OpenCV3.0+Python3.6實(shí)現(xiàn)特定顏色的物體追蹤

更新時(shí)間：2019年07月23日 10:13:12 作者：qq_20156437

這篇文章主要為大家詳細(xì)介紹了OpenCV3.0+Python3.6實(shí)現(xiàn)特定顏色的物體追蹤，具有一定的參考價(jià)值，感興趣的小伙伴們可以參考一下

一、環(huán)境

win10、Python3.6、OpenCV3.x；編譯器：pycharm5.0.3

二、實(shí)現(xiàn)目標(biāo)

根據(jù)需要追蹤的物體顏色，設(shè)定閾值，在視頻中框選出需要追蹤的物體。

三、實(shí)現(xiàn)步驟

1）根據(jù)需要追蹤的物體顏色，設(shè)定顏色閾值，獲取追蹤物體的掩膜

代碼：generate_threshold.py

# -*- coding : utf-8 -*-
# Author: Tom Yu
import cv2
import numpy as np
 
cap = cv2.VideoCapture(0)#獲取攝像頭圖像
# img = cv2.imread("timg1.jpg")
# hsv_img = cv2.cvtColor(img,cv2.COLOR_BGR2HSV)
 
def nothing(x):
  pass
def createbars():
  """
  實(shí)現(xiàn)創(chuàng)建六個(gè)滑塊的作用，分別控制H、S、V的最高值與最低值
  """
  cv2.createTrackbar("H_l","image",0,180,nothing)
  cv2.createTrackbar("H_h","image",0,180,nothing)
  cv2.createTrackbar("S_l","image",0,255,nothing)
  cv2.createTrackbar("S_h","image",0,255,nothing)
  cv2.createTrackbar("V_l","image",0,255,nothing)
  cv2.createTrackbar("V_h","image",0,255,nothing)
cv2.namedWindow("image")
createbars()#創(chuàng)建六個(gè)滑塊
 
lower = np.array([0,0,0])#設(shè)置初始值
upper = np.array([0,0,0])
while True:
  ret,frame = cap.read()
  hsv_frame = cv2.cvtColor(frame,cv2.COLOR_BGR2HSV)#將圖片由BGR顏色空間轉(zhuǎn)化成HSV空間，HSV可以更好地分割顏色圖形
  lower[0]=cv2.getTrackbarPos("H_l","image")#獲取"H_l"滑塊的實(shí)時(shí)值
  upper[0]=cv2.getTrackbarPos("H_h","image")#獲取"H_h"滑塊的實(shí)時(shí)值
  lower[1]=cv2.getTrackbarPos("S_l","image")
  upper[1]=cv2.getTrackbarPos("S_h","image")
  lower[2]=cv2.getTrackbarPos("V_l","image")
  upper[2]=cv2.getTrackbarPos("V_h","image")
 
  mask = cv2.inRange(hsv_frame,lower,upper)#cv2.inrange()函數(shù)通過(guò)設(shè)定的最低、最高閾值獲得圖像的掩膜
  cv2.imshow("img",frame)
  cv2.imshow("mask",mask)
  if cv2.waitKey(1)&0xff == 27:
    break
 
cv2.destroyAllWindows()

實(shí)現(xiàn)效果：獲取需要追蹤的物體顏色閾值

2）根據(jù)獲取到的閾值，設(shè)定閾值范圍，在視頻中追蹤特定顏色的物體并用框選框出所需追蹤的物體

代碼：tracking_object.py

# -*- coding : utf-8 -*-
# Author: Tom Yu
import cv2
import numpy as np
 
cap = cv2.VideoCapture(0)#獲取攝像頭視頻
 
while True:
  ret,frame = cap.read()#讀取每一幀圖片
  hsv_frame = cv2.cvtColor(frame,cv2.COLOR_BGR2HSV)#將每一幀圖片轉(zhuǎn)化HSV空間顏色
  """
  依據(jù)之前的腳本獲取的閾值設(shè)置最高值與最低值
  """
  lower = np.array([0,104,205])
  upper = np.array([15,208,255])
 
  mask = cv2.inRange(hsv_frame,lower,upper)
  img,conts,hier = cv2.findContours(mask,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)#找出邊界
  cv2.drawContours(frame,conts,-1,(0,255,0),3)#畫(huà)出邊框
  dst = cv2.bitwise_and(frame,frame,mask=mask)#對(duì)每一幀進(jìn)行位與操作，獲取追蹤圖像的顏色
  #cv2.imshow("mask",mask)
  #cv2.imshow("dst",dst)
  cv2.imshow("frame",frame)
  if cv2.waitKey(1)&0xff == 27:
    break
 
cv2.destroyAllWindows()

實(shí)現(xiàn)效果：