ocv_deslant: Deslant images by putting cursive text upright

In image.textlinedetector: Segment Images in Text Lines and Words

Deslant images by putting cursive text upright

Description

This algorithm sets handwritten text in images upright by removing cursive writing style. One can use it as a preprocessing step for handwritten text recognition.

• image_deslant expects a magick-image and performs grayscaling before doing deslanting

• ocv_deslant expects a ocv-image and does not perform grayscaling before doing deslanting

Usage

ocv_deslant(image, bgcolor = 255, lower_bound = -1, upper_bound = 1)

image_deslant(image, bgcolor = 255, lower_bound = -1, upper_bound = 1)

Arguments

image	an object of class opencv-image (for ocv_deslant) with pixel values between 0 and 255 or a magick-image (for image_deslant)
bgcolor	integer value with the background color to use to fill the gaps of the sheared image that is returned. Defaults to white: 255
lower_bound	lower bound of shear values. Defaults to -1
upper_bound	upper bound of shear values. Defaults to 1

Value

an object of class opencv-image or magick-image with the deslanted image

Examples

library(magick)
library(opencv)
library(image.textlinedetector)
path <- system.file(package = "image.textlinedetector", "extdata", "cursive.png")
img  <- ocv_read(path)
img  <- ocv_grayscale(img)
img
up   <- ocv_deslant(img)
up

img  <- image_read(path)
img
image_deslant(img)

image.textlinedetector documentation built on Nov. 10, 2023, 1:09 a.m.