????????兩個(gè)關(guān)鍵子系統(tǒng)： 車牌定位 和?字符識(shí)別
一，主要步驟
- 原始車輛圖像采集
- 汽車牌照區(qū)域定位
- 汽車牌照內(nèi)字符的分割
- 汽車牌照內(nèi)字符的識(shí)別
車牌定位方法：
- 基于邊緣檢測(cè)的車牌定位方法
- 基于遺傳算法的車牌定位方法
- 基于紋理特征的車牌定位方法
- 基于數(shù)學(xué)形態(tài)學(xué)的車牌定位方法
- 基于小波分析和變換的車牌定位方法
- 基于神經(jīng)網(wǎng)絡(luò)的車牌定位方法
車牌定位技術(shù)：
- 車牌圖像的濾波
- 車牌圖像的二值化
- 車牌圖像的邊緣檢測(cè)
- 車牌圖像的灰度映射
- 車牌圖像的改進(jìn)型投影法定位
- 水平投影?：一階差分運(yùn)算于汽車牌照?qǐng)D像，在水平方向上進(jìn)行。 累加位于水平差分圖像中的像素，累加沿水平方向進(jìn)行。 水平投影表得以產(chǎn)生，利用該表并集合如前所述的汽車牌照在水平投影后在投影值上表現(xiàn)出來的特征確定汽車牌照的大概位置。??????
- 垂直投影：方法與水平投影類似，做垂直方向的差分運(yùn)算，進(jìn)行平滑，得到左右邊界。
- 傳統(tǒng)車牌投影順序 ：水平投影、水平搜索、水平提取、垂直投影、垂直搜索、垂直提取。
車牌圖像預(yù)處理：
- 車牌圖像的灰度化
- 車牌圖像的直方圖均衡化
車牌字符分割技術(shù)：
- 車牌傾斜度檢測(cè)方法
- 車牌傾斜的矯正方法
- 車牌邊框和鉚釘?shù)娜コ?/strong>
- 基于垂直投影和先驗(yàn)知識(shí)的車牌字符分割
- 計(jì)算垂直投影
- 初步垂直切分
- 粘連車牌字符的分???????
- 割斷裂車牌字符的合并
- 對(duì)車牌字符的切分結(jié)果進(jìn)行確認(rèn)
車牌字符識(shí)別技術(shù)：
- 模式識(shí)別
? ? ? ? ? ?模式就是一種對(duì)某種對(duì)象（一些敏感的客體）結(jié)構(gòu)或者定量的描述，是一種集合（由具有某些共同特定性質(zhì)的模式構(gòu)成）。目前模式識(shí)別主要有4種方法：基于神經(jīng)網(wǎng)絡(luò)的識(shí)別方法、基于句法模式的識(shí)別方法、基于統(tǒng)計(jì)模式的識(shí)別方法和基于模糊模式的識(shí)別方法。
- 字符識(shí)別
- 基于神經(jīng)網(wǎng)絡(luò)的識(shí)別方法
- 基于特征分析的匹配方法
- 基于模版的匹配方法
- 英文、數(shù)字識(shí)別
????????目前，小波識(shí)別法、模板匹配法與神經(jīng)網(wǎng)絡(luò)法等常被作為汽車牌照字符識(shí)別的主要方法
- 漢字識(shí)別
二，車牌定位
# -*- coding: utf-8 -*-
import cv2
import numpy as np
def stretch(img):
'''
圖像拉伸函數(shù)
'''
maxi=float(img.max())
mini=float(img.min())
for i in range(img.shape[0]):
for j in range(img.shape[1]):
img[i,j]=(255/(maxi-mini)*img[i,j]-(255*mini)/(maxi-mini))
return img
def dobinaryzation(img):
'''
二值化處理函數(shù)
'''
maxi=float(img.max())
mini=float(img.min())
x=maxi-((maxi-mini)/2)
#二值化,返回閾值ret 和 二值化操作后的圖像thresh
ret,thresh=cv2.threshold(img,x,255,cv2.THRESH_BINARY)
#返回二值化后的黑白圖像
return thresh
def find_rectangle(contour):
'''
尋找矩形輪廓
'''
y,x=[],[]
for p in contour:
y.append(p[0][0])
x.append(p[0][1])
return [min(y),min(x),max(y),max(x)]
def locate_license(img,afterimg):
'''
定位車牌號(hào)
'''
contours,hierarchy=cv2.findContours(img,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
#找出最大的三個(gè)區(qū)域
block=[]
for c in contours:
#找出輪廓的左上點(diǎn)和右下點(diǎn)，由此計(jì)算它的面積和長(zhǎng)度比
r=find_rectangle(c)
a=(r[2]-r[0])*(r[3]-r[1]) #面積
