目标识别

目标识别

图像轮廓

图像轮廓：具有相同颜色或强度的连续点的曲线
作用：
	可用于图形分析
	物体的识别和检测
注意点：
	为了检测的准确性，需要先对图像进行二值化或Canny操作
	画轮廓时会修改输入的图像

findContours(img,mode,ApproximationMode)
	mode:模式
		RETR_EXTERNAL=0表示只检测外轮廓
		RETR_LIST=1检测的轮廓不建立等级关系
		RETR_CCOMP=2每层最多两级
		RETR_TREE=3按树形存储轮廓
	ApproximationMode
		CHAIN_APPROX_NONE:保存所有轮廓上的点
		CHAIN_APPROX_SIMPLE只保存角点
		
		
import cv2
import numpy as np

img = cv2.imread('C:\\Users\dazhi\Desktop\lunkuo.png')
###转变成单通道
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

##二值化
ret,binary = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

###轮廓查找
contours,hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
print(contours)

cv2.imshow('img',img)

# cv2.imshow('bin',binary)

cv2.waitKey(0)

绘制轮廓

drawContours(img,contours,contourIdx,color,thickness..)
	contourIdx -1表示绘制所有轮廓
	color 颜色(0,0,255)
	thickness,线宽，-1是全部填充
	
import cv2
import numpy as np

img = cv2.imread('C:\\Users\dazhi\Desktop\lunkuo.png')
###转变成单通道
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

##二值化
ret,binary = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

###轮廓查找
contours,hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
# print(contours)

###绘制轮廓
cv2.drawContours(img,contours,-1,(0,0,255),1)

cv2.imshow('img',img)

# cv2.imshow('bin',binary)

cv2.waitKey(0)

轮廓的面积和周长

###轮廓的面积
contourArea(contour)
	contour:轮廓
###轮廓的周长
arcLength(curve,closed)
	curve:轮廓
	closed：闭合还是非闭合
	
import cv2
import numpy as np

img = cv2.imread('C:\\Users\dazhi\Desktop\lunkuo.png')
###转变成单通道
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

##二值化
ret,binary = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

###轮廓查找
contours,hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
# print(contours)

###绘制轮廓
cv2.drawContours(img,contours,-1,(0,0,255),1)

###计算面积
area = cv2.contourArea(contours[0])
print("area=%d"%(area))


###计算周长
len = cv2.arcLength(contours[0],True)
print("len=%d"%(len))

cv2.imshow('img',img)

# cv2.imshow('bin',binary)

cv2.waitKey(0)

多边形逼近与凸包

###多边形逼近

approxPolyDP(curve,epsilon,closed)
	curve:曲线，轮廓
	epsilon：精度
	closed：是否闭合
	
###凸包
convexHull(points,clockwise)
	points:点，轮廓
	clockwise：顺时针绘制
	
import cv2
import numpy as np

def darwShape(src,points):
    i = 0
    ###画线
    while i <len(points):
        if(i == len(points) -1):
            x,y = points[i][0]
            x1,y1 = points[0][0]
            cv2.line(src,(x,y),(x1,y1),(0,0,255),3)
        else:
            x,y = points[i][0]
            x1,y1 = points[i+1][0]
            cv2.line(src,(x,y),(x1,y1),(0,0,255),3)
        i=i+1


img = cv2.imread('C:\\Users\dazhi\Desktop\-tubao.png')
###转变成单通道
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

##二值化
ret,binary = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

###轮廓查找
contours,hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
# print(contours)

###绘制轮廓
cv2.drawContours(img,contours,0,(0,255,0),1)

###多边形逼近
e = 20  ###e值变小线会变得平滑
approx = cv2.approxPolyDP(contours[0],e,True)
darwShape(img,approx)

###凸包
hull = cv2.convexHull(contours[0])
darwShape(img,hull)

cv2.imshow('img',img)

cv2.waitKey(0)

外接矩阵

最小外接矩形
	minAreaRect(points)
		points:坐标点。轮廓
最大外接矩形
	boundingRect(array)
		array:轮廓

import cv2
import numpy as np

def darwShape(src,points):
    i = 0
    ###画线
    while i <len(points):
        if(i == len(points) -1):
            x,y = points[i][0]
            x1,y1 = points[0][0]
            cv2.line(src,(x,y),(x1,y1),(0,0,255),3)
        else:
            x,y = points[i][0]
            x1,y1 = points[i+1][0]
            cv2.line(src,(x,y),(x1,y1),(0,0,255),3)
        i=i+1


img = cv2.imread('C:\\Users\dazhi\Desktop\Hello.png')
###转变成单通道
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

##二值化
ret,binary = cv2.threshold(gray,150,255,cv2.THRESH_BINARY)

###轮廓查找
contours,hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
# print(contours)

###最小外接矩形
r = cv2.minAreaRect(contours[1])
###由于上面的是带角度的，下面只取起始点和宽高
box = cv2.boxPoints(r)
###强制转换
box = np.int0(box)
cv2.drawContours(img,[box],0,(0,0,255),2)

