#-*- coding: utf-8 -*-#feimengjuan#利用python实现多种方法来实现图像识别
importcv2importnumpy as npfrom matplotlib importpyplot as plt, image#最简单的以灰度直方图作为相似比较的实现
def classify_gray_hist(image1, image2, size=(256, 256)):#先计算直方图
#几个参数必须用方括号括起来
#这里直接用灰度图计算直方图,所以是使用第一个通道,
#也可以进行通道分离后,得到多个通道的直方图
#bins 取为16
image1 =cv2.resize(image1, size)
image2=cv2.resize(image2, size)
hist1= cv2.calcHist([image1], [0], None, [256], [0.0, 255.0])
hist2= cv2.calcHist([image2], [0], None, [256], [0.0, 255.0])#可以比较下直方图
plt.plot(range(256), hist1, 'r')
plt.plot(range(256), hist2, 'b')
plt.show()#计算直方图的重合度
degree =0for i inrange(len(hist1)):if hist1[i] !=hist2[i]:
degree= degree + (1 - abs(hist1[i] - hist2[i]) /max(hist1[i], hist2[i]))else:
degree= degree + 1degree= degree /len(hist1)returndegree#计算单通道的直方图的相似值
defcalculate(image1, image2):
hist1= cv2.calcHist([image1], [0], None, [256], [0.0, 255.0])
hist2= cv2.calcHist([image2], [0], None, [256], [0.0, 255.0])#计算直方图的重合度
degree =0for i inrange(len(hist1)):if hist1[i] !=hist2[i]:
degree= degree + (1 - abs(hist1[i] - hist2[i]) /max(hist1[i], hist2[i]))else:
degree= degree + 1degree= degree /len(hist1)returndegree#通过得到每个通道的直方图来计算相似度
def classify_hist_with_split(image1, image2, size=(256, 256)):#将图像resize后,分离为三个通道,再计算每个通道的相似值
image1 =cv2.resize(image1, size)
image2=cv2.resize(image2, size)
sub_image1=cv2.split(image1)
sub_image2=cv2.split(image2)
sub_data=0for im1, im2 inzip(sub_image1, sub_image2):
sub_data+=calculate(im1, im2)
sub_data= sub_data / 3
returnsub_data#平均哈希算法计算
defclassify_aHash(image1, image2):
image1= cv2.resize(image1, (8, 8))
image2= cv2.resize(image2, (8, 8))
gray1=cv2.cvtColor(image1, cv2.COLOR_BGR2GRAY)
gray2=cv2.cvtColor(image2, cv2.COLOR_BGR2GRAY)
hash1=getHash(gray1)
hash2=getHash(gray2)returnHamming_distance(hash1, hash2)defclassify_pHash(image1, image2):
image1= cv2.resize(image1, (32, 32))
image2= cv2.resize(image2, (32, 32))
gray1=cv2.cvtColor(image1, cv2.COLOR_BGR2GRAY)
gray2=cv2.cvtColor(image2, cv2.COLOR_BGR2GRAY)#将灰度图转为浮点型,再进行dct变换
dct1 =cv2.dct(np.float32(gray1))
dct2=cv2.dct(np.float32(gray2))#取左上角的8*8,这些代表图片的最低频率
#这个操作等价于c++中利用opencv实现的掩码操作
#在python中进行掩码操作,可以直接这样取出图像矩阵的某一部分
dct1_roi = dct1[0:8, 0:8]
dct2_roi= dct2[0:8, 0:8]
hash1=getHash(dct1_roi)
hash2=getHash(dct2_roi)returnHamming_distance(hash1, hash2)#输入灰度图,返回hash
defgetHash(image):
avreage=np.mean(image)
hash=[]for i inrange(image.shape[0]):for j in range(image.shape[1]):if image[i, j] >avreage:
hash.append(1)else:
hash.append(0)returnhash#计算汉明距离
defHamming_distance(hash1, hash2):
num=0for index inrange(len(hash1)):if hash1[index] !=hash2[index]:
num+= 1
returnnumif __name__ == '__main__':
face_cascade= cv2.CascadeClassifier('haarcascade_frontalface_alt2.xml')
face_cascade.load('F:\pycharm\py2_7\Lib\site-packages\cv2\data\haarcascade_frontalface_default.xml')
scaling_factor= 0.5img1= cv2.imread('d://68.jpg')#img1 = cv2.resize(img1,(300,300))
img1 = cv2.resize(img1, None, fx=scaling_factor * 3, fy=scaling_factor * 3, interpolation=cv2.INTER_AREA)
gray=cv2.cvtColor(img1, cv2.COLOR_BGR2GRAY)
face_rects= face_cascade.detectMultiScale(gray, 1.1, 5)for (x, y, w, h) inface_rects:
img1= img1[y:y + h,x:x +w]#cv2.rectangle(img1, (x, y), (x + w, y + h), (0, 255, 0), 3)
img2= cv2.imread('d://69.jpg')#img2 = cv2.resize(img2,(300,300))
img2 = cv2.resize(img2, None, fx=scaling_factor * 3, fy=scaling_factor * 3, interpolation=cv2.INTER_AREA)
gray2=cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY)
face_rects2= face_cascade.detectMultiScale(gray2, 1.1, 5)for (x, y, w, h) inface_rects2:
img2= img2[y:y + h,x:x +w]#cv2.rectangle(img2, (x, y), (x + w, y + h), (0, 255, 0), 3)
#degree = classify_gray_hist(img1, img2)
degree =classify_hist_with_split(img1,img2)#degree = classify_aHash(img1,img2)
#degree = classify_pHash(img1,img2)
out = format('相似度:%f' % (float)(degree*100))printout
img1= cv2.resize(img1, (300, 300))
cv2.imshow('img1', img1)
img2= cv2.resize(img2, (300, 300))
cv2.imshow('img2',img2)#degree = classify_hist_with_split(img1,img2)
#degree = classify_aHash(img1,img2)
#degree = classify_pHash(img1,img2)
#print degree
cv2.waitKey(0)