s=(r[2]-r[0])*(r[3]-r[1]) #長(zhǎng)度比
block.append([r,a,s])
#選出面積最大的3個(gè)區(qū)域
block=sorted(block,key=lambda b: b[1])[-3:]
#使用顏色識(shí)別判斷找出最像車牌的區(qū)域
maxweight,maxindex=0,-1
for i in range(len(block)):
b=afterimg[block[i][0][1]:block[i][0][3],block[i][0][0]:block[i][0][2]]
#BGR轉(zhuǎn)HSV
hsv=cv2.cvtColor(b,cv2.COLOR_BGR2HSV)
#藍(lán)色車牌的范圍
lower=np.array([100,50,50])
upper=np.array([140,255,255])
#根據(jù)閾值構(gòu)建掩膜
mask=cv2.inRange(hsv,lower,upper)
#統(tǒng)計(jì)權(quán)值
w1=0
for m in mask:
w1+=m/255
w2=0
for n in w1:
w2+=n
#選出最大權(quán)值的區(qū)域
if w2>maxweight:
maxindex=i
maxweight=w2
return block[maxindex][0]
def find_license(img):
'''
預(yù)處理函數(shù)
'''
m=400*img.shape[0]/img.shape[1]
#壓縮圖像
img=cv2.resize(img,(400,int(m)),interpolation=cv2.INTER_CUBIC)
#BGR轉(zhuǎn)換為灰度圖像
gray_img=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
#灰度拉伸
stretchedimg=stretch(gray_img)
'''進(jìn)行開運(yùn)算，用來去除噪聲'''
r=16
h=w=r*2+1
kernel=np.zeros((h,w),np.uint8)
cv2.circle(kernel,(r,r),r,1,-1)
#開運(yùn)算
openingimg=cv2.morphologyEx(stretchedimg,cv2.MORPH_OPEN,kernel)
#獲取差分圖，兩幅圖像做差 cv2.absdiff('圖像1','圖像2')
strtimg=cv2.absdiff(stretchedimg,openingimg)
#圖像二值化
binaryimg=dobinaryzation(strtimg)
#canny邊緣檢測(cè)
canny=cv2.Canny(binaryimg,binaryimg.shape[0],binaryimg.shape[1])
'''消除小的區(qū)域，保留大塊的區(qū)域，從而定位車牌'''
#進(jìn)行閉運(yùn)算
kernel=np.ones((5,19),np.uint8)
closingimg=cv2.morphologyEx(canny,cv2.MORPH_CLOSE,kernel)
#進(jìn)行開運(yùn)算
openingimg=cv2.morphologyEx(closingimg,cv2.MORPH_OPEN,kernel)
#再次進(jìn)行開運(yùn)算
kernel=np.ones((11,5),np.uint8)
openingimg=cv2.morphologyEx(openingimg,cv2.MORPH_OPEN,kernel)
#消除小區(qū)域，定位車牌位置
rect=locate_license(openingimg,img)
return rect,img
def cut_license(afterimg,rect):
'''
圖像分割函數(shù)
'''
#轉(zhuǎn)換為寬度和高度
rect[2]=rect[2]-rect[0]
rect[3]=rect[3]-rect[1]
rect_copy=tuple(rect.copy())
rect=[0,0,0,0]
#創(chuàng)建掩膜
mask=np.zeros(afterimg.shape[:2],np.uint8)
#創(chuàng)建背景模型 大小只能為13*5，行數(shù)只能為1，單通道浮點(diǎn)型
bgdModel=np.zeros((1,65),np.float64)
#創(chuàng)建前景模型
fgdModel=np.zeros((1,65),np.float64)
#分割圖像
cv2.grabCut(afterimg,mask,rect_copy,bgdModel,fgdModel,5,cv2.GC_INIT_WITH_RECT)
mask2=np.where((mask==2)|(mask==0),0,1).astype('uint8')
img_show=afterimg*mask2[:,:,np.newaxis]
