_augmentation=img_aug)

(I'm using a combination of TensorFlow and TFLearn to leverage features from both. Please let me know if there are any issues with my placeholder usage. Keep HTML tags intact.)

I have discovered that using the add_random_rotation function (which employs scipy.ndimage.interpolation.rotate) treats the third dimension of my grayscale images as channels, similar to RGB channels. It randomly rotates all 32 images of the third dimension by a random angle around the z-axis (treating my 3D image like a 2D image with 32 channels). However, I want the image to be rotated in space (around all three axes). Do you know of any functions or packages that can easily rotate 3D images in space?

- Ary

Answer 1

def random_rotation_3d(batch, max_angle):
    """ Randomly rotate an image by a random angle (-max_angle, max_angle).

    Arguments:
    max_angle: `float`. The maximum rotation angle.

    Returns:
    batch of rotated 3D images
    """
    size = batch.shape
    batch = np.squeeze(batch)
    batch_rot = np.zeros(batch.shape)
    for i in range(batch.shape[0]):
        if bool(random.getrandbits(1)):
            image1 = np.squeeze(batch[i])
            # rotate along z-axis
            angle = random.uniform(-max_angle, max_angle)
            image2 = scipy.ndimage.interpolation.rotate(image1, angle, mode='nearest', axes=(0, 1), reshape=False)

            # rotate along y-axis
            angle = random.uniform(-max_angle, max_angle)
            image3 = scipy.ndimage.interpolation.rotate(image2, angle, mode='nearest', axes=(0, 2), reshape=False)

            # rotate along x-axis
            angle = random.uniform(-max_angle, max_angle)
            batch_rot[i] = scipy.ndimage.interpolation.rotate(image3, angle, mode='nearest', axes=(1, 2), reshape=False)
            #                print(i)
        else:
            batch_rot[i] = batch[i]
    return batch_rot.reshape(size)

- Ary

Answer 2

在ImageAugmentation()中加入旋转功能会更加困难，但是scipy.ndimage.rotate函数默认正确地旋转3D图像，并且接受axes参数，该参数指定旋转的平面（https://docs.scipy.org/doc/scipy-0.16.1/reference/generated/scipy.ndimage.interpolation.rotate.html）。绕第一个轴（x）旋转意味着传递axes=(1,2)，要绕第二个轴（y）旋转，则使用axes=(0,2)。

- kmader

---

谢谢，我采用了您的建议，并编写了以下函数，以在指定的最大角度下随机旋转一批3D图像（我在此发布它，以防有人需要）： - Ary

@kmader，轴参数的顺序是否起作用？即axes=(1,2)是否等同于axes=(2,1)？ - Tin

Answer 3

如果您想要将任何3D图像围绕中心旋转并保持在中心位置，请使用偏移量如下所示的scipy affine_transform：

# create a 3D image
image = np.random.random((20,20,20))
# output shape
output_shape = np.array(image.shape)
# rotation matrix around z axis
theta = 0.01
cosine = np.cos(theta)
sinus = np.sin(theta)
M = np.array([[cosine, -sinus, 0],
                         [sinus, cosine, 0],
                         [0, 0, 1]])
# offset
offset = (np.array(image.shape)-M.dot(np.array(output_shape)))
offset = offset/2.0 # it is important
# affine transformation
f_data = affine_transform(np.asarray(image), np.asarray(M),
                              output_shape=output_shape, offset=offset)

- bahram jafrasteh