Opencv基本操作（基于Python）

鍪韓瞿

已于 2022-09-07 14:04:34 修改

阅读量933

点赞数 1

分类专栏： cv 文章标签： opencv python 计算机视觉

于 2022-09-07 13:59:09 首次发布

本文链接：https://blog.csdn.net/qq_39644822/article/details/126743824

版权

cv 专栏收录该内容

1 篇文章 0 订阅

订阅专栏

显示图片

img2 = cv2.imread('cat.jpg', cv2.IMREAD_COLOR)
img3 = cv2.imread('cat.jpg', cv2.IMREAD_GRAYSCALE) # 灰度
cv2.show(img2)
cv2.waitKey(0)
cv2.destroyAllWindows()

显示视频

vc = cv2.VideoCapture('video.mp4')
if vc.isOpened():  # 检查是否打开正确
    open, frame = vc.read()  # open是bool值， frame是一帧
    print(open)
else:
    open = False
while open:
    ret, frame = vc.read()
    if frame is None:
        break
    if ret:
        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)  # 灰度处理
        cv2.imshow('result', gray)
        if cv2.waitKey(100) & 0xFF == 27:  # 退出键退出
            break
vc.release()
cv2.destroyAllWindows()

截取部分图像数据

img = cv2.imread('cat.jpg')
cat = img[0:100, 0:200]
cv2.imshow('cat', cat)
cv2.waitKey(500)

颜色通道提取

img = cv2.imread('cat.jpg')
b, g, r = cv2.split(img)
print(b, g, r)
# bgr合成
img = cv2.merge((b, g, r))
# 只保留r
cur_img = img.copy()
cur_img[:, :, 0] = 0
cur_img[:, :, 1] = 0
cv_show('cat', cur_img)

边界填充

img = cv2.imread('cat.jpg')
top_size, bottom_size, left_size, right_size = (50, 50, 50, 50)
replicate = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, borderType=cv2.BORDER_REPLICATE)
reflect = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_REFLECT)
reflect101 = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_REFLECT101)
wrap = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_WRAP)
constant = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_CONSTANT, value = 5)
# 放在图表里好比较
plt.subplot(231), plt.imshow(img, 'gray'), plt.title('ORIGINAL')
plt.subplot(232), plt.imshow(replicate, 'gray'), plt.title('REPLICATE')
plt.subplot(233), plt.imshow(reflect, 'gray'), plt.title('REFLECT')
plt.subplot(234), plt.imshow(reflect101, 'gray'), plt.title('REFLECT101')
plt.subplot(235), plt.imshow(wrap, 'gray'), plt.title('WRAP')
plt.subplot(236), plt.imshow(constant, 'gray'), plt.title('CONSTANT')
plt.show()

