基于OpenCV实现对图片及视频中感兴趣区域颜色识别

基于OpenCV实现图片及视频中选定区域颜色识别

近期，需要实现检测摄像头中指定坐标区域内的主体颜色，通过查阅大量相关的内容，最终实现代码及效果如下，具体的实现步骤在代码中都详细注释，代码还可以进一步优化，但提升有限。

主要实现过程：按不同颜色的取值范围，对图像进行循环遍历，转换为灰度图，将本次遍历的颜色像素转换为白色，对白色部分进行膨胀处理，使其更加连续，计算白色部分外轮廓包围的面积累加求和，比较每种颜色围起来面积，保存最大值及其颜色，所有颜色遍历完后，返回最大值对应的颜色，显示在图像上

如果有类似的颜色识别的任务，可参考以下代码修改后实现具体需求

colorList.py

import numpy as np
import collections

# 将rgb图像转换为hsv图像后，确定不同颜色的取值范围
def getColorList():
    dict = collections.defaultdict(list)

    # black
    lower_black = np.array([0, 0, 0])
    upper_black = np.array([180, 255, 46])
    color_list_black = []
    color_list_black.append(lower_black)
    color_list_black.append(upper_black)
    dict['black'] = color_list_black

    # gray
    lower_gray = np.array([0, 0, 46])
    upper_gray = np.array([180, 43, 220])
    color_list_gray= []
    color_list_gray.append(lower_gray)
    color_list_gray.append(upper_gray)
    dict['gray'] = color_list_gray

    # white
    lower_white = np.array([0, 0, 221])
    upper_white = np.array([180, 30, 255])
    color_list_white = []
    color_list_white.append(lower_white)
    color_list_white.append(upper_white)
    dict['white'] = color_list_white

    # red
    lower_red = np.array([156, 43, 46])
    upper_red = np.array([180, 255, 255])
    color_list_red = []
    color_list_red.append(lower_red)
    color_list_red.append(upper_red)
    dict['red'] = color_list_red

    # red2
    lower_red = np.array([0, 43, 46])
    upper_red = np.array([10, 255, 255])
    color_list_red2 = []
    color_list_red2.append(lower_red)
    color_list_red2.append(upper_red)
    dict['red2'] = color_list_red2

    # orange
    lower_orange = np.array([11, 43, 46])
    upper_orange = np.array([25, 255, 255])
    color_list_orange = []
    color_list_orange.append(lower_orange)
    color_list_orange.append(upper_orange)
    dict['orange'] = color_list_orange

    # yellow
    lower_yellow = np.array([26, 43, 46])
    upper_yellow = np.array([34, 255, 255])
    color_list_yellow = []
    color_list_yellow.append(lower_yellow)
    color_list_yellow.append(upper_yellow)
    dict['yellow'] = color_list_yellow

    # green
    lower_green = np.array([35, 43, 46])
    upper_green = np.array([77, 255, 255])
    color_list_green = []
    color_list_green.append(lower_green)
    color_list_green.append(upper_green)
    dict['green'] = color_list_green

    # cyan
    lower_cyan = np.array([78, 43, 46])
    upper_cyan = np.array([99, 255, 255])
    color_list_cyan = []
    color_list_cyan.append(lower_cyan)
    color_list_cyan.append(upper_cyan)
    dict['cyan'] = color_list_cyan

    # blue
    lower_blue = np.array([100, 43, 46])
    upper_blue = np.array([124, 255, 255])
    color_list_blue = []
    color_list_blue.append(lower_blue)
    color_list_blue.append(upper_blue)
    dict['blue'] = color_list_blue

    # purple
    lower_purple = np.array([125, 43, 46])
    upper_purple = np.array([155, 255, 255])
    color_list_purple = []
    color_list_purple.append(lower_purple)
    color_list_purple.append(upper_purple)
    dict['purple'] = color_list_purple

    return dict

if __name__ == '__main__':
    color_dict = getColorList()
    print(color_dict)

    num = len(color_dict)
    print('num=', num)

    for d in color_dict:
        print('key=', d)
        print('value=', color_dict[d][1])

image_color_realize.py

import cv2
import colorList

# 实现对图片中目标区域颜色的识别
def get_color(frame):
    print('go in get_color')
    hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
    maxsum = 0
    color = None
    color_dict = colorList.getColorList()

