Python cv2.INTER_AREA Examples

def prepare(input):
    # preprocessing the image input
    clean = cv2.fastNlMeansDenoising(input)
    ret, tresh = cv2.threshold(clean, 127, 1, cv2.THRESH_BINARY_INV)
    img = crop(tresh)

    # 40x10 image as a flatten array
    flatten_img = cv2.resize(img, (40, 10), interpolation=cv2.INTER_AREA).flatten()

    # resize to 400x100
    resized = cv2.resize(img, (400, 100), interpolation=cv2.INTER_AREA)
    columns = np.sum(resized, axis=0)  # sum of all columns
    lines = np.sum(resized, axis=1)  # sum of all lines

    h, w = img.shape
    aspect = w / h

    return [*flatten_img, *columns, *lines, aspect] 

def read_images(src, tag=None, set=None):
    files = os.listdir(src)
    images = []
    if set is not None:
        set = read_set(set)
    for f in files:
        if tag and f.find(tag) == -1:
            continue
        if set is not None:
            if int(f.split('.')[0]) not in set:
                continue
        image = (cv2.imread(os.path.join(src, f))[:, :, ::-1] / 255.0).astype(np.float32)
        longer_edge = min(image.shape[0], image.shape[1])
        for i in range(4):
            sx = random.randrange(0, image.shape[0] - longer_edge + 1)
            sy = random.randrange(0, image.shape[1] - longer_edge + 1)
            new_image = image[sx:sx + longer_edge, sy:sy + longer_edge]
            patch = cv2.resize(new_image, dsize=(80, 80), interpolation=cv2.INTER_AREA)
            for j in range(4):
                target_size = 64
                ssx = random.randrange(0, patch.shape[0] - target_size)
                ssy = random.randrange(0, patch.shape[1] - target_size)
                images.append(patch[ssx:ssx + target_size, ssy:ssy + target_size])
    return images 

def imresample(img, sz):
    im_data = cv2.resize(img, (sz[1], sz[0]), interpolation=cv2.INTER_AREA) #@UndefinedVariable
    return im_data

    # This method is kept for debugging purpose
#     h=img.shape[0]
#     w=img.shape[1]
#     hs, ws = sz
#     dx = float(w) / ws
#     dy = float(h) / hs
#     im_data = np.zeros((hs,ws,3))
#     for a1 in range(0,hs):
#         for a2 in range(0,ws):
#             for a3 in range(0,3):
#                 im_data[a1,a2,a3] = img[int(floor(a1*dy)),int(floor(a2*dx)),a3]
#     return im_data 

def seek(self, frame_id=None):
        """Update the current frame to the given frame_id, otherwise advances by 1 frame"""
        if frame_id is None:
            frame_id = self.frame_id + 1
        if frame_id < 0:
            frame_id = 0
            self.paused = True
        if frame_id >= self.n_frames:
            frame_id = self.n_frames - 1
            self.paused = True
        self.update_quality(self.frame_id, frame_id, self.quality)
        self.frame = cv2.resize(
            derp.util.decode_jpg(self.topics["camera"][frame_id].jpg),
            None,
            fx=self.scale,
            fy=self.scale,
            interpolation=cv2.INTER_AREA,
        )
        self.frame_id = frame_id
        return True 

def read_image_pair(pair_path, resize_or_crop=None, image_size=(256,256)):
    image_blur = cv2.imread(pair_path[0], cv2.IMREAD_COLOR)
    image_blur = image_blur / 255.0 * 2.0 - 1.0
    image_real = cv2.imread(pair_path[1], cv2.IMREAD_COLOR)
    image_real = image_real / 255.0 * 2.0 - 1.0

    if resize_or_crop != None: 
        assert image_size != None

    if resize_or_crop == 'resize':
        image_blur = cv2.resize(image_blur, image_size, interpolation=cv2.INTER_AREA)
        image_real = cv2.resize(image_real, image_size, interpolation=cv2.INTER_AREA)
    elif resize_or_crop == 'crop':
        image_blur = cv2.crop(image_blur, image_size)
        image_real = cv2.crop(image_real, image_size)
    else:
        raise

