604-1189960-阈值ThreshHold.sy.md

show	version	enable_checker
step	1.0	true

opencv

回忆

• 上次研究了视频的修改
  • 视频另存
  • 视频大小调整
  • 视频空间裁剪
  • 视频时间截取
  • 在原视频上画别的
  • 在原视频上添加字幕
• opencv 确实很强大
• 还能做点什么？🤔

设置阈值

import cv2 
image = cv2.imread("/home/shiyanlou/gear.jpg")
t1, dst1 = cv2.threshold(image,127,255,cv2.THRESH_BINARY)
cv2.imshow("img",image) 
cv2.imshow("dst1",dst1)
cv2.waitKey()
cv2.destroyAllWindows()

• 效果

• 如何理解threshold

threshold

• 阈值
  • 就像一个门槛
  • 过了门槛
  • 就算进屋了
  • 到了另一个阶段了

• 手册里面怎么说呢？

查看手册

firefox https://docs.opencv.org/4.x/d7/d4d/tutorial_py_thresholding.html

• 具体手册内容

• 如何理解type呢？

type

• 可以根据阈值
  • 做出不同的操作

建立渐变

import cv2
import numpy
width = height = 200
img = numpy.zeros((height,width),dtype=numpy.uint8)
for num in range(width):
    img[:,num] = 0 + num /(width-1) * 255
cv2.imshow("1",img)
cv2.waitKey()
cv2.destroyAllWindows()

• 效果

• 可以改用matplotlib来输出

matplotlib

import cv2
import numpy
from matplotlib import pyplot as plt
width = height = 200
img = numpy.zeros((height,width),dtype=numpy.uint8)
for num in range(width):
    img[:,num] = 0 + num /(width-1) * 255
plt.imshow(img,'gray')
plt.show()

• 输出结果

具体效果

import cv2 as cv
import numpy
from matplotlib import pyplot as plt
width = height = 200
img = numpy.zeros((height,width),dtype=numpy.uint8)
for num in range(width):
    img[:,num] = 0 + num /(width-1) * 255
ret,thresh1 = cv.threshold(img,127,255,cv.THRESH_BINARY)
ret,thresh2 = cv.threshold(img,127,255,cv.THRESH_BINARY_INV)
ret,thresh3 = cv.threshold(img,127,255,cv.THRESH_TRUNC)
ret,thresh4 = cv.threshold(img,127,255,cv.THRESH_TOZERO)
ret,thresh5 = cv.threshold(img,127,255,cv.THRESH_TOZERO_INV)
titles = ['Original Image','BINARY','BINARY_INV','TRUNC','TOZERO','TOZERO_INV']
images = [img, thresh1, thresh2, thresh3, thresh4, thresh5]
for i in range(6):
    plt.subplot(2,3,i+1),plt.imshow(images[i],'gray',vmin=0,vmax=255)
    plt.title(titles[i])
    plt.xticks([]),plt.yticks([])
plt.show()

• 执行结果

• 具体什么意义呢？

效果

• 具体公式

这个阈值设置为多少合适呢？

import numpy as np
import cv2
src = cv2.imread("./gear.jpg", cv2.IMREAD_GRAYSCALE)
cv2.imshow("origin",src)

#最大值最小值和相应的位置
min, max, minLoc, maxLoc = cv2.minMaxLoc(src)
print("min: %.2f, max: %.2f"% (min, max))
print("min loc: ", minLoc)
print("max loc: ", maxLoc)

#均值和标准差
means, stddev = cv2.meanStdDev(src)
print("mean: %.2f, stddev: %.2f"% (means, stddev))
#二值化
src[np.where(src < means)] = 0
src[np.where(src > means)] = 255
cv2.imshow("final",src)

cv2.waitKey(0)
cv2.destroyAllWindows()

• 通过函数找到画面的均值
• 然后将阈值设置为均值

• 有没有什么自动的阈值呢？

自动阈值

import numpy as np
from matplotlib import pyplot as plt
import cv2 as cv

src = cv.imread("gear.jpg")
h, w = src.shape[:2]
gray = cv.cvtColor(src, cv.COLOR_BGR2GRAY)
ret, binary = cv.threshold(gray, 0, 255, cv.THRESH_BINARY | cv.THRESH_OTSU)
print("ret :", ret)
result = np.zeros([h, w*2, 3], dtype=src.dtype)
result[0:h,0:w,:] = src
result[0:h,w:2*w,:] = cv.cvtColor(binary, cv.COLOR_GRAY2BGR)
cv.putText(result, "input", (10, 30), cv.FONT_ITALIC, 1.0, (0, 0, 255), 2)
cv.putText(result, "binary, threshold = " + str(ret), (w+10, 30), cv.FONT_ITALIC, 1.0, (0, 0, 255), 2)
plt.imshow(result)
plt.show()
cv.imwrite("binary_result.png", result)
cv.waitKey(0)
cv.destroyAllWindows()

• 效果

• 可以通过滑块控制阈值吗？

总结

• 这次研究了
  • 阈值
  • threshold
• 可以说是分界点
  • 就像二极管一样
  • 一分为二
  • 区分两种状态
    • 可感知的刺激
    • 阈下刺激

• 图像的阈值时根据亮度划分的
  • 可以是具体值
  • 也可以根据图像画面的均值
• 可以根据一个变化的数值来观看吗？🤔
• 下次再说👋