habana_frameworks.mediapipe.fn.Saturation

Class:
  • habana_frameworks.mediapipe.fn.Saturation(**kwargs)

Define graph call:
  • __call__(input, saturation_level_tensor)

Parameter:
  • input - Input tensor to operator. Supported dimensions: minimum = 4, maximum = 4. Supported data types: UINT8, UINT16.

  • (optional) saturation_level_tensor - Tensor containing saturation level for each image in the batch. Supported dimensions: minimum = 1, maximum = 1. Supported data types: FLOAT32.

Description:

Saturation operator modifies the saturation level of an image by specified change factor. It applies the HSV space modification to an RGB value to get the modified RGB value.

Supported backend:
  • HPU

Keyword Arguments

kwargs

Description

saturation_level

Change factor by which the saturation level of an image needs to be modified.

  • Type: float

  • Default: 0.0

  • Optional: yes

dtype

Output data type.

  • Type: habana_frameworks.mediapipe.media_types.dtype

  • Default: UINT8

  • Optional: yes

  • Supported data type:

    • UINT8

    • UINT16

Note

  1. Input tensor and Output tensor must be of the same data type.

  2. Input1 and output tensor should be in NCHW layout. Input2 is one dimensional tensor of size N.

  3. Saturation level can be provided either as scalar for all batches or 1D tensor.

  4. Number of channels in input and output should be 3.

  5. saturation_level should be >= 0

    • 0: completely de-saturated image

    • 1: no change to image’s saturation

Example: Saturation Operator

The following code snippet shows usage of Saturation operator:

from habana_frameworks.mediapipe import fn
from habana_frameworks.mediapipe.mediapipe import MediaPipe
from habana_frameworks.mediapipe.media_types import imgtype as it
from habana_frameworks.mediapipe.media_types import dtype as dt
import matplotlib.pyplot as plt
import os


g_display_timeout = os.getenv("DISPLAY_TIMEOUT") or 5

# Create MediaPipe derived class
class myMediaPipe(MediaPipe):
    def __init__(self, device, queue_depth, batch_size, num_threads,
                op_device, dir, img_h, img_w):
        super(
            myMediaPipe,
            self).__init__(
            device,
            queue_depth,
            batch_size,
            num_threads,
            self.__class__.__name__)

        self.input = fn.ReadImageDatasetFromDir(shuffle=False,
                                                dir=dir,
                                                format="jpg",
                                                device="cpu")

        # WHCN
        self.decode = fn.ImageDecoder(device="hpu",
                                      output_format=it.RGB_P,
                                      resize=[img_w, img_h])

        self.saturation = fn.Saturation(saturation_level=0.3,
                                        dtype=dt.UINT8,
                                        device=op_device)

        # WHCN -> CWHN
        self.transpose = fn.Transpose(permutation=[2, 0, 1, 3],
                                      tensorDim=4,
                                      dtype=dt.UINT8,
                                      device=op_device)

    def definegraph(self):
        images, labels = self.input()
        images = self.decode(images)
        images = self.saturation(images)
        images = self.transpose(images)
        return images, labels


def display_images(images, batch_size, cols):
    rows = (batch_size + 1) // cols
    plt.figure(figsize=(10, 10))
    for i in range(batch_size):
        ax = plt.subplot(rows, cols, i + 1)
        plt.imshow(images[i])
        plt.axis("off")
    plt.show(block=False)
    plt.pause(g_display_timeout)
    plt.close()


def run(device, op_device):
    batch_size = 6
    queue_depth = 2
    num_threads = 1
    img_width = 200
    img_height = 200
    base_dir = os.environ['DATASET_DIR']
    dir = base_dir + "/img_data/"
    columns = 3

    # Create MediaPipe object
    pipe = myMediaPipe(device, queue_depth, batch_size,
                      num_threads, op_device, dir,
                      img_height, img_width)

    # Build MediaPipe
    pipe.build()

    # Initialize MediaPipe iterator
    pipe.iter_init()

    # Run MediaPipe
    images, labels = pipe.run()

    def as_cpu(tensor):
        if (callable(getattr(tensor, "as_cpu", None))):
            tensor = tensor.as_cpu()
        return tensor

    # Copy data to host from device as numpy array
    images = as_cpu(images).as_nparray()
    labels = as_cpu(labels).as_nparray()
    del pipe

    # Display images
    display_images(images, batch_size, columns)


if __name__ == "__main__":
    dev_opdev = {'mixed': ['hpu'],
                'legacy': ['hpu']}

    for dev in dev_opdev.keys():
        for op_dev in dev_opdev[dev]:
            run(dev, op_dev)
Copy to clipboard

Images with Changed Saturation 1

Image1 of saturation
Image2 of saturation
Image3 of saturation
Image4 of saturation
Image5 of saturation
Image6 of saturation
1

Licensed under a CC BY SA 4.0 license. The images used here are taken from https://data.caltech.edu/records/mzrjq-6wc02.