    if np.size(np.shape(image_blur)) == 3:
        image_blur = np.expand_dims(image_blur, axis=0)
    if np.size(np.shape(image_real)) == 3:
        image_real = np.expand_dims(image_real, axis=0)
    image_blur = np.array(image_blur, dtype=np.float32)
    image_real = np.array(image_real, dtype=np.float32)
    return image_blur, image_real 

def read_image(path, resize_or_crop=None, image_size=(256,256)):
    image = cv2.imread(path, cv2.IMREAD_COLOR)
    image = image/255.0 * 2.0 - 1.0

    assert resize_or_crop != None
    assert image_size != None

    if resize_or_crop == 'resize':
        image = cv2.resize(image, image_size, interpolation=cv2.INTER_AREA)
    elif resize_or_crop == 'crop':
        image = cv2.crop(image, image_size)

    if np.size(np.shape(image)) == 3: 
        image = np.expand_dims(image, axis=0)

    image = np.array(image, dtype=np.float32)
    return image 

def resize(image, width=None, height=None, inter=cv2.INTER_AREA):
    # Grab the image size
    (h, w) = image.shape[:2]

    # If both the width and height are None, then return the original image
    if width is None and height is None:
        return image

    # Check to see if the width is None
    if width is None:
        # Calculate the ratio of the height and construct the dimensions
        r = height / float(h)
        dim = (int(w * r), height)

    # Otherwise, the height is None
    else:
        # Calculate the ratio of the width and construct the dimensions
        r = width / float(w)
        dim = (width, int(h * r))

    # Resize the image
    resized = cv2.resize(image, dim, interpolation=inter)

    # Return the resized image
    return resized 

def resize(image, width=None, height=None, inter=cv2.INTER_AREA):
    # Grab the image size
    (h, w) = image.shape[:2]

    # If both the width and height are None, then return the original image
    if width is None and height is None:
        return image

    # Check to see if the width is None
    if width is None:
        # Calculate the ratio of the height and construct the dimensions
        r = height / float(h)
        dim = (int(w * r), height)

    # Otherwise, the height is None
    else:
        # Calculate the ratio of the width and construct the dimensions
        r = width / float(w)
        dim = (width, int(h * r))

    # Resize the image
    resized = cv2.resize(image, dim, interpolation=inter)

    # Return the resized image
    return resized 

def resize(image, width=None, height=None, inter=cv2.INTER_AREA):
    # Grab the image size
    (h, w) = image.shape[:2]

    # If both the width and height are None, then return the original image
    if width is None and height is None:
        return image

    # Check to see if the width is None
    if width is None:
        # Calculate the ratio of the height and construct the dimensions
        r = height / float(h)
        dim = (int(w * r), height)

    # Otherwise, the height is None
    else:
        # Calculate the ratio of the width and construct the dimensions
        r = width / float(w)
        dim = (width, int(h * r))

    # Resize the image
    resized = cv2.resize(image, dim, interpolation=inter)

    # Return the resized image
    return resized 

def get_mnist_data(is_train, image_size, batchsize):
    ds = MNISTCh('train' if is_train else 'test', shuffle=True)

    if is_train:
        augs = [
            imgaug.RandomApplyAug(imgaug.RandomResize((0.8, 1.2), (0.8, 1.2)), 0.3),
            imgaug.RandomApplyAug(imgaug.RotationAndCropValid(15), 0.5),
            imgaug.RandomApplyAug(imgaug.SaltPepperNoise(white_prob=0.01, black_prob=0.01), 0.25),
            imgaug.Resize((224, 224), cv2.INTER_AREA)
        ]
        ds = AugmentImageComponent(ds, augs)
        ds = PrefetchData(ds, 128*10, multiprocessing.cpu_count())
        ds = BatchData(ds, batchsize)
        ds = PrefetchData(ds, 256, 4)
    else:
        # no augmentation, only resizing
        augs = [
            imgaug.Resize((image_size, image_size), cv2.INTER_CUBIC),
        ]
        ds = AugmentImageComponent(ds, augs)
        ds = BatchData(ds, batchsize)
        ds = PrefetchData(ds, 20, 2)
    return ds 

