最新的国产成人精品2020,免费久久国产精品无码一区

在計算機視覺中，顏色提取是一種常用的圖像處理技術，其中HSV（色相、飽和度、明度）是一種常見的顏色空間，用于描述和提取圖像中的顏色信息。

HSV顏色空間將顏色表示為三個分量：色相（H）、飽和度（S）和明度（V）。

?色相（Hue）：色相是顏色的類型或類別，描述了顏色在光譜中的位置。在HSV模型中，色相表示為一個角度值，通常在0到360度之間，將整個顏色光譜分為不同的顏色區(qū)域，如紅、橙、黃、綠、藍、紫等。

?飽和度（Saturation）：飽和度表示顏色的純度或強度，即顏色的深淺程度。飽和度為0時，顏色變?yōu)榛译A；飽和度為最大值時，顏色呈現(xiàn)出最鮮艷的狀態(tài)。飽和度的取值范圍通常在0到1之間，也可以表示為0%到100%。

?明度（Value）：明度表示顏色的明亮度，即顏色的亮度和深度。明度為0時，顏色為黑色；明度為最大值時，顏色為白色。明度的取值范圍通常在0到1之間，也可以表示為0%到100%。

通過這三個參數(shù)的組合，可以在HSV模型中精確地描述各種顏色，使得調整和理解顏色更加直觀和容易。在圖像處理中，HSV模型常用于調整圖像的顏色平衡、色調和亮度等方面。

硬件環(huán)境

操作環(huán)境及軟硬件配置如下：

?OriginEye 智能相機

?PC：Ubuntu (≥20.04) + ROS2 (≥Foxy)

代碼

 import cv2
  import numpy as np
  import rclpy
  from rclpy.node import Node
  from sensor_msgs.msg import Image
  from cv_bridge import CvBridge
  from collections import deque
  import textwrap


  class GetHsv(Node):
    def __init__(self):
      super().__init__('get_hsv')
      self.get_logger().info("Start get HSV")
      self.bridge = CvBridge()
      self.image_sub = self.create_subscription(Image, '/image_raw', self.image_callback, 10)
      self.pub = self.create_publisher(Image, '/camera/process_image', 10)
      self.count = 0
      self.box_width = 60
      self.collect_times = 300
      self.HSV_value = deque(maxlen=self.collect_times)
      self.H_range, self.S_range, self.V_range = 10, 80, 80
      self.lower_HSV = None
      self.upper_HSV = None


    def image_callback(self, msg):
      image = self.bridge.imgmsg_to_cv2(msg, 'bgr8')


      if self.count < self.collect_times:
        self.draw_rectangle(image)
      elif self.count < self.collect_times + self.collect_times:
        self.collect_hsv_value(image)
      else:
        self.save_hsv_value(image)


      self.count += 1
      self.pub.publish(self.bridge.cv2_to_imgmsg(image, 'bgr8'))


    def draw_rectangle(self, image):
      text = 'please put the color being tested in the rectangle box!'
      self.add_text(image, text)
      rect_x = (image.shape[1] - self.box_width) // 2
      rect_y = (image.shape[0] - self.box_width) // 2
      cv2.rectangle(image, (rect_x, rect_y), (
          rect_x + self.box_width, rect_y + self.box_width), (0, 255, 0), 3)


    def collect_hsv_value(self, image):
      text = 'HSV_value is collecting!'
      self.add_text(image, text)
      frame_x = (image.shape[1] - self.box_width) // 2
      frame_y = (image.shape[0] - self.box_width) // 2
      frame = image[frame_y:frame_y + self.box_width, frame_x:frame_x + self.box_width]
      self.HSV_value.append(self.calculate_mean_hsv(frame))


    def save_hsv_value(self, image):
      text = 'HSV_value is collected!'
      self.add_text(image,text)
      mean_HSV = np.mean(self.HSV_value, axis=0)
      self.lower_HSV, self.upper_HSV = self.calculate_hsv_range(mean_HSV)
      self.write_hsv_to_file(self.lower_HSV, self.upper_HSV)


    def calculate_mean_hsv(self, img):
      hsv_image = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
      return np.mean(hsv_image, axis=(0, 1))


    def calculate_hsv_range(self, HSV_value):
      lower_H = np.clip(int(HSV_value[0] - self.H_range), 0, 180)
      upper_H = np.clip(int(HSV_value[0] + self.H_range), 0, 180)
      lower_S = np.clip(int(HSV_value[1] - self.S_range), 40, 255)
      upper_S = np.clip(int(HSV_value[1] + self.S_range), 40, 255)
      lower_V = np.clip(int(HSV_value[2] - self.V_range), 40, 255)
      upper_V = np.clip(int(HSV_value[2] + self.V_range), 40, 255)
      return np.array([lower_H, lower_S, lower_V]), np.array([upper_H, upper_S, upper_V])


    def write_hsv_to_file(self, lower_HSV, upper_HSV):
      content = f"block_HSV is : {lower_HSV[0]},
        {lower_HSV[1]},{lower_HSV[2]} {upper_HSV[0]},{upper_HSV[1]},{upper_HSV[2]}
"
      filename = "/userdata/dev_ws/color_block_HSV.txt"
      with open(filename, "w+") as f:
        f.write(content)


      self.get_logger().info(f"HSV value has been saved in {filename}")
      self.get_logger().info(content)


    def add_text(self, image, text):
      text = textwrap.wrap(text, width = 80)
      text_y = image.shape[0] // 2 - 30
      for line in text:
        text_size, _ = cv2.getTextSize(line, cv2.FONT_HERSHEY_SIMPLEX, 1.2, 2)
        text_x = (image.shape[1] - text_size[0]) // 2
        cv2.putText(image, line, (text_x, text_y), 
          cv2.FONT_HERSHEY_SIMPLEX, 1.2, (0, 255, 0), 2, cv2.LINE_AA)


  def main(args=None):
    rclpy.init(args=args)
    get_hsv = GetHsv()
    rclpy.spin(get_hsv)
    get_hsv.destroy_node()
    rclpy.shutdown()


  if __name__ == '__main__':
    main()