return img_show
def deal_license(licenseimg):
'''
車牌圖片二值化
'''
#車牌變?yōu)榛叶葓D像
gray_img=cv2.cvtColor(licenseimg,cv2.COLOR_BGR2GRAY)
#均值濾波 去除噪聲
kernel=np.ones((3,3),np.float32)/9
gray_img=cv2.filter2D(gray_img,-1,kernel)
#二值化處理
ret,thresh=cv2.threshold(gray_img,120,255,cv2.THRESH_BINARY)
return thresh
def find_end(start,arg,black,white,width,black_max,white_max):
end=start+1
for m in range(start+1,width-1):
if (black[m] if arg else white[m])>(0.98*black_max if arg else 0.98*white_max):
end=m
break
return end
if __name__=='__main__':
img=cv2.imread('car.jpg',cv2.IMREAD_COLOR)
#預(yù)處理圖像
rect,afterimg=find_license(img)
#框出車牌號(hào)
cv2.rectangle(afterimg,(rect[0],rect[1]),(rect[2],rect[3]),(0,255,0),2)
cv2.imshow('afterimg',afterimg)
#分割車牌與背景
cutimg=cut_license(afterimg,rect)
cv2.imshow('cutimg',cutimg)
#二值化生成黑白圖
thresh=deal_license(cutimg)
cv2.imshow('thresh',thresh)
cv2.imwrite("cp.jpg",thresh)
cv2.waitKey(0)
#分割字符
'''
判斷底色和字色
'''
#記錄黑白像素總和
white=[]
black=[]
height=thresh.shape[0] #263
width=thresh.shape[1] #400
#print('height',height)
#print('width',width)
white_max=0
black_max=0
#計(jì)算每一列的黑白像素總和
for i in range(width):
line_white=0
line_black=0
for j in range(height):
if thresh[j][i]==255:
line_white+=1
if thresh[j][i]==0:
line_black+=1
white_max=max(white_max,line_white)
black_max=max(black_max,line_black)
white.append(line_white)
black.append(line_black)
print('white',white)
print('black',black)
#arg為true表示黑底白字，F(xiàn)alse為白底黑字
arg=True
if black_max<white_max:
arg=False
n=1
start=1
end=2
while n<width-2:
n+=1
#判斷是白底黑字還是黑底白字 0.05參數(shù)對(duì)應(yīng)上面的0.95 可作調(diào)整
if(white[n] if arg else black[n])>(0.02*white_max if arg else 0.02*black_max):
start=n
end=find_end(start,arg,black,white,width,black_max,white_max)
n=end
if end-start>5:
cj=thresh[1:height,start:end]
cv2.imshow('cutlicense',cj)
cv2.waitKey(0)
cv2.waitKey(0)
cv2.destroyAllWindows()三，字符識(shí)別
? ? ? ? 1, 安裝包?pytesseract?pillow
????????pytesseract 可用于驗(yàn)證碼識(shí)別?【精選】Python OCR工具pytesseract詳解_測(cè)試開發(fā)小記的博客-CSDN博客
? ? ? ? pillow?百度安全驗(yàn)證
'''
是基于Python的OCR工具， 底層使用的是Google的Tesseract-OCR 引擎，支持識(shí)別圖片中的文字，
支持jpeg, png, gif, bmp, tiff等圖片格式。本文介紹如何使用pytesseract 實(shí)現(xiàn)圖片文字識(shí)別。
'''
pip install pytesseract?