def get_heatmap(self, target_size):
        heatmap = np.zeros((CocoMetadata.__coco_parts, self.height, self.width), dtype=np.float32)

        for joints in self.joint_list:
            for idx, point in enumerate(joints):
                if point[0] < 0 or point[1] < 0:
                    continue
                CocoMetadata.put_heatmap(heatmap, idx, point, self.sigma)

        heatmap = heatmap.transpose((1, 2, 0))

        # background
        heatmap[:, :, -1] = np.clip(1 - np.amax(heatmap, axis=2), 0.0, 1.0)

        if target_size:
            heatmap = cv2.resize(heatmap, target_size, interpolation=cv2.INTER_AREA)

        return heatmap.astype(np.float16) 

def imresample(img, sz):
    im_data = cv2.resize(img, (sz[1], sz[0]), interpolation=cv2.INTER_AREA) #@UndefinedVariable
    return im_data

    # This method is kept for debugging purpose
#     h=img.shape[0]
#     w=img.shape[1]
#     hs, ws = sz
#     dx = float(w) / ws
#     dy = float(h) / hs
#     im_data = np.zeros((hs,ws,3))
#     for a1 in range(0,hs):
#         for a2 in range(0,ws):
#             for a3 in range(0,3):
#                 im_data[a1,a2,a3] = img[int(floor(a1*dy)),int(floor(a2*dx)),a3]
#     return im_data 

def _resize(img, shape):
        """Resize the specified image.

        :param img: image to resize
        :param shape: desired shape in the format (rows, columns)
        :return: resized image
        """
        if not (OPENCV_AVAILABLE or PILLOW_AVAILABLE):
            raise ValueError('No image library backend found.'' Install either '
                             'OpenCV or Pillow to support image processing.')

        if OPENCV_AVAILABLE:
            return cv2.resize(img, shape, interpolation=cv2.INTER_AREA)

        if PILLOW_AVAILABLE:
            return np.array(PIL.Image.fromarray(img).resize(shape))

        raise NotImplementedError 

def observation(self, obs):
        if self._key is None:
            frame = obs
        else:
            frame = obs[self._key]

        if self._grayscale:
            frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(
            frame, (self._width, self._height), interpolation=cv2.INTER_AREA
        )
        if self._grayscale:
            frame = np.expand_dims(frame, -1)

        if self._key is None:
            obs = frame
        else:
            obs = obs.copy()
            obs[self._key] = frame
        return obs 

def custom_hashing(image, hash_size=8):
    image = cv2.resize(image, (hash_size + 1, hash_size), cv2.INTER_AREA)
    pixel = []
    [rows, cols] = image.shape
    for i in range(0, rows):
        for j in range(0, cols):
            pixel.append(image.item(i, j))
    pixels = list(pixel)

    difference = []
    for row in range(hash_size - 1):
        for col in range(hash_size - 1):
            pixel_left = image.item(row, col)
            pixel_right = image.item(row, col + 1)
            difference.append(pixel_left > pixel_right)
    decimal_value = 0
    hex_string = []
    for index, value in enumerate(difference):
        if value:
            decimal_value += 2 ** (index % 8)
        if (index % 8) == 7:
            hex_string.append(hex(decimal_value)[2:].rjust(2, "0"))
            decimal_value = 0
    return "".join(hex_string) 

def process_image(self, img):
        assert self._bot + self._top < img.shape[0], "Overcrop! bot + top crop >= image height!"
        assert self._right + self._left < img.shape[1], "Overcrop! right + left crop >= image width!"

