opencv如何填充阈值图像（一些白色部分）？

发布于 2025-01-18 03:05:06 字数 992 浏览 2 评论 0原文

我创建了一个脚本来阈值微观图像，我想知道如何填充图像的某些白色空间。如您在图片中所见，我的目的是制作一个过滤器，以仅分开蓝色区域。应将所有其他白色区域（红色箭头）擦除或转换为黑色。

这是我的代码：

original = cv2.imread(inputfile + ".jpg")
#cv2.imread(inputfile + ".jpg")
original = cv2.resize(original, (1280, 960)) #resize of original image
img = cv2.cvtColor(original, cv2.COLOR_BGR2GRAY)
median = cv2.medianBlur(img, 3)
ret, th = cv2.threshold(median, 160, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)
kernel = np.ones((1,1), np.uint8)
opening = cv2.morphologyEx(th, cv2.MORPH_OPEN, kernel)
edge_touching_removed = clear_border(opening)
contours, hierarchy = cv2.findContours(edge_touching_removed, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

事实是...应用此阈值后，我将在蓝色区域应用轮廓以使区域陷入困境。没关系...但是我不知道如何擦除其他区域。

我应该导出该阈值图像，然后手动绘画？我不想这样做...我在想也许有手动填写代码。我使用了CV2形态，但我不知道我是否需要为此进行另一个过滤器。

谢谢

原文

I<ve created a script to threshold a microscopic image and I would like to know how to fill some white spaces of the image. As you see in the picture my aim is to make a filter to separate only the blue zones. All other white zones (red arrow) should be erased or converted (filled out) to black.

Here is my code:

original = cv2.imread(inputfile + ".jpg")
#cv2.imread(inputfile + ".jpg")
original = cv2.resize(original, (1280, 960)) #resize of original image
img = cv2.cvtColor(original, cv2.COLOR_BGR2GRAY)
median = cv2.medianBlur(img, 3)
ret, th = cv2.threshold(median, 160, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)
kernel = np.ones((1,1), np.uint8)
opening = cv2.morphologyEx(th, cv2.MORPH_OPEN, kernel)
edge_touching_removed = clear_border(opening)
contours, hierarchy = cv2.findContours(edge_touching_removed, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

The thing is ... after apply this threshold I will apply contours to the blue zone to caculate area. That is ok... but the thing is I don<t know how to erase the others zones.

Should I export that threshold image and I painted manually? I dont wanna do that... I was thinking that maybe there is a manual fill out code. I used the cv2 morphology but I dont know if I need to do another filter for that.

Thanks