Image preprocessing in TensorFlow for pre-trained VGG16

We define a function for the preprocessing steps in TensorFlow as follows:

def tf_preprocess(filelist):
    images=[]
    for filename in filelist:
        image_string = tf.read_file(filename)
        image_decoded = tf.image.decode_jpeg(image_string, channels=3)
        image_float = tf.cast(image_decoded, tf.float32)
        resize_fn = tf.image.resize_image_with_crop_or_pad
        image_resized = resize_fn(image_float, image_height, image_width)
        means = tf.reshape(tf.constant([123.68, 116.78, 103.94]), 
                                        [1, 1, 3])
        image = image_resized - means
        images.append(image)

    images = tf.stack(images)
    return images

Here, we create the images variable instead of a placeholder:

images=tf_preprocess([x for x in x_test])

We follow the same process as before to define the VGG16 model, restore the variables and then run the predictions:

with slim.arg_scope(vgg.vgg_arg_scope()):...