在这里插入图片描述

改变大小

res = cv2.resize(img, (0, 0), fx=2, fy=1)  # 长度扩大三倍
res2 = cv2.resize(img, (500, 500))  # 改变为括号的大小
plt.imshow(res2)
plt.show()
``
## 数值计算

```python
cat = cv2.imread("cat.jpg")
dog = cv2.imread("dog.jpg")
dog = cv2.resize(dog, (500, 414))  # 需要数据大小相同才能相加
# 两种不同的加法
cat_dog = cat + dog
cat_dog2 = cv2.add(cat, dog)
print(cat_dog[:5, :, 0])
print(cat_dog2[:5, :, 0])

图像融合

注意两张图片要大小一样

res = cv2.addWeighted(cat, 0.3, dog, 0.7, 0)  # z = x*a + y*b + c
cv_show("result", res)

腐蚀操作

img = cv2.imread('dige.png')
kernel = np.ones((3, 3), np.uint8)  # 可以理解为腐蚀框的大小 越大腐蚀越强
erosion = cv2.erode(img, kernel, iterations=1)  # iterations表示腐蚀次数
res = np.hstack((erosion, img))
cv_show('res', res)

膨胀操作

dilation = cv2.dilate(img, kernel, iterations=1)
res = np.hstack((dilation, img))
cv_show('res', res)

开运算与闭运算

开：先腐蚀再膨胀；闭：先膨胀再腐蚀

opening = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)
closing = cv2.morphologyEx(img, cv2.MORPH_CLOSE, kernel)
res = np.hstack((opening, closing))
cv_show('res', res)

梯度运算

梯度=膨胀-腐蚀

erosion = cv2.erode(img, kernel, iterations=1)
dilation = cv2.dilate(img, kernel, iterations=1)
res = dilation - erosion
gradient = cv2.morphologyEx(img, cv2.MORPH_GRADIENT, kernel, iterations=2)
res = np.hstack((res, gradient))
cv_show('res', res)

礼帽和黑帽

礼帽=原始输入-开运算结果；黑帽=闭运算-原始输入

tophat = cv2.morphologyEx(img, cv2.MORPH_TOPHAT, kernel)
blackhat = cv2.morphologyEx(img, cv2.MORPH_BLACKHAT, kernel)
res = np.hstack((tophat, blackhat))
cv_show('res', res)

图像梯度-sobel算子

矩阵与图片相乘，使图片上下或者左右进行运算用于得到边界
dst = cv2.sobel(src, ddepth, dx, dy, ksize)
分别对横向和纵行处理图像叠加比同时进行横向和纵向处理的效果要好

img = cv2.imread('lena.jpg', cv2.IMREAD_GRAYSCALE)
sobelx = cv2.Sobel(img, cv2.CV_64F, 1, 0, ksize=3)
sobelx = cv2.convertScaleAbs(sobelx)  # 补全另一半
sobely = cv2.Sobel(img, cv2.CV_64F, 0, 1, ksize=3)
sobely = cv2.convertScaleAbs(sobely)
sobelxy = cv2.addWeighted(sobelx, 0.5, sobely, 0.5, 0)
plt.subplot(221)
plt.imshow(cv2.cvtColor(sobelx, code=cv2.COLOR_BGR2RGB))
plt.subplot(222)
plt.imshow(cv2.cvtColor(sobely, code=cv2.COLOR_BGR2RGB))
plt.subplot(223)
plt.imshow(cv2.cvtColor(sobelxy, code=cv2.COLOR_BGR2RGB))
plt.show()

在这里插入图片描述

Scharr、laplacian、sobel算子比较

sobelx = cv2.Sobel(img, cv2.CV_64F, 1, 0, ksize=3)
sobelx = cv2.convertScaleAbs(sobelx)  # 补全另一半
sobely = cv2.Sobel(img, cv2.CV_64F, 0, 1, ksize=3)
sobely = cv2.convertScaleAbs(sobely)
sobelxy = cv2.addWeighted(sobelx, 0.5, sobely, 0.5, 0)
scharrx = cv2.Scharr(img, cv2.CV_64F, 1, 0)
scharrx = cv2.convertScaleAbs(scharrx)  # 补全另一半
scharry = cv2.Scharr(img, cv2.CV_64F, 0, 1)
scharry = cv2.convertScaleAbs(scharry)
scharrxy = cv2.addWeighted(scharrx, 0.5, scharry, 0.5, 0)
laplacian = cv2.Laplacian(img, cv2.CV_64F)
laplacian = cv2.convertScaleAbs(laplacian)  # 补全另一半
plt.subplot(221)
plt.imshow(sobelxy, 'gray')
plt.subplot(222)
plt.imshow(scharrxy, 'gray')
plt.subplot(223)
plt.imshow(laplacian, 'gray')
plt.show()

在这里插入图片描述

图像平滑处理

图像降噪

# 均值滤波 简单的平均卷积操作
blur = cv2.blur(img, (3, 3))
# 方框滤波 基本和均值一样，可以选择归一化，容易越界
box1 = cv2.boxFilter(img, -1, (3, 3))
box2 = cv2.boxFilter(img, -1, (3, 3), normalize=True)
# 高斯滤波 采用高斯分布，不以像素点位置为中心
gaussian = cv2.GaussianBlur(img, (5, 5), 1)
# 中值滤波 效果最好
median = cv2.medianBlur(img, 5)
# 画图对比一下
plt.subplot(231), plt.imshow(cv2.cvtColor(blur, code=cv2.COLOR_BGR2RGB))
plt.subplot(232), plt.imshow(cv2.cvtColor(box1, code=cv2.COLOR_BGR2RGB))
plt.subplot(233), plt.imshow(cv2.cvtColor(box2, cv2.COLOR_BGR2RGB))
plt.subplot(234), plt.imshow(cv2.cvtColor(gaussian, cv2.COLOR_BGR2RGB))
plt.subplot(235), plt.imshow(cv2.cvtColor(median, cv2.COLOR_BGR2RGB))
plt.show()

在这里插入图片描述

图像阈值

只能输入单通道图像

img = cv2.imread('cat.jpg', cv2.IMREAD_GRAYSCALE)
ret, thresh1 = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY)
ret, thresh2 = cv2.threshold(img, 127, 255, cv2.THRESH_OTSU)
ret, thresh3 = cv2.threshold(img, 127, 255, cv2.THRESH_MASK)
ret, thresh4 = cv2.threshold(img, 127, 255, cv2.THRESH_TRUNC)
ret, thresh5 = cv2.threshold(img, 127, 255, cv2.THRESH_TOZERO)
ret, thresh6 = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY_INV)
ret, thresh7 = cv2.threshold(img, 127, 255, cv2.THRESH_TRIANGLE)
ret, thresh8 = cv2.threshold(img, 127, 255, cv2.THRESH_TOZERO_INV)

images = [img, thresh1, thresh2, thresh3, thresh4, thresh5, thresh6, thresh7, thresh8]
titiles = ["Original","Binary", "Otsu", "Mask", "Trunc", "Tozero", "BinaryINV", "Triangle", "TozeroINV"]
for i in range(9):
    plt.subplot(3, 3, i+1)
    plt.imshow(cv2.cvtColor(images[i], cv2.COLOR_BGR2RGB))
    plt.title(titiles[i])
plt.show()