使用 Python OpenCV 的收缩/凸起失真

Question

我想使用 Python OpenCV 在图像上应用收缩/凸出滤镜。结果应该是这个例子的某种形式：

https://pixijs.io/pixi-filters/tools/screenshots/dist/bulge-pinch.gif

我读过以下 stackoverflow 帖子，它应该是过滤器的正确公式：Formulas for Barrel/Pincushion Distortion

但我正在努力在 Python OpenCV 中实现这一点。

我读过有关在图像上应用滤镜的地图：使用 OpenCv-python 的扭曲效果

根据我的理解，代码可能如下所示：

import numpy as np

import cv2 as cv

f_img = 'example.jpg'

im_cv = cv.imread(f_img)

# grab the dimensions of the image

(h, w, _) = im_cv.shape

# set up the x and y maps as float32

flex_x = np.zeros((h, w), np.float32)

flex_y = np.zeros((h, w), np.float32)

# create map with the barrel pincushion distortion formula

for y in range(h):

    for x in range(w):

        flex_x[y, x] = APPLY FORMULA TO X

        flex_y[y, x] = APPLY FORMULA TO Y

# do the remap  this is where the magic happens

dst = cv.remap(im_cv, flex_x, flex_y, cv.INTER_LINEAR)

cv.imshow('src', im_cv)

cv.imshow('dst', dst)

cv.waitKey(0)

cv.destroyAllWindows()

这是实现示例图像中呈现的失真的正确方法吗？非常感谢有关有用资源或最好的示例的任何帮助。

Answer 1

在熟悉了 ImageMagick 源代码后，我找到了一种应用失真公式的方法。借助OpenCV 重映射函数，这是一种扭曲图像的方法：import numpy as npimport cv2 as cvf_img = 'example.jpg'im_cv = cv.imread(f_img)# grab the dimensions of the image(h, w, _) = im_cv.shape# set up the x and y maps as float32flex_x = np.zeros((h, w), np.float32)flex_y = np.zeros((h, w), np.float32)# create map with the barrel pincushion distortion formulafor y in range(h):    delta_y = scale_y * (y - center_y)    for x in range(w):        # determine if pixel is within an ellipse        delta_x = scale_x * (x - center_x)        distance = delta_x * delta_x + delta_y * delta_y        if distance >= (radius * radius):            flex_x[y, x] = x            flex_y[y, x] = y        else:            factor = 1.0            if distance > 0.0:                factor = math.pow(math.sin(math.pi * math.sqrt(distance) / radius / 2), -amount)            flex_x[y, x] = factor * delta_x / scale_x + center_x            flex_y[y, x] = factor * delta_y / scale_y + center_y# do the remap  this is where the magic happensdst = cv.remap(im_cv, flex_x, flex_y, cv.INTER_LINEAR)cv.imshow('src', im_cv)cv.imshow('dst', dst)cv.waitKey(0)cv.destroyAllWindows()这与使用 ImageMagick 中的Convert -implode函数具有相同的效果。

Answer 2

您可以使用 Python Wand（使用 ImageMagick）中的内爆和爆炸选项来完成此操作。输入：from wand.image import Imageimport numpy as npimport cv2with Image(filename='zelda1.jpg') as img:    img.virtual_pixel = 'black'    img.implode(0.5)    img.save(filename='zelda1_implode.jpg')    # convert to opencv/numpy array format    img_implode_opencv = np.array(img)    img_implode_opencv = cv2.cvtColor(img_implode_opencv, cv2.COLOR_RGB2BGR)with Image(filename='zelda1.jpg') as img:    img.virtual_pixel = 'black'    img.implode(-0.5 )    img.save(filename='zelda1_explode.jpg')    # convert to opencv/numpy array format    img_explode_opencv = np.array(img)    img_explode_opencv = cv2.cvtColor(img_explode_opencv, cv2.COLOR_RGB2BGR)# display result with opencvcv2.imshow("IMPLODE", img_implode_opencv)cv2.imshow("EXPLODE", img_explode_opencv)cv2.waitKey(0)内爆：爆炸：