'''?PIL?軟件包提供了基本的圖像處理功能，如：改變圖像大小，旋轉(zhuǎn)圖像，圖像格式轉(zhuǎn)換，
場(chǎng)空間轉(zhuǎn)換，圖像增強(qiáng)，直方圖處理，插值和濾波等等。
'''
pip install pillow?? ? ? ?2，安裝pytesseract?
????????選擇合適的版本，安裝包地址：
????????Home · UB-Mannheim/tesseract Wiki · GitHub
?????????3， 修改 pytesseract?包源文件?
? ? ? ? 修改為指向剛才的安裝地址
四，車牌識(shí)別系統(tǒng)
? ? ? 主界面文章來源：http://www.zghlxwxcb.cn/news/detail-831743.html
import tkinter as tk
from tkinter.filedialog import *
from tkinter import ttk
import tkinter.messagebox as mBox
import predict
import cv2
from PIL import Image, ImageTk
import threading
import time
class Surface(ttk.Frame):
pic_path = ""
viewhigh = 600
viewwide = 600
update_time = 0
thread = None
thread_run = False
camera = None
color_transform = {"green":("綠牌","#55FF55"), "yello":("黃牌","#FFFF00"), "blue":("藍(lán)牌","#6666FF")}
def __init__(self, win):
ttk.Frame.__init__(self, win)
frame_left = ttk.Frame(self)
frame_right1 = ttk.Frame(self)
frame_right2 = ttk.Frame(self)
win.title("車牌識(shí)別")
win.state("zoomed")
self.pack(fill=tk.BOTH, expand=tk.YES, padx="5", pady="5")
frame_left.pack(side=tk.LEFT,expand=1,fill=tk.BOTH)
frame_right1.pack(side=tk.TOP,expand=1,fill=tk.Y)
frame_right2.pack(side=tk.RIGHT,expand=0)
ttk.Label(frame_left, text='原圖：').pack(anchor="nw")
ttk.Label(frame_right1, text='車牌位置：').grid(column=0, row=0, sticky=tk.W)
from_pic_ctl = ttk.Button(frame_right2, text="來自圖片", width=20, command=self.from_pic)
from_vedio_ctl = ttk.Button(frame_right2, text="來自攝像頭", width=20, command=self.from_vedio)
self.image_ctl = ttk.Label(frame_left)
self.image_ctl.pack(anchor="nw")
self.roi_ctl = ttk.Label(frame_right1)
self.roi_ctl.grid(column=0, row=1, sticky=tk.W)
ttk.Label(frame_right1, text='識(shí)別結(jié)果：').grid(column=0, row=2, sticky=tk.W)
self.r_ctl = ttk.Label(frame_right1, text="")
self.r_ctl.grid(column=0, row=3, sticky=tk.W)
self.color_ctl = ttk.Label(frame_right1, text="", width="20")
self.color_ctl.grid(column=0, row=4, sticky=tk.W)
from_vedio_ctl.pack(anchor="se", pady="5")
from_pic_ctl.pack(anchor="se", pady="5")
self.predictor = predict.CardPredictor()
self.predictor.train_svm()
def get_imgtk(self, img_bgr):
img = cv2.cvtColor(img_bgr, cv2.COLOR_BGR2RGB)
im = Image.fromarray(img)
imgtk = ImageTk.PhotoImage(image=im)
wide = imgtk.width()
high = imgtk.height()
if wide > self.viewwide or high > self.viewhigh:
wide_factor = self.viewwide / wide
high_factor = self.viewhigh / high
factor = min(wide_factor, high_factor)
wide = int(wide * factor)
if wide <= 0 : wide = 1
high = int(high * factor)
if high <= 0 : high = 1
im=im.resize((wide, high), Image.LANCZOS) #在pillow的10.0.0版本中，ANTIALIAS方法被刪除了，使用新的方法即可：