        bot, right = self._bot, self._right
        if self._bot <= 0:
            bot = -(img.shape[0] + 10)
        if self._right <= 0:
            right = -(img.shape[1] + 10)
        img = img[self._top:-bot, self._left:-right]

        if self.flip:
            img = img[::-1, ::-1]

        if (self.height, self.width) != img.shape[:2]:
            return cv2.resize(img, (self.width, self.height), interpolation=cv2.INTER_AREA)
        return img 

def _modify_observation(self, observation):
        # convert color to grayscale using luma component
        observation = (
            observation[:, :, 0] * 0.299 + observation[:, :, 1] * 0.587 +
            observation[:, :, 2] * 0.114
        )

        observation = cv2.resize(
            observation, (84, 110), interpolation=cv2.INTER_AREA
        )
        observation = observation[18:102, :]
        assert observation.shape == (84, 84)

        # convert to values between 0 and 1
        observation = np.array(observation, dtype=np.uint8)

        return observation 

def letterbox(img, new_shape=416, color=(128, 128, 128), mode='auto', interp=cv2.INTER_AREA):
    # Resize a rectangular image to a 32 pixel multiple rectangle
    # https://github.com/ultralytics/yolov3/issues/232
    shape = img.shape[:2]  # current shape [height, width]

    if isinstance(new_shape, int):
        r = float(new_shape) / max(shape)  # ratio  = new / old
    else:
        r = max(new_shape) / max(shape)
    ratio = r, r  # width, height ratios
    new_unpad = (int(round(shape[1] * r)), int(round(shape[0] * r)))

    # Compute padding https://github.com/ultralytics/yolov3/issues/232
    if mode is 'auto':  # minimum rectangle
        dw = np.mod(new_shape - new_unpad[0], 32) / 2  # width padding
        dh = np.mod(new_shape - new_unpad[1], 32) / 2  # height padding
    elif mode is 'square':  # square
        dw = (new_shape - new_unpad[0]) / 2  # width padding
        dh = (new_shape - new_unpad[1]) / 2  # height padding
    elif mode is 'rect':  # square
        dw = (new_shape[1] - new_unpad[0]) / 2  # width padding
        dh = (new_shape[0] - new_unpad[1]) / 2  # height padding
    elif mode is 'scaleFill':
        dw, dh = 0.0, 0.0
        new_unpad = (new_shape, new_shape)
        ratio = new_shape / shape[1], new_shape / shape[0]  # width, height ratios

    if shape[::-1] != new_unpad:  # resize
        img = cv2.resize(img, new_unpad, interpolation=interp)  # INTER_AREA is better, INTER_LINEAR is faster
    top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
    left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
    img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color)  # add border
    return img, ratio, dw, dh 

def observation(self, frame):
        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(frame, (self.width, self.height), interpolation=cv2.INTER_AREA)
        return frame[:, :, None] 

def observation(self, frame):
        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(frame, (self.width, self.height), interpolation=cv2.INTER_AREA)
        return frame[:, :, None] 

def _observation(self, frame):
    frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
    frame = cv2.resize(frame, (self.width, self.height),
                       interpolation=cv2.INTER_AREA)
    return frame[:, :, None].transpose(2, 0, 1) 

def observation(self, frame):
        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(frame, (self.width, self.height),
                           interpolation=cv2.INTER_AREA)
        return frame.reshape(self.observation_space.low.shape) 

def __init__(self,
               folder,
               read_limit=-1,
               main_size=80,
               crop_size=64,
               augmentation_factor=4,
               *args,
               **kwargs):
    files = os.listdir(folder)
    files = sorted(files)

