九、图像直方图

一、图像直方图的属性

说白了就是将图像上的各个颜色通道上的像素点的像素值进行统计，例如：像素值为14的像素点个数有几个，进行显示。
在这里插入图片描述
图像的像素值取值范围为[0,255]，这个范围也成为直方图的range也就是直方图的横坐标轴
每一个像素值所对应的个数称之为bin

二、对图像进行直方图统计

image.ravel()把图像的所有像素点信息进行统计
plt.hist(image.ravel(),256,[0,256])将图像信息进行统计，统计成256个bin，范围为[0,255]
cv2.calcHist([image],[i],None,[256],[0,256])[image]为当前出来图像，[i]这里使用了一个循环也就是依次BGR三个通道，None是掩膜信息这里没有用到，[256]表示直方图的size，[0,256]BGR三颜色的像素值的范围

import cv2
import numpy as np
from matplotlib import pyplot as pltdef plot(image):plt.hist(image.ravel(),256,[0,256])plt.show("matlab自带直方图")def hist(image):color = ('blue','green','red')for i,color in enumerate(color):hist = cv2.calcHist([image],[i],None,[256],[0,256])plt.plot(hist,color=color)plt.xlim([0,256])plt.show()src = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\a1.jpg")
cv2.imshow("image",src)
cv2.namedWindow("image",cv2.WINDOW_AUTOSIZE)plot(src)
hist(src)cv2.waitKey(0)
cv2.destroyAllWindows()

效果图如下：
在这里插入图片描述

三、直方图的均衡化

OpenCV中的直方图均衡化针对的都是灰度图

Ⅰ全局直方图均衡化

import cv2
import numpy as np
from matplotlib import pyplot as pltdef equalizeHist(image):gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)dst = cv2.equalizeHist(gray)#yy = cv2.cvtColor(dst,cv2.COLOR_GRAY2BGR)cv2.imshow("equalizeHist",dst)src = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\mi.jpg")
cv2.imshow("image",src)
cv2.namedWindow("image",cv2.WINDOW_AUTOSIZE)equalizeHist(src)cv2.waitKey(0)
cv2.destroyAllWindows()

效果图如下：
在这里插入图片描述

Ⅱ局部直方图均衡化

import cv2
import numpy as np
from matplotlib import pyplot as pltdef clahe(image):gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)clahe = cv2.createCLAHE(clipLimit=2.0,tileGridSize=(8,8))dst = clahe.apply(gray)cv2.imshow("clahe",dst)src = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\mi.jpg")
cv2.imshow("image",src)
cv2.namedWindow("image",cv2.WINDOW_AUTOSIZE)clahe(src)cv2.waitKey(0)
cv2.destroyAllWindows()

效果图如下：
在这里插入图片描述

四、直方图反向投影

Ⅰ2D直方图

cv2.calcHist([image],[0,1],None,[180,256],[0,180,0,256])其中[180,256]表示bin的个数，可以修改，当然范围越小越精确

import cv2
import numpy as np
from matplotlib import pyplot as pltdef hist2d(image):hsv = cv2.cvtColor(image,cv2.COLOR_BGR2HSV)hist = cv2.calcHist([image],[0,1],None,[180,256],[0,180,0,256])cv2.imshow("hist2d",hist)def hist2d_1(image):hsv = cv2.cvtColor(image,cv2.COLOR_BGR2HSV)hist = cv2.calcHist([image],[0,1],None,[180,256],[0,180,0,256])plt.imshow(hist,interpolation='nearest')plt.title("2D Histogram")plt.show()src = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\l.png")
cv2.imshow("image",src)
cv2.namedWindow("image",cv2.WINDOW_AUTOSIZE)
hist2d(src)
hist2d_1(src)
cv2.waitKey(0)
cv2.destroyAllWindows()

效果图如下：
在这里插入图片描述

Ⅱ直方图反向投影

cv2.calcHist([roi_hsv],[0,1],None,[32,48],[0,180,0,256])其中[32,48]表示bin的个数，可以修改，当然范围越小越精确

import cv2
import numpy as np
from matplotlib import pyplot as pltdef back_projection():sample = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\yg1.jpg")target = cv2.imread(r"G:\Juptyer_workspace\study\opencv\opencv3\yg.jpg")roi_hsv = cv2.cvtColor(sample,cv2.COLOR_BGR2HSV)target_hsv = cv2.cvtColor(target,cv2.COLOR_BGR2HSV)cv2.imshow("sample",sample)cv2.imshow("target",target)roiHist = cv2.calcHist([roi_hsv],[0,1],None,[32,48],[0,180,0,256])cv2.normalize(roiHist,roiHist,0,255,cv2.NORM_MINMAX)dst = cv2.calcBackProject([target_hsv],[0,1],roiHist,[0,180,0,256],1)cv2.imshow("back_projection",dst)back_projection()
cv2.waitKey(0)
cv2.destroyAllWindows()