imgtk = ImageTk.PhotoImage(image=im)
return imgtk
def show_roi(self, r, roi, color):
if r :
roi = cv2.cvtColor(roi, cv2.COLOR_BGR2RGB)
roi = Image.fromarray(roi)
self.imgtk_roi = ImageTk.PhotoImage(image=roi)
self.roi_ctl.configure(image=self.imgtk_roi, state='enable')
self.r_ctl.configure(text=str(r))
self.update_time = time.time()
try:
c = self.color_transform[color]
self.color_ctl.configure(text=c[0], background=c[1], state='enable')
except:
self.color_ctl.configure(state='disabled')
elif self.update_time + 8 < time.time():
self.roi_ctl.configure(state='disabled')
self.r_ctl.configure(text="")
self.color_ctl.configure(state='disabled')
def from_vedio(self):
if self.thread_run:
return
if self.camera is None:
self.camera = cv2.VideoCapture(0)
if not self.camera.isOpened():
mBox.showwarning('警告', '攝像頭打開失??！')
self.camera = None
return
self.thread = threading.Thread(target=self.vedio_thread, args=(self,))
self.thread.setDaemon(True)
self.thread.start()
self.thread_run = True
def from_pic(self):
self.thread_run = False
self.pic_path = askopenfilename(title="選擇識(shí)別圖片", filetypes=[("jpg圖片", "*.jpg")])
if self.pic_path:
img_bgr = predict.imreadex(self.pic_path)
self.imgtk = self.get_imgtk(img_bgr)
self.image_ctl.configure(image=self.imgtk)
resize_rates = (1, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4)
for resize_rate in resize_rates:
print("resize_rate:", resize_rate)
try:
r, roi, color = self.predictor.predict(img_bgr, resize_rate)
except:
continue
if r:
break
#r, roi, color = self.predictor.predict(img_bgr, 1)
self.show_roi(r, roi, color)
@staticmethod
def vedio_thread(self):
self.thread_run = True
predict_time = time.time()
while self.thread_run:
_, img_bgr = self.camera.read()
self.imgtk = self.get_imgtk(img_bgr)
self.image_ctl.configure(image=self.imgtk)
if time.time() - predict_time > 2:
r, roi, color = self.predictor.predict(img_bgr)
self.show_roi(r, roi, color)
predict_time = time.time()
print("run end")
def close_window():
print("destroy")
if surface.thread_run :
surface.thread_run = False
surface.thread.join(2.0)
win.destroy()
if __name__ == '__main__':
win=tk.Tk()
surface = Surface(win)
win.protocol('WM_DELETE_WINDOW', close_window)
win.mainloop()
? ? ? ?預(yù)測(cè)文件?predict.py文章來源地址http://www.zghlxwxcb.cn/news/detail-831743.html
import cv2
import numpy as np
from numpy.linalg import norm
import sys
import os
import json
SZ = 20 #訓(xùn)練圖片長(zhǎng)寬
MAX_WIDTH = 1000 #原始圖片最大寬度
Min_Area = 2000 #車牌區(qū)域允許最大面積
PROVINCE_START = 1000
#讀取圖片文件
def imreadex(filename):
return cv2.imdecode(np.fromfile(filename, dtype=np.uint8), cv2.IMREAD_COLOR)