    # count = 0

    for d in color_dict:
        mask = cv2.inRange(hsv, color_dict[d][0], color_dict[d][1])  # 在后两个参数范围内的值变成255
        
        # mask = cv2.medianBlur(mask, 3)  # 中值滤波

        binary = cv2.threshold(mask, 127, 255, cv2.THRESH_BINARY)[1]  # 在灰度图片中，像素值大于127的都变成255，[1]表示调用图像，也就是该函数第二个返回值

        # cv2.imshow("0",binary)
        # cv2.waitKey(0)
        # count+=1

        binary = cv2.dilate(binary, None, iterations=2)  # 使用默认内核进行膨胀操作，操作两次，使缝隙变小，图像更连续
        cnts = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)[-2]  # 获取该函数倒数第二个返回值轮廓
        sum = 0
        for c in cnts:
            sum += cv2.contourArea(c)  # 获取该颜色所有轮廓围成的面积的和
        # print("%s  , %d" %(d, sum ))
        if sum > maxsum:
            maxsum = sum
            color = d
            if color == 'red2':
                color = 'red'
            elif color == 'orange':
                color = 'yellow'
            elif color == 'purple' or color == 'blue' or color == 'cyan' or color == 'white' or color == 'green':
                color = 'normal'
    return color

if __name__ == '__main__':
    filename = "C:/Users/admin/Desktop/water_samples/live01.jpg"
    frame = cv2.imread(filename)
    # frame = frame[180:280, 180:380]  # [y:y+h, x:x+w] 注意x,y顺序
    color = get_color(frame)

    # 绘制文本
    cv2.putText(img=frame,text=color,org=(20,50),fontFace=cv2.FONT_HERSHEY_SIMPLEX,
                fontScale=1.0,color=(0,255,0),thickness=2)

    # cv2.namedWindow('frame',cv2.WINDOW_NORMAL)  # 设置显示窗口可调节
    cv2.imshow('frame',frame)
    cv2.waitKey(0)

video_color_realize.py

import cv2
import xf_color


# 对视频或摄像头获取的影像目标区域颜色进行识别

cap = cv2.VideoCapture("C:/Users/admin/Desktop/water_samples/01.mp4")
# cap = cv2.VideoCapture(0)
cap.set(cv2.CAP_PROP_FRAME_WIDTH, 1100)  # 这里窗口大小调节只对摄像头有效
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 750)

while cap.isOpened():
    ret, frame0 = cap.read()
    # 对图像帧进行翻转（因为opencv图像和我们正常是反着的） 视频是正常的，摄像头是反转的
    # frame0 = cv2.flip(src=frame0, flipCode=2)

    # frame = frame[180:280, 180:380]  # [y:y+h, x:x+w]
    # frame = frame0[200:400, 100:300]  # 设置检测颜色的区域，四个顶点坐标
    frame = frame0

    # frame=cv2.resize(src=frame,dsize=(750,600))
    # hsv_frame = cv2.cvtColor(src=frame, code=cv2.COLOR_BGR2HSV)
    # 获取读取的帧的高宽
    height, width, channel = frame.shape
    # color = xf_color.get_color(hsv_frame)
    color = xf_color.get_color(frame)
    # 绘制文本
    cv2.putText(img=frame0, text=color, org=(20, 50), fontFace=cv2.FONT_HERSHEY_SIMPLEX,
                fontScale=1.0, color=(0, 255, 0), thickness=2)
    cv2.imshow('frame', frame0)
    key = cv2.waitKey(1)
    if key == 27:
        break

cap.release()
cv2.destroyAllWindows()


if __name__ == '__main__':
    print('Pycharm')

效果如下：

示例图片1

示例图片2

示例图片3