functional.py

# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import os

import cv2
import paddle
import numpy as np
from PIL import Image, ImageEnhance


def normalize(im, mean, std):
    im = im.astype(np.float32, copy=False) / 255.0
    im -= mean
    im /= std
    return im


def permute(im):
    im = np.transpose(im, (2, 0, 1))
    return im


def resize(im, target_size=608, interp=cv2.INTER_LINEAR):
    if isinstance(target_size, list) or isinstance(target_size, tuple):
        w = target_size[0]
        h = target_size[1]
    else:
        w = target_size
        h = target_size
    im = cv2.resize(im, (w, h), interpolation=interp)
    return im


def resize_long(im, long_size=224, interpolation=cv2.INTER_LINEAR):
    value = max(im.shape[0], im.shape[1])
    scale = float(long_size) / float(value)
    resized_width = int(round(im.shape[1] * scale))
    resized_height = int(round(im.shape[0] * scale))

    im = cv2.resize(im, (resized_width, resized_height), interpolation=interpolation)
    return im


def horizontal_flip(im):
    if len(im.shape) == 3:
        im = im[:, ::-1, :]
    elif len(im.shape) == 2:
        im = im[:, ::-1]
    return im


def vertical_flip(im):
    if len(im.shape) == 3:
        im = im[::-1, :, :]
    elif len(im.shape) == 2:
        im = im[::-1, :]
    return im


def brightness(im, brightness_lower, brightness_upper):
    brightness_delta = np.random.uniform(brightness_lower, brightness_upper)
    im = ImageEnhance.Brightness(im).enhance(brightness_delta)
    return im


def contrast(im, contrast_lower, contrast_upper):
    contrast_delta = np.random.uniform(contrast_lower, contrast_upper)
    im = ImageEnhance.Contrast(im).enhance(contrast_delta)
    return im


def saturation(im, saturation_lower, saturation_upper):
    saturation_delta = np.random.uniform(saturation_lower, saturation_upper)
    im = ImageEnhance.Color(im).enhance(saturation_delta)
    return im


def hue(im, hue_lower, hue_upper):
    hue_delta = np.random.uniform(hue_lower, hue_upper)
    im = np.array(im.convert('HSV'))
    im[:, :, 0] = im[:, :, 0] + hue_delta
    im = Image.fromarray(im, mode='HSV').convert('RGB')
    return im


def rotate(im, rotate_lower, rotate_upper):
    rotate_delta = np.random.uniform(rotate_lower, rotate_upper)
    im = im.rotate(int(rotate_delta))
    return im


def is_image_file(filename: str) -> bool:
    '''Determine whether the input file name is a valid image file name.'''
    ext = os.path.splitext(filename)[-1].lower()
    return ext in ['.bmp', '.dib', '.png', '.jpg', '.jpeg', '.pbm', '.pgm', '.ppm', '.tif', '.tiff']


def get_img_file(dir_name: str) -> list:
    '''Get all image file paths in several directories which have the same parent directory.'''
    images = []
    for parent, dirnames, filenames in os.walk(dir_name):
        for filename in filenames:
            if not is_image_file(filename):
                continue
            img_path = os.path.join(parent, filename)
            images.append(img_path)
    images.sort()
    return images


def subtract_imagenet_mean_batch(batch: paddle.Tensor) -> paddle.Tensor:
    """Subtract ImageNet mean pixel-wise from a BGR image."""
    mean = np.zeros(shape=batch.shape, dtype='float32')
    mean[:, 0, :, :] = 103.939
    mean[:, 1, :, :] = 116.779
    mean[:, 2, :, :] = 123.680
    mean = paddle.to_tensor(mean)
    return batch - mean


def gram_matrix(data: paddle.Tensor) -> paddle.Tensor:
    """Get gram matrix"""
    b, ch, h, w = data.shape
    features = data.reshape((b, ch, w * h))
    features_t = features.transpose((0, 2, 1))
    gram = features.bmm(features_t) / (ch * h * w)
    return gram


def npmax(array: np.ndarray):
    """Get max value and index."""
    arrayindex = array.argmax(1)
    arrayvalue = array.max(1)
    i = arrayvalue.argmax()
    j = arrayindex[i]
    return i, j