如何创建 "bluescale" 图像

Question

如何使用opencv制作类似本站的图片处理效果：

https://manytools.org/image/colorize-filter/

原创

已处理

我试过保留蓝色通道，或者给它填充灰度图，红色和绿色通道都减少到0，但是没有帮助，任何帮助都不胜感激。

Answer 1

您可以尝试使用 cv2.cvtColor 将 RGB 转换为 BGR 或使用 cv2.applyColorMap

应用颜色图

import cv2
img = cv2.imread("car.jpg")
# First method
img_bgr = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
# Second method
img_ocean = cv2.applyColorMap(img, cv2.COLORMAP_OCEAN)

以下是 img_bgr 和 img_ocean 的示例输出。希望对你有帮助。

已编辑：您也可以尝试自定义您自己的 ColorMap

Answer 2

我设法通过首先将图像转换为 HSV colorspace，然后将色调值调整为 120（代表蓝色 [255, 0, 0] 的 OpenCV 转换）来获得类似的效果。

有了这个我得到了这张照片：

使用以下代码：

import cv2

img = cv2.imread("./EZCUD.jpg")
img_blue = img.copy()
img_blue = cv2.cvtColor(img_blue, cv2.COLOR_BGR2HSV)
img_blue[:, :, 0] = 120
cv2.imwrite("./test.jpg", cv2.cvtColor(img_blue, cv2.COLOR_HSV2BGR))

编辑：

通过在 HSV 处理后对独立 BGR 通道应用伽马校正，我设法获得了更好的结果。我将图像标准化到范围 [0,1] 并在 RG 通道中取平方幂并在蓝色通道中取平方根。

这是结果：

使用以下代码：

import cv2
import numpy as np

img = cv2.imread("./EZCUD.jpg")
img_blue = img.copy()

img_blue = cv2.cvtColor(img_blue, cv2.COLOR_BGR2HSV)
img_blue[:, :, 0] = 120
img_blue = cv2.cvtColor(img_blue, cv2.COLOR_HSV2BGR)
img_blue = img_blue.astype("float")
img_blue /= 255
img_blue[:, :, 1:] = img_blue[:, :, 1:] ** 2
img_blue[:, :, 0] = img_blue[:, :, 0] ** (1 / 2)
img_blue *= 255

cv2.imwrite("./test.jpg", img_blue.astype("uint8"))

您可以尝试使用这些参数，直到结果更接近您的预期。

Answer 3

但没有充实它。我使用您的示例图像创建了自定义颜色图。结果并不完美，但非常接近。差异在轮舱边缘最为明显。我怀疑这是您的参考软件和 cv2 将图像转换为灰度的方式不同。