####最大外接矩形
x,y,w,h = cv2.boundingRect(contours[1])
###(x+w,y+h)在起始点的基础上加上
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)

cv2.imshow('img',img)

cv2.waitKey(0)

车辆识别

涉及到的内容
	窗口的展示
	图像/视频的加载
	基本图形的绘制
	车辆识别
		基本图像运算与处理
		形态学
		轮廓查找
		

###视频加载
import cv2
import numpy as np

cap = cv2.VideoCapture('C://Users/dazhi/Desktop/video.mp4')

###通过while循环将视频一幅一幅的读取出来(读取每一帧)
while True:
    ret,frame = cap.read()

    if(ret == True):
        cv2.imshow('video',frame)
    
    key = cv2.waitKey(1)
    ###27=左上角esc
    if(key == 27):
        break

cap.release()
cv2.destroyAllWindows()

###去除背景
createBackgroundSubtractorMOG()
	history=200 缓冲5帧

import cv2
import numpy as np

cap = cv2.VideoCapture('C://Users/dazhi/Desktop/video.mp4')

bgsubmog = cv2.createBackgroundSubtractorMOG2()
# test = cv2.createBackgroundSubtractorMOG2()
# bgsubmog = cv2.bgsegm.createBackgroundSubtractorGMG()
###通过while循环将视频一幅一幅的读取出来(读取每一帧)
while True:
    ret,frame = cap.read()

    if(ret == True):

        ###灰度化
        cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)

        ###去噪(高斯)
        blur = cv2.GaussianBlur(frame,(3,3),5)
        ###去背景
        mask = bgsubmog.apply(blur)
        ###做对比 没有经过去噪的
        # t1 = test.apply(frame)
        # cv2.imshow('t1',t1)

        cv2.imshow('video',mask)
    
    key = cv2.waitKey(1)
    ###27=左上角esc
    if(key == 27):
        break

cap.release()
cv2.destroyAllWindows()

###形态处理
import cv2
import numpy as np

cap = cv2.VideoCapture('C://Users/dazhi/Desktop/video.mp4')

bgsubmog = cv2.createBackgroundSubtractorMOG2()
# test = cv2.createBackgroundSubtractorMOG2()
# bgsubmog = cv2.bgsegm.createBackgroundSubtractorGMG()
###通过while循环将视频一幅一幅的读取出来(读取每一帧)

###形态学kernel
kernel = cv2.getStructuringElement(cv2.MORPH_RECT,(5,5))

while True:
    ret,frame = cap.read()

    if(ret == True):

        ###灰度化
        cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)

        ###去噪(高斯)
        blur = cv2.GaussianBlur(frame,(3,3),5)
        ###去背景
        mask = bgsubmog.apply(blur)
        ###做对比 没有经过去噪的
        # t1 = test.apply(frame)
        # cv2.imshow('t1',t1)

        ##腐蚀(去掉一些小的斑块)
        erode = cv2.erode(mask,kernel)
        # cv2.imshow('erode',erode)

        ###膨胀（还原放大）
        dilate = cv2.dilate(erode,kernel,iterations=3)
        # cv2.imshow('erode',dilate)

        ###闭操作 去掉物体内部的小块
        close = cv2.morphologyEx(dilate,cv2.MORPH_CLOSE,kernel)
        close = cv2.morphologyEx(close,cv2.MORPH_CLOSE,kernel)
        # cv2.imshow('erode',close) 

        cnts,h = cv2.findContours(close,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)

        for (i,c) in enumerate(cnts):
            (x,y,w,h) = cv2.boundingRect(c)
            cv2.rectangle(frame,(x,y),(x+w,y+h),(0,0,255),2)

        cv2.imshow('video',frame)
    
    key = cv2.waitKey(1)
    ###27=左上角esc
    if(key == 27):
        break

cap.release()
cv2.destroyAllWindows()

###车辆统计
import cv2
import numpy as np

###检测线的高度 （可以在灰度那里打印出(720, 1280, 3)）
line_high = 600

###存放有效车辆的数组
cars = []

###统计车的数量
carno = 0
###计算中心点

def center(x,y,w,h):
    x1 = int(w/2)
    y1 = int(h/2)
    cx = x+x1
    cy = y+y1

    return cx,cy

cap = cv2.VideoCapture('C://Users/dazhi/Desktop/video.mp4')

bgsubmog = cv2.createBackgroundSubtractorMOG2()
# test = cv2.createBackgroundSubtractorMOG2()
# bgsubmog = cv2.bgsegm.createBackgroundSubtractorGMG()
###通过while循环将视频一幅一幅的读取出来(读取每一帧)

###形态学kernel
kernel = cv2.getStructuringElement(cv2.MORPH_RECT,(5,5))

while True:
    ret,frame = cap.read()

    if(ret == True):

