需求

 需要获取图片的红框的内容，实体的图片我就不放了

获取红框

先截取获得图片的多个轮廓

import cv2  
import numpy as np  # 加载图像并转换为灰度图像  
image = cv2.imread('image6.jpg')  
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)  # 应用高斯模糊以减少噪声  
blur = cv2.GaussianBlur(gray, (5, 5), 0)  # 应用HSV颜色空间转换  
hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)  
lower_red = np.array([0, 50, 50])  
upper_red = np.array([10, 255, 255])  
mask = cv2.inRange(hsv, lower_red, upper_red)  # 应用膨胀操作来放大边框内的内容和边框  
kernel = np.ones((5,5),np.uint8)  
dilated = cv2.dilate(mask,kernel,iterations = 1)  # 获取边界框坐标  
contours, hierarchy = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)  # 遍历每个轮廓并找到最大的红色边框  
max_contour = None  
max_area = 0  
for contour in contours:  area = cv2.contourArea(contour)  # if area > max_area:  #     max_contour = contour  #     max_area = area  x, y, w, h = cv2.boundingRect(contour)  # 裁剪图像以显示边界框内的内容及其周围10px内容  crop_image = image[max(y-10, 0):min(y+h+10, image.shape[0]), max(x-10, 0):min(x+w+10, image.shape[1])]  # 在裁剪后的图像上绘制红色矩形框以突出显示边界框内的内容及其周围10px内容  cv2.rectangle(crop_image, (max(x-10, 0), max(y-10, 0)), (min(x+w+10, image.shape[1]), min(y+h+10, image.shape[0])), (0, 0, 255), 2)  # 在裁剪后的图像上绘制红色矩形框以突出显示边界框内的内容及其周围10px内容  #cv2.imshow('Content with Border and Surrounding Area', crop_image)  # 显示带有红色边框和周围10px内容的裁剪后的图像  cv2.imwrite(f'red_border_{x}_{y}_{w}_{h}.jpg', crop_image)  cv2.waitKey(0)  cv2.destroyAllWindows()# 获取最大轮廓的边界框坐标  
# x, y, w, h = cv2.boundingRect(max_contour)  # # 裁剪图像以显示边界框内的内容及其周围10px内容  
# crop_image = image[max(y-10, 0):min(y+h+10, image.shape[0]), max(x-10, 0):min(x+w+10, image.shape[1])]  # # 在裁剪后的图像上绘制红色矩形框以突出显示边界框内的内容及其周围10px内容  
# cv2.rectangle(crop_image, (max(x-10, 0), max(y-10, 0)), (min(x+w+10, image.shape[1]), min(y+h+10, image.shape[0])), (0, 0, 255), 2)  # 在裁剪后的图像上绘制红色矩形框以突出显示边界框内的内容及其周围10px内容  
# cv2.imshow('Content with Border and Surrounding Area', crop_image)  # 显示带有红色边框和周围10px内容的裁剪后的图像  # cv2.waitKey(0)  
# cv2.destroyAllWindows()

识别红框

import cv2
import numpy as np# 加载图像
image = cv2.imread('red_border_1038_1886_6_6.jpg')# 将图像转换为灰度
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)# 二值化图像
_, threshold = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)# 找到图像中的轮廓
contours, _ = cv2.findContours(threshold, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)# 遍历每个轮廓，判断是否是闭合的圆
for contour in contours:# 进行轮廓近似，获取近似的多边形轮廓epsilon = 0.01 * cv2.arcLength(contour, True)approx = cv2.approxPolyDP(contour, epsilon, True)# 计算近似轮廓的周长approx_length = cv2.arcLength(approx, True)# 计算原始轮廓的周长contour_length = cv2.arcLength(contour, True)# 判断近似轮廓的周长是否接近于原始轮廓的周长if approx_length >= 0.9 * contour_length:# 绘制闭合的圆cv2.drawContours(image, [approx], -1, (0, 255, 0), 2)cv2.putText(image, 'Closed Circle', (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 0), 2)print("存在")# 显示结果图像
cv2.imshow('Result', image)
cv2.waitKey(0)
cv2.destroyAllWindows()