图像灰度上移变换
该算法将实现图像灰度值的上移,从而提升图像的亮度,由于图像的灰度值位于0到255之间,需要对灰度值进行溢出判断。
代码如下:
import cv2 import numpy as np import matplotlib.pyplot as plt img = cv2.imread("src.png") grayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) height, width = grayImage.shape[:2] result = np.zeros((height, width), np.uint8) # 图像灰度上移变换 for i in range(height): for j in range(width): if int(grayImage[i, j] + 50) > 255: gray = 255 else: gray = grayImage[i, j] + 50 result[i, j] = np.uint8(gray) cv2.imshow("src", grayImage) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
效果如下:
图像对比度增强变换
import cv2 import numpy as np import matplotlib.pyplot as plt img = cv2.imread("src.png") grayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) height, width = grayImage.shape[:2] result = np.zeros((height, width), np.uint8) # 图像灰度上移变换 for i in range(height): for j in range(width): if int(grayImage[i, j]*1.5 + 50) > 255: gray = 255 else: gray = grayImage[i, j]*1.5 + 50 result[i, j] = np.uint8(gray) cv2.imshow("src", grayImage) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
效果如下:
图像对比度增强减弱
import cv2 import numpy as np import matplotlib.pyplot as plt img = cv2.imread("src.png") grayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) height, width = grayImage.shape[:2] result = np.zeros((height, width), np.uint8) # 图像灰度上移变换 for i in range(height): for j in range(width): if int(grayImage[i, j]*0.8 + 50) > 255: gray = 255 else: gray = grayImage[i, j]*0.8 + 50 result[i, j] = np.uint8(gray) cv2.imshow("src", grayImage) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
效果如下:
图像灰度反色变换
反色变换又称为线性灰度补变换,它是对原图像的像素值进行反转,即黑色变为白色,白色变为黑色
import cv2 import numpy as np import matplotlib.pyplot as plt img = cv2.imread("src.png") grayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) height, width = grayImage.shape[:2] result = np.zeros((height, width), np.uint8) # 图像灰度上移变换 for i in range(height): for j in range(width): gray = 255 - int(grayImage[i,j]) result[i, j] = np.uint8(gray) cv2.imshow("src", grayImage) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
效果如下:
图像灰度非线性变换: DB=DAxDA/255
图像的灰度非线性变换主要包括对数变换、幂次变换、指数变换、分段函数变换,通过非线性关系对图像进行灰度处理,下面主要讲解三种常见类型的灰度非线性变换。
原始图像的灰度值按照DB=DA*DA/255
import cv2 import numpy as np import matplotlib.pyplot as plt img = cv2.imread("src.png") grayImage = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) height, width = grayImage.shape[:2] result = np.zeros((height, width), np.uint8) # 图像灰度上移变换 for i in range(height): for j in range(width): gray = int(grayImage[i, j])*int(grayImage[i, j])/255 result[i, j] = np.uint8(gray) cv2.imshow("src", grayImage) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
效果如下:
图像灰度对数变换
由于对数曲线在像素值较低的区域斜率大,在像素值较高的区域斜率较小,所以图像经过对数变换后,较暗区域的对比度将有所提升。这种变换可用于增强图像的暗部细节,从而用来扩展被压缩的高值图像中的较暗像素。
对数变换实现了扩展低灰度值而压缩高灰度值的效果,被广泛地应用于频谱图像的显示中。一个典型的应用是傅立叶频谱,其动态范围可能宽达0~106直接显示频谱时,图像显示设备的动态范围往往不能满足要求,从而丢失大量的暗部细节;而在使用对数变换之后,图像的动态范围被合理地非线性压缩,从而可以清晰地显示。在下图中,未经变换的频谱经过对数变换后,增加了低灰度区域的对比度,从而增强暗部的细节。
import cv2 import numpy as np import matplotlib.pyplot as plt def log_plot(c): x = np.arange(0, 256, 0.01) y = c * np.log(1+x) plt.plot(x, y, "r", linewidth=1) plt.rcParams["font.sans-serif"] = ["SimHei"] plt.title("对数变换函数") plt.xlim(0, 255) plt.ylim(0, 255) plt.show() # 对数变换 def log(c, img): output = c * np.log(1.0+img) output = np.uint8(output) return output img = cv2.imread("src.png") log_plot(42) result = log(42, img) cv2.imshow("src", img) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()
图像灰度伽玛变换
伽玛变换又称为指数变换或幂次变换,另一种常用的灰度非线性变换。
Db = cXDa^y
- 当γ>1时,会拉伸图像中灰度级较高的区域,压缩灰度级较低的部分。
- 当γ<1时,会拉伸图像中灰度级较低的区域,压缩灰度级较高的部分。
- 当γ=1时,该灰度变换是线性的,此时通过线性方式改变原图像。
import cv2 import numpy as np import matplotlib.pyplot as plt def gamma_plot(c, v): x = np.arange(0, 256, 0.01) y = c *x**v plt.plot(x, y, "r", linewidth=1) plt.rcParams["font.sans-serif"] = ["SimHei"] plt.title("对数变换函数") plt.xlim([0, 255]) plt.ylim([0, 255]) plt.show() # 对数变换 def gamma(img, c, v): lut = np.zeros(256, dtype=np.float32) for i in range(256): lut[i] = c*i**v # 灰度值的映射 output = cv2.LUT(img, lut) output = np.uint8(output + 0.5) return output img = cv2.imread("src.png") gamma_plot(0.0000005, 4) result = gamma(img, 0.0000005, 4) cv2.imshow("src", img) cv2.imshow("result", result) if cv2.waitKey() == 27: cv2.destroyAllWindows()