    if read_limit != -1:
      files = files[:read_limit]
    data = []
    files.sort()
    for f in files:
      image = (cv2.imread(os.path.join(folder, f))[:, :, ::-1] /
               255.0).astype(np.float32)
      image = get_image_center(image)
      # image = cv2.resize(image, (64, 64), interpolation=cv2.INTER_AREA)
      # data.append(image)
      image = cv2.resize(
          image, (main_size, main_size), interpolation=cv2.INTER_AREA)
      for i in range(augmentation_factor):
        new_image = image
        if random.random() < 0.5:
          new_image = new_image[:, ::-1, :]
        sx, sy = random.randrange(main_size - crop_size + 1), random.randrange(
            main_size - crop_size + 1)
        data.append(new_image[sx:sx + crop_size, sy:sy + crop_size])
    data = np.stack(data, axis=0)
    print("# image after augmentation =", len(data))
    super(FolderDataProvider, self).__init__(data, *args, bnw=False,
                                             augmentation=1.0,
                                             output_size=crop_size,
                                             **kwargs) 

def _resize_subtract_mean(image, insize, rgb_mean):
    interp_methods = [cv2.INTER_LINEAR, cv2.INTER_CUBIC, cv2.INTER_AREA, cv2.INTER_NEAREST, cv2.INTER_LANCZOS4]
    interp_method = interp_methods[random.randrange(5)]
    image = cv2.resize(image, (insize, insize), interpolation=interp_method)
    image = image.astype(np.float32)
    image -= rgb_mean
    return image.transpose(2, 0, 1) 

def resize(image, size):
    """ Resize the image to the target (w, h) """
    is_larger = size[0] > image.shape[1] or size[1] > image.shape[0]
    interpolation = cv2.INTER_LINEAR if is_larger else cv2.INTER_AREA
    return cv2.resize(image, size, interpolation=interpolation) 

def resizeAndPad(img, size, padColor=0):

    h, w = img.shape[:2]
    sh, sw = size

    # interpolation method
    if h > sh or w > sw: # shrinking image
        interp = cv2.INTER_AREA
    else: # stretching image
        interp = cv2.INTER_CUBIC

    # aspect ratio of image
    aspect = w/h  # if on Python 2, you might need to cast as a float: float(w)/h

    # compute scaling and pad sizing
    if aspect > 1: # horizontal image
        new_w = sw
        new_h = np.round(new_w/aspect).astype(int)
        pad_vert = (sh-new_h)/2
        pad_top, pad_bot = np.floor(pad_vert).astype(int), np.ceil(pad_vert).astype(int)
        pad_left, pad_right = 0, 0
    elif aspect < 1: # vertical image
        new_h = sh
        new_w = np.round(new_h*aspect).astype(int)
        pad_horz = (sw-new_w)/2
        pad_left, pad_right = np.floor(pad_horz).astype(int), np.ceil(pad_horz).astype(int)
        pad_top, pad_bot = 0, 0
    else: # square image
        new_h, new_w = sh, sw
        pad_left, pad_right, pad_top, pad_bot = 0, 0, 0, 0

    # set pad color
    if len(img.shape) is 3 and not isinstance(padColor, (list, tuple, np.ndarray)): # color image but only one color provided
        padColor = [padColor]*3

    # scale and pad
    scaled_img = cv2.resize(img, (new_w, new_h), interpolation=interp)
    scaled_img = cv2.copyMakeBorder(scaled_img, pad_top, pad_bot, pad_left, pad_right, borderType=cv2.BORDER_CONSTANT, value=padColor)

    return scaled_img 

def resizeAndPad(img, size, padColor=0):

    h, w = img.shape[:2]
    sh, sw = size

    # interpolation method
    if h > sh or w > sw: # shrinking image
        interp = cv2.INTER_AREA
    else: # stretching image
        interp = cv2.INTER_CUBIC