def point_limit(point):
if point[0] < 0:
point[0] = 0
if point[1] < 0:
point[1] = 0
#根據(jù)設(shè)定的閾值和圖片直方圖，找出波峰，用于分隔字符
def find_waves(threshold, histogram):
up_point = -1#上升點(diǎn)
is_peak = False
if histogram[0] > threshold:
up_point = 0
is_peak = True
wave_peaks = []
for i,x in enumerate(histogram):
if is_peak and x < threshold:
if i - up_point > 2:
is_peak = False
wave_peaks.append((up_point, i))
elif not is_peak and x >= threshold:
is_peak = True
up_point = i
if is_peak and up_point != -1 and i - up_point > 4:
wave_peaks.append((up_point, i))
return wave_peaks
#根據(jù)找出的波峰，分隔圖片，從而得到逐個(gè)字符圖片
def seperate_card(img, waves):
part_cards = []
for wave in waves:
part_cards.append(img[:, wave[0]:wave[1]])
return part_cards
#來自opencv的sample，用于svm訓(xùn)練
def deskew(img):
m = cv2.moments(img)
if abs(m['mu02']) < 1e-2:
return img.copy()
skew = m['mu11']/m['mu02']
M = np.float32([[1, skew, -0.5*SZ*skew], [0, 1, 0]])
img = cv2.warpAffine(img, M, (SZ, SZ), flags=cv2.WARP_INVERSE_MAP | cv2.INTER_LINEAR)
return img
#來自opencv的sample，用于svm訓(xùn)練
def preprocess_hog(digits):
samples = []
for img in digits:
gx = cv2.Sobel(img, cv2.CV_32F, 1, 0)
gy = cv2.Sobel(img, cv2.CV_32F, 0, 1)
mag, ang = cv2.cartToPolar(gx, gy)
bin_n = 16
bin = np.int32(bin_n*ang/(2*np.pi))
bin_cells = bin[:10,:10], bin[10:,:10], bin[:10,10:], bin[10:,10:]
mag_cells = mag[:10,:10], mag[10:,:10], mag[:10,10:], mag[10:,10:]
hists = [np.bincount(b.ravel(), m.ravel(), bin_n) for b, m in zip(bin_cells, mag_cells)]
hist = np.hstack(hists)
# transform to Hellinger kernel
eps = 1e-7
hist /= hist.sum() + eps
hist = np.sqrt(hist)
hist /= norm(hist) + eps
samples.append(hist)
return np.float32(samples)
#不能保證包括所有省份
provinces = [
"zh_cuan", "川",
"zh_e", "鄂",
"zh_gan", "贛",
"zh_gan1", "甘",
"zh_gui", "貴",
"zh_gui1", "桂",
"zh_hei", "黑",
"zh_hu", "滬",
"zh_ji", "冀",
"zh_jin", "津",
"zh_jing", "京",
"zh_jl", "吉",
"zh_liao", "遼",
"zh_lu", "魯",
"zh_meng", "蒙",
"zh_min", "閩",
"zh_ning", "寧",
"zh_qing", "靑",
"zh_qiong", "瓊",
"zh_shan", "陜",
"zh_su", "蘇",
"zh_sx", "晉",
"zh_wan", "皖",
"zh_xiang", "湘",
"zh_xin", "新",
"zh_yu", "豫",
"zh_yu1", "渝",
"zh_yue", "粵",
"zh_yun", "云",
"zh_zang", "藏",
"zh_zhe", "浙"
]
class StatModel(object):
def load(self, fn):
self.model = self.model.load(fn)
def save(self, fn):
self.model.save(fn)
class SVM(StatModel):
def __init__(self, C = 1, gamma = 0.5):
self.model = cv2.ml.SVM_create()
self.model.setGamma(gamma)
self.model.setC(C)
self.model.setKernel(cv2.ml.SVM_RBF)
self.model.setType(cv2.ml.SVM_C_SVC)
#訓(xùn)練svm
def train(self, samples, responses):
self.model.train(samples, cv2.ml.