        ###灰度化
        cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
        ###查看视频的数据
        # print(frame.shape)
        # exit()

        ###去噪(高斯)
        blur = cv2.GaussianBlur(frame,(3,3),5)
        ###去背景
        mask = bgsubmog.apply(blur)
        ###做对比 没有经过去噪的
        # t1 = test.apply(frame)
        # cv2.imshow('t1',t1)

        ##腐蚀(去掉一些小的斑块)
        erode = cv2.erode(mask,kernel)
        # cv2.imshow('erode',erode)

        ###膨胀（还原放大）
        dilate = cv2.dilate(erode,kernel,iterations=3)
        # cv2.imshow('erode',dilate)

        ###闭操作 去掉物体内部的小块
        close = cv2.morphologyEx(dilate,cv2.MORPH_CLOSE,kernel)
        close = cv2.morphologyEx(close,cv2.MORPH_CLOSE,kernel)
        # cv2.imshow('erode',close) 

        cnts,h = cv2.findContours(close,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)

        ###画一条线
        cv2.line(frame,(10,line_high),(1200,line_high),(255,255,0),3)
        for (i,c) in enumerate(cnts):
            (x,y,w,h) = cv2.boundingRect(c)
            

            ###过滤(对车辆的宽高进行判断，已验证是否是有效的车辆)
            isValid = (w >= 90) and (h >= 90)
            if(not isValid):
                continue
            ###到这里是有效的车

            cv2.rectangle(frame,(x,y),(x+w,y+h),(0,0,255),2) ##绘制车
            cpoint = center(x,y,w,h)
            cars.append(cpoint)

            for(x,y) in cars:
                 ##要有一条线
                ##有范围 6
                ##从数组中减去

                if ((y > line_high -6) and (y < line_high + 6)):
                    carno += 1
                    cars.remove((x,y))
                    print(carno)
               
        cv2.imshow('video',frame)
    
    key = cv2.waitKey(1)
    ###27=左上角esc
    if(key == 27):
        break

cap.release()
cv2.destroyAllWindows()

###显示车辆统计信息
import cv2
import numpy as np

###检测线的高度 （可以在灰度那里打印出(720, 1280, 3)）
line_high = 550

###存放有效车辆的数组
cars = []

###统计车的数量
carno = 0
###计算中心点
def center(x,y,w,h):
    x1 = int(w/2)
    y1 = int(h/2)
    cx = x+x1
    cy = y+y1
    return cx,cy

cap = cv2.VideoCapture('C://Users/dazhi/Desktop/video.mp4')

bgsubmog = cv2.createBackgroundSubtractorMOG2()
# bgsubmog = cv2.bgsegm.createBackgroundSubtractorGMG()
###通过while循环将视频一幅一幅的读取出来(读取每一帧)

###形态学kernel
kernel = cv2.getStructuringElement(cv2.MORPH_RECT,(5,5))

while True:
    ret,frame = cap.read()

    if(ret == True):

        ###灰度化
        cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)

        ###去噪(高斯)
        blur = cv2.GaussianBlur(frame,(3,3),5)
        ###去背景
        mask = bgsubmog.apply(blur)

        ##腐蚀(去掉一些小的斑块)
        erode = cv2.erode(mask,kernel)
        # cv2.imshow('erode',erode)

        ###膨胀（还原放大）
        dilate = cv2.dilate(erode,kernel,iterations=3)
        # cv2.imshow('erode',dilate)

        ###闭操作 去掉物体内部的小块
        close = cv2.morphologyEx(dilate,cv2.MORPH_CLOSE,kernel)
        close = cv2.morphologyEx(close,cv2.MORPH_CLOSE,kernel)
        # cv2.imshow('erode',close) 

        cnts,h = cv2.findContours(close,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)

        ###画一条线
        cv2.line(frame,(10,line_high),(1200,line_high),(255,255,0),3)
        for (i,c) in enumerate(cnts):
            (x,y,w,h) = cv2.boundingRect(c)
            

            ###过滤(对车辆的宽高进行判断，已验证是否是有效的车辆)
            isValid = (w >= 90) and (h >= 90)
            if(not isValid):
                continue
            ###到这里是有效的车

            cv2.rectangle(frame,(x,y),(x+w,y+h),(0,0,255),2) ##绘制车
            cpoint = center(x,y,w,h)
            cars.append(cpoint)

            for(x,y) in cars:
                 ##要有一条线
                ##有范围 6
                ##从数组中减去
                if ((y > line_high - 7) and (y < line_high + 7)):
                    carno += 1
                    cars.remove((x,y))
                    print(carno)
                    
        ###将信息打印出来       
        cv2.putText(frame,"Cars count:"+str(carno),(500,60),cv2.FONT_HERSHEY_SIMPLEX,2,(255,0,0),5)
        cv2.imshow('video',frame)
    
    key = cv2.waitKey(1)
    ###27=左上角esc
    if(key == 27):
        break

cap.release()
cv2.destroyAllWindows()