PaperFactory

class augraphy.base.paperfactory.PaperFactory(texture_path='./paper_textures', blend_texture=1, blend_method='random', p=1)

Bases: Augmentation

Replaces the starting paper image with a texture randomly chosen from a directory and resized to fit or cropped and tiled to fit.

Parameters:

• texture_path (string, optional) – Directory location to pull paper textures from.

• blend_texture (int, optional) – Flag to blend multiple textures.

• blend_method (string, optional) – The method to blend multiple textures.

• p (float, optional) – The probability that this Augmentation will be applied.

check_paper_edges(texture)

Crop image section with better texture.

Parameters:

texture (numpy array) – Texture image.

resize(texture, shape)

Scales and zooms a given texture to fit a given shape.

Parameters:

• texture (numpy array.3.) – Texture image.

• shape (list or tuple) – x and y shape of scaled image.

retrieve_texture(image)

Overview

PaperFactory is an augmentation specific to paper phase of the augmnentation pipeline. It applies texture into image based on a random image chosen from the provided directory.

Example

In this example, PaperFactory is use to apply texture extracted from an image in the chosen texture image directory into an input image.

# import libraries
from augraphy import *
import cv2
import numpy as np
import os

# create directory for paper texture
paper_texture_dir = os.path.join(os.getcwd(), "paper_textures/")
os.makedirs(paper_texture_dir, exist_ok = True)

# create an empty image
image_texture = np.full((1000,1000), fill_value=255, dtype="uint8")

# create texture
for i in range(5):
    i += 8
    image_texture[::i,::i] = 0

# save texture image in the paper texture directory
cv2.imwrite(paper_texture_dir+"texture.png", image_texture)



ink_phase   = []
paper_phase = [PaperFactory(texture_path=paper_texture_dir, p=1)]
post_phase  = []
pipeline    = AugraphyPipeline(ink_phase, paper_phase, post_phase)

image = np.full((1200, 1200,3), 250, dtype="uint8")
cv2.putText(
    image,
    "Lorem ipsum dolor sit amet, consectetur adipiscing elit",
    (80, 250),
    cv2.FONT_HERSHEY_SIMPLEX,
    1.2,
    0,
    3,
)

augmented_image = pipeline.augment(image)["output"]

Input image:

Texture image:

Augmented image: