opencv計算機視覺學習筆記三

轉載來自https://blog.csdn.net/retacn_yue/article/details/53608377
第四章 深度估計和分割

1 捕獲深度攝像頭的幀

深度圖 灰度 每個像素都是攝像頭到物體表面的距離 毫米

點雲圖 彩色 每種顏色對應一個維度空間 米

視差圖 灰度 每個像素代表物體表面的立體視差 近大遠小

有效深度掩模一個給定像素的深度信息是否有效

2 從視差圖中得到掩模

!/usr/bin/env python

-- coding: utf-8 --

@Time : 2016/12/1 10:16

@Author : Retacn

@Site : 深度攝像頭數據的處理

@File : depth.py

@Software: PyCharm

author = “retacn”
copyright = “property of mankind.”
license = “CN”
version = “0.0.1”
maintainer = “retacn”
email = “[email protected]”
status = “Development”

import numpy as np

設備

CAP_OPENNI=900
CAP_OPENNI_ASUS=910

通道(基於浮點數的距離)

CAP_OPENNI_DEPTH_MAP=0

會得到bgr圖像

CAP_OPENNI_POINT_CLOUD_MAP=1

XYZ

CAP_OPENNI_DISPARITY_MAP=2

#
CAP_OPENNI_DISPARITY_MAP_32F=3
CAP_OPENNI_VALID_DEPTH_MASK=4

CAP_OPENNI_BGR_IMAGE=5
CAP_OPENNI_GRAY_IMAGE=6

生成掩模

def createMedianMask(disparityMap,#視差圖
validDepthMask,#有效深度掩模
rect=None):#矩形
if rect is not None:
x,y,w,h=rect
disparityMap=disparityMap[y:y+h,x:x+w]
validDepthMask=validDepthMask[y:y+h,x:x+w]
#得到中值
median=np.median(disparityMap)
#生成掩模,逐像素進行布爾操作
return np.where((validDepthMask==0) | (abs(disparityMap-median)<12),#值爲真假的數組,當有效視差與平均視差>=12,看作噪聲
1.0,#爲真時，數組相應元素爲該值
0.0)#爲假時，爲該值

3 對複製操作進行掩模

!/usr/bin/env python

-- coding: utf-8 --

@Time : 2016/12/1 10:22

@Author : Retacn

@Site : 矩形區域複製

@File : rects.py

@Software: PyCharm

author = “retacn”
copyright = “property of mankind.”
license = “CN”
version = “0.0.1”
maintainer = “retacn”
email = “[email protected]”
status = “Development”

import cv2
import numpy as np
from Three import utils

對複製操作執行掩模

def copyRect(src,
dst,
srcRect,
dstRect,
mask=None,#掩模參數,掩模要和圖像有相同的通道數
interpolation=cv2.INTER_LINEAR):#插值方法爲線性插值
x0,y0,w0,h0=srcRect
x1,y1,w1,h1=dstRect

#如果掩模爲空,則執行復制操作
if mask is None:
    dst[y1:y1+h1,x1:x1+w1]=cv2.resize(src[y0:y0+h0,y0:y0+h0],#源圖像
                                      (w1,h1),#目標圖像
                                      interpolation=interpolation)#插值方法
else:
    #如果掩模爲單通道,則複製通道
    if not utils.isGray(src):
        mask=mask.repeat(3).reshape(h0,w0,3)
    dst[y1:y1+h1,x1:x1+w1]=np.where(cv2.resize(mask,(w1,h1),interpolation=cv2.INTER_NEAREST),
                                    cv2.resize(src[y0:y0+h0,x0:x0+w0],(w1,h1),interpolation=interpolation),
                                    dst[y1:y1 + h1, x1:x1 + w1]
                                    )

一組矩形的循環交換

def swqpRects(src,dst,rects,masks=None,interpolation=cv2.INTER_LINEAR):
if dst is not src:
dst[:]=src

numRects=len(rects)
if numRects<2:
    return

if masks is None:
    masks=[None]*numRects

x,y,w,h=rects[numRects-1]
temp=src[y:y+h,x:x+w].copy()

i=numRects-2
while i>=0:
    copyRect(src,dst,rects[i],rects[i+1],masks[i],interpolation)
    i-=1
copyRect(temp,dst,(0,0,w,h),rects[0],masks[numRects-1],interpolation)

4 使用普通攝像頭進行深試評估

!/usr/bin/env python

-- coding: utf-8 --

@Time : 2016/12/1 11:40

@Author : Retacn

@Site : 使用普通攝像頭進行深試估計

@File : commonCamera2Depth.py

@Software: PyCharm

author = “retacn”
copyright = “property of mankind.”
license = “CN”
version = “0.0.1”
maintainer = “retacn”
email = “[email protected]”
status = “Development”

import cv2
import numpy as np

def update(val=0):
stereo.setBlockSize(cv2.getTrackbarPos(‘window_size’,’disparity’))
stereo.setUniquenessRatio(cv2.getTrackbarPos(‘uniquenessRatio’,’disparity’))
stereo.setSpeckleWindowSize(cv2.getTrackbarPos(‘speckleWindowSize’,’disparity’))
stereo.setSpeckleRange(cv2.getTrackbarPos(‘speckleRange’,’disparity’))
stereo.setDisp12MaxDiff(cv2.getTrackbarPos(‘disp12MaxDiff’,’disparity’))
print(‘computing disparity…’)
disp = stereo.compute(imgL, imgR).astype(np.float32) / 16.0
cv2.imshow(‘left’, imgL)
cv2.imshow(‘disparity’, (disp - min_disp) / num_disp)

if name==’main‘:
windows_size=5 #一個匹配塊的大小,大於1的奇數
min_disp=16 #最小視差值
num_disp=192-min_disp #最大視差值和最小視差值的差
blockSize=windows_size
uniquenessRatio=1
speckleRange=3 #每個已連接部分的最大視差變化
speckleWindowSize=3 #平滑視差區域的最大窗口尺寸
disp12MaxDiff=200
P1=600 #控制視差圖平滑度有第一個參數
P2=2400#第二個參數,值越大視差圖越平滑

#讀入圖像
imgL=cv2.imread('../imgl.jpg')
imgR=cv2.imread('../imgr.jpg')

cv2.namedWindow('disparity')
cv2.createTrackbar('speckleRange','disparity',speckleRange,50,update)
cv2.createTrackbar('window_size','disparity',windows_size,21,update)
cv2.createTrackbar('speckleWindowSize','disparity',speckleWindowSize,200,update)
cv2.createTrackbar('uniquenessRatio','disparity',uniquenessRatio,50,update)
cv2.createTrackbar('disp12MaxDiff','disparity',disp12MaxDiff,250,update)

stereo=cv2.StereoSGBM_create(minDisparity=min_disp,
                             numDisparities=num_disp,
                             blockSize=blockSize,
                             uniquenessRatio=uniquenessRatio,
                             speckleRange=speckleRange,
                             speckleWindowSize=speckleWindowSize,
                             disp12MaxDiff=disp12MaxDiff,
                             P1=P1,
                             P2=P2)

update()
cv2.waitKey()

5 使用分水嶺和grabcut算法進行物體分割

A 使用brabcut進行前景檢測

!/usr/bin/env python

-- coding: utf-8 --

@Time : 2016/12/2 11:17

@Author : Retacn

@Site : 使用grubCut進行前景檢測

@File : grabcutNew.py

@Software: PyCharm

author = “retacn”
copyright = “property of mankind.”
license = “CN”
version = “0.0.1”
maintainer = “retacn”
email = “[email protected]”
status = “Development”

import cv2
import numpy as np
import sys

定義顏色

BLUE = [255, 0, 0] # rectangle color
RED = [0, 0, 255] # PR BG
GREEN = [0, 255, 0] # PR FG
BLACK = [0, 0, 0] # sure BG
WHITE = [255, 255, 255] # sure FG

DRAW_BG = {‘color’: BLACK, ‘val’: 0}
DRAW_FG = {‘color’: WHITE, ‘val’: 1}
DRAW_PR_FG = {‘color’: GREEN, ‘val’: 3}
DRAW_PR_BG = {‘color’: RED, ‘val’: 2}

設置標誌位

rect = (0, 0, 1, 1) # 隔離對像矩形
drawing = False # 繪製標誌位
rectangle = False # 矩形繪製標誌
rect_over = False # 檢查隔離矩形是否繪製
rect_or_mask = 100 # 掩模模式
value = DRAW_FG #
thickness = 3 # 畫筆寬度

自定義鼠標回調函數

def onmouse(event, x, y, flags, param):
# 定義全局變量
global img, img2, drawing, value, mask, rectangle, rect, rect_or_mask, ix, iy, rect_over

# 繪製隔離矩形
if event == cv2.EVENT_RBUTTONDOWN:  # 鼠標右健按下,開始繪製矩形
    rectangle = True
    ix, iy = x, y

elif event == cv2.EVENT_MOUSEMOVE:  # 鼠標移動事件
    if rectangle == True:  # 繪製矩形
        img = img2.copy()
        cv2.rectangle(img,  # 源圖像
                      (ix, iy),  # 開始點
                      (x, y),  # 結束點
                      BLUE,  # 畫筆顏色
                      2)  # 畫筆寬度
        rect = (min(ix, x), min(iy, y), abs(ix - x), abs(iy - y))
        rect_or_mask = 0

elif event == cv2.EVENT_RBUTTONUP:  # 右鍵擡起
    rectangle = False  # 設置標誌位,矩形繪製完成
    rect_over = True
    cv2.rectangle(img, (ix, iy), (x, y), BLUE, 2)
    rect = (min(ix, x), min(iy, y), abs(ix - x), abs(iy - y))
    rect_or_mask=0
    print('按下 n , 開始繪製')