    # aspect ratio of image
    aspect = w/h  # if on Python 2, you might need to cast as a float: float(w)/h

    # compute scaling and pad sizing
    if aspect > 1: # horizontal image
        new_w = sw
        new_h = np.round(new_w/aspect).astype(int)
        pad_vert = (sh-new_h)/2
        pad_top, pad_bot = np.floor(pad_vert).astype(int), np.ceil(pad_vert).astype(int)
        pad_left, pad_right = 0, 0
    elif aspect < 1: # vertical image
        new_h = sh
        new_w = np.round(new_h*aspect).astype(int)
        pad_horz = (sw-new_w)/2
        pad_left, pad_right = np.floor(pad_horz).astype(int), np.ceil(pad_horz).astype(int)
        pad_top, pad_bot = 0, 0
    else: # square image
        new_h, new_w = sh, sw
        pad_left, pad_right, pad_top, pad_bot = 0, 0, 0, 0

    # set pad color
    if len(img.shape) is 3 and not isinstance(padColor, (list, tuple, np.ndarray)): # color image but only one color provided
        padColor = [padColor]*3

    # scale and pad
    scaled_img = cv2.resize(img, (new_w, new_h), interpolation=interp)
    scaled_img = cv2.copyMakeBorder(scaled_img, pad_top, pad_bot, pad_left, pad_right, borderType=cv2.BORDER_CONSTANT, value=padColor)

    return scaled_img 

def affine(img, angle, translate, scale, shear, interpolation=cv2.INTER_LINEAR, mode=cv2.BORDER_CONSTANT, fillcolor=0):
    """Apply affine transformation on the image keeping image center invariant
    Args:
        img (numpy ndarray): numpy ndarray to be transformed.
        angle (float or int): rotation angle in degrees between -180 and 180, clockwise direction.
        translate (list or tuple of integers): horizontal and vertical translations (post-rotation translation)
        scale (float): overall scale
        shear (float): shear angle value in degrees between -180 to 180, clockwise direction.
        interpolation (``cv2.INTER_NEAREST` or ``cv2.INTER_LINEAR`` or ``cv2.INTER_AREA``, ``cv2.INTER_CUBIC``):
            An optional resampling filter.
            See `filters`_ for more information.
            If omitted, it is set to ``cv2.INTER_CUBIC``, for bicubic interpolation.
        mode (``cv2.BORDER_CONSTANT`` or ``cv2.BORDER_REPLICATE`` or ``cv2.BORDER_REFLECT`` or ``cv2.BORDER_REFLECT_101``)
            Method for filling in border regions. 
            Defaults to cv2.BORDER_CONSTANT, meaning areas outside the image are filled with a value (val, default 0)
        val (int): Optional fill color for the area outside the transform in the output image. Default: 0
    """
    if not _is_numpy_image(img):
        raise TypeError('img should be numpy Image. Got {}'.format(type(img)))

    assert isinstance(translate, (tuple, list)) and len(translate) == 2, \
        "Argument translate should be a list or tuple of length 2"

    assert scale > 0.0, "Argument scale should be positive"

    output_size = img.shape[0:2]
    center = (img.shape[1] * 0.5 + 0.5, img.shape[0] * 0.5 + 0.5)
    matrix = _get_affine_matrix(center, angle, translate, scale, shear)
    
    if img.shape[2]==1:
        return cv2.warpAffine(img, matrix, output_size[::-1],interpolation, borderMode=mode, borderValue=fillcolor)[:,:,np.newaxis]
    else:
        return cv2.warpAffine(img, matrix, output_size[::-1],interpolation, borderMode=mode, borderValue=fillcolor) 

def preproc_for_test(image, insize, mean):
    interp_methods = [cv2.INTER_LINEAR, cv2.INTER_CUBIC, cv2.INTER_AREA, cv2.INTER_NEAREST, cv2.INTER_LANCZOS4]
    interp_method = interp_methods[random.randrange(5)]
    image = cv2.resize(image, (insize[0], insize[1]),interpolation=interp_method)
    image = image.astype(np.float32)
    image -= mean
    return image.transpose(2, 0, 1) 

def observation(self, frame):
        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        frame = cv2.resize(frame, (self.width, self.height), interpolation=cv2.INTER_AREA)
        return frame[:, :, None]