import cv2
img = cv2.imread(r'car.jpg', cv2.IMREAD_GRAYSCALE)
lut = np.array([
    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],
    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],
    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],
    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],
    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],    [[0,0,0]],
    [[0,0,0]],    [[1,0,0]],    [[1,0,0]],    [[2,0,0]],
    [[2,0,0]],    [[3,0,0]],    [[7,0,0]],    [[13,0,0]],
    [[19,0,0]],    [[23,0,0]],    [[24,0,0]],    [[24,0,0]],
    [[25,0,0]],    [[27,0,0]],    [[27,0,0]],    [[28,0,0]],
    [[30,0,0]],    [[32,0,0]],    [[33,0,0]],    [[33,0,0]],
    [[34,0,0]],    [[36,0,0]],    [[39,0,0]],    [[40,0,0]],
    [[41,0,0]],    [[45,0,0]],    [[48,0,0]],    [[48,0,0]],
    [[49,0,0]],    [[53,0,0]],    [[56,0,0]],    [[58,0,0]],
    [[64,0,0]],    [[67,0,0]],    [[69,0,0]],    [[71,0,0]],
    [[75,0,0]],    [[78,0,0]],    [[80,0,0]],    [[82,0,0]],
    [[85,0,0]],    [[88,0,0]],    [[89,0,0]],    [[91,0,0]],
    [[94,0,0]],    [[96,0,0]],    [[99,0,0]],    [[103,0,0]],
    [[106,0,0]],    [[107,0,0]],    [[110,0,0]],    [[111,0,0]],
    [[113,0,0]],    [[113,0,0]],    [[117,0,0]],    [[117,0,0]],
    [[121,0,0]],    [[124,0,0]],    [[129,0,0]],    [[132,0,0]],
    [[136,0,0]],    [[138,0,0]],    [[138,0,0]],    [[139,0,0]],
    [[141,0,0]],    [[144,0,0]],    [[144,0,0]],    [[144,0,0]],
    [[145,0,0]],    [[149,0,0]],    [[149,0,0]],    [[151,0,0]],
    [[154,0,0]],    [[155,0,0]],    [[157,0,0]],    [[159,0,0]],
    [[161,0,0]],    [[164,0,0]],    [[165,0,0]],    [[167,0,0]],
    [[169,0,0]],    [[172,0,0]],    [[174,0,0]],    [[175,0,0]],
    [[178,0,0]],    [[181,0,0]],    [[184,0,0]],    [[186,0,0]],
    [[189,0,0]],    [[191,0,0]],    [[195,0,0]],    [[198,0,0]],
    [[201,0,0]],    [[203,0,0]],    [[206,0,0]],    [[209,0,0]],
    [[212,0,0]],    [[216,0,0]],    [[218,0,0]],    [[222,0,0]],
    [[224,0,0]],    [[227,1,1]],    [[230,1,1]],    [[233,1,1]],
    [[236,1,1]],    [[239,2,2]],    [[241,2,2]],    [[243,3,3]],
    [[246,4,4]],    [[248,5,5]],    [[250,6,6]],    [[252,7,7]],
    [[253,9,9]],    [[254,12,12]],    [[255,15,15]],    [[255,18,18]],
    [[255,21,21]],    [[255,25,25]],    [[255,29,29]],    [[255,31,31]],
    [[255,36,36]],    [[255,38,38]],    [[255,42,42]],    [[255,46,46]],
    [[255,50,50]],    [[255,53,53]],    [[255,57,57]],    [[255,60,60]],
    [[255,62,62]],    [[255,65,65]],    [[255,68,68]],    [[255,71,71]],
    [[255,74,74]],    [[255,76,76]],    [[255,80,80]],    [[255,82,82]],
    [[255,86,86]],    [[255,88,88]],    [[255,91,91]],    [[255,93,93]],
    [[255,97,97]],    [[255,99,99]],    [[255,102,102]],    [[255,105,105]],
    [[255,108,108]],    [[255,112,112]],    [[255,117,117]],    [[255,118,118]],
    [[255,121,121]],    [[255,122,122]],    [[255,125,125]],    [[255,126,126]],
    [[255,129,129]],    [[255,131,131]],    [[255,133,133]],    [[255,135,135]],
    [[255,139,139]],    [[255,144,144]],    [[255,147,147]],    [[255,152,152]],
    [[255,154,154]],    [[255,157,157]],    [[255,162,162]],    [[255,164,164]],
    [[255,168,168]],    [[255,170,170]],    [[255,174,174]],    [[255,177,177]],
    [[255,180,180]],    [[255,185,185]],    [[255,188,188]],    [[255,191,191]],
    [[255,193,193]],    [[255,196,196]],    [[255,198,198]],    [[255,200,200]],
    [[255,202,202]],    [[255,204,204]],    [[255,206,206]],    [[255,209,209]],
    [[255,212,212]],    [[255,214,214]],    [[255,216,216]],    [[255,218,218]],
    [[255,221,221]],    [[255,223,223]],    [[255,225,225]],    [[255,227,227]],
    [[255,228,228]],    [[255,230,230]],    [[255,231,231]],    [[255,234,234]],
    [[255,235,235]],    [[255,238,238]],    [[255,238,238]],    [[255,241,241]],
    [[255,242,242]],    [[255,244,244]],    [[255,247,247]],    [[255,249,249]],
    [[255,250,250]],    [[255,253,253]],    [[255,253,253]],    [[255,254,254]],
    [[255,254,254]],    [[255,253,253]],    [[255,252,252]],    [[255,253,253]],
    [[255,254,254]],    [[255,254,254]],    [[255,254,254]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],
    [[255,255,255]],    [[255,255,255]],    [[255,255,255]],    [[255,255,255]]],
    dtype=np.uint8)
cv2.imshow('cv2', cv2.applyColorMap(img, lut))

很明显，我对转换进行了硬编码 look-up table，但是通过更多的分析，也许可以创建一个简单地生成它的公式。

Answer 4

生成模板图像：

import cv2

img=np.array([[x for x in range(256)]]*128,dtype='uint8')

cv2.imshow('',img)
cv2.imwrite('gradient.jpg',img)

从网站按需生成结果

将模板应用于输入图像：

import numpy as np
import cv2

img=cv2.imread('processed.jpg')
img2=cv2.imread('in2.jpg')

line=img[0]
line=line.reshape(line.shape[0],1,-1)

result=cv2.applyColorMap(img2,line)

cv2.imshow('result',result)
cv2.waitKey()

输入图像：

如何创建 "bluescale" 图像

How to create a "bluescale" image

python

opencv