#繪製隔離圓形
if event==cv2.EVENT_LBUTTONDOWN: #左鍵按下
    if rect_over==False:
        print("請先繪製圓形")
    else:
        drawing=True
        cv2.circle(img,(x,y),thickness,value['color'],-1)
        cv2.circle(mask,(x,y),thickness,value['color'],-1)

elif event==cv2.EVENT_MOUSEMOVE:
    if drawing==True:
        cv2.circle(img, (x, y), thickness, value['color'], -1)
        cv2.circle(mask, (x, y), thickness, value['color'], -1)

elif event==cv2.EVENT_LBUTTONUP:
    if drawing==True:
        drawing=False
        cv2.circle(img, (x, y), thickness, value['color'], -1)
        cv2.circle(mask, (x, y), thickness, value['color'], -1)

if name == ‘main‘:
fileName = ‘../test1.jpg’

img = cv2.imread(fileName)
img2 = img.copy()
mask = np.zeros(img.shape[:2], dtype=np.uint8)
output = np.zeros(img.shape, np.uint8)

# 定義輸入輸出窗口
cv2.namedWindow("output")
cv2.namedWindow('input')
#輸出窗口註冊鼠標事件
cv2.setMouseCallback('input', onmouse)
cv2.moveWindow('input',img.shape[1]+10,90)

print("操作指南:\n")
print('使用鼠標右健在源圖像中繪製矩形\n')

while(1):
    #顯示圖像
    cv2.imshow('output',output)
    cv2.imshow('input',img)
    k=0xFF&cv2.waitKey(1)

    if k==27: #ESC鍵退出
        break
    elif k==ord('0'):#繪製背景
        print("左鍵繪製background標識區域 \n")
        value=DRAW_BG
    elif k==ord('1'):#繪製前景
        print('左鍵繪製foreground標識區域 \n')
        value=DRAW_FG
    elif k==ord('2'):
        value=DRAW_PR_BG
    elif k == ord('3'):
        value = DRAW_PR_FG
    elif k==ord('s'):# 保存圖像
        bar=np.zeros((img.shape[0],5,3),np.uint8)
        res=np.hstack((img2.bar,img,bar,output))
        cv2.imwrite('grabcut_output.png',res)
        print('保存圖像')
    elif k==ord('r'):#重置
        print('開始重置 \n')
        rect = (0, 0, 1, 1)
        drawing = False
        rectangle = False
        rect_or_mask = 100
        rect_over = False
        value = DRAW_FG
        img = img2.copy()
        mask = np.zeros(img.shape[:2], dtype=np.uint8)
        output = np.zeros(img.shape, np.uint8)
    elif k == ord('n'):  # 圖像截取
        print(""" For finer touchups, mark foreground and background after pressing keys 0-3
               and again press 'n' \n""")
        if (rect_or_mask == 0):  # 設置掩模
            bgdmodel = np.zeros((1, 65), np.float64)
            fgdmodel = np.zeros((1, 65), np.float64)
            cv2.grabCut(img2, mask, rect, bgdmodel, fgdmodel, 1, cv2.GC_INIT_WITH_RECT)
            rect_or_mask = 1
        elif rect_or_mask == 1:  # 設置掩模
            bgdmodel = np.zeros((1, 65), np.float64)
            fgdmodel = np.zeros((1, 65), np.float64)
            cv2.grabCut(img2, mask, rect, bgdmodel, fgdmodel, 1, cv2.GC_INIT_WITH_MASK)

    mask2 = np.where((mask == 1) + (mask == 3), 255, 0).astype('uint8')
    output = cv2.bitwise_and(img2, img2, mask=mask2)

cv2.destroyAllWindows()

B 使用分水嶺算法進行圖像分割

示例代碼如下:

!/usr/bin/env python

-- coding: utf-8 --

@Time : 2016/12/2 12:21

@Author : Retacn

@Site : 使用分水嶺算法進行圖像分割

@File : watershed.py

@Software: PyCharm

author = “retacn”
copyright = “property of mankind.”
license = “CN”
version = “0.0.1”
maintainer = “retacn”
email = “[email protected]”
status = “Development”

import cv2
import numpy as np
from matplotlib import pyplot as plt

img =cv2.imread(‘../test1.jpg’)
gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

轉換爲灰度圖,設置閾值

ret,thresh=cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)

kernel=np.ones((3,3),np.uint8)
opening=cv2.morphologyEx(thresh,#源圖像
cv2.MORPH_OPEN, #開運算
kernel, #核
iterations=2)#迭代次數

取得確定的前景區域

sure_bg=cv2.dilate(opening,kernel,iterations=3)
dist_transform=cv2.distanceTransform(opening,cv2.DIST_L2,5)
ret,sure_fg=cv2.threshold(dist_transform,0.7*dist_transform.max(),255,0)

sure_fg=np.uint8(sure_fg)
unknown=cv2.subtract(sure_bg,sure_fg)

ret,markers=cv2.connectedComponents(sure_fg)
markers=markers+1
markers[unknown==255]=0

markers=cv2.watershed(img,markers)
img[markers==-1]=[255,0,0]
plt.imshow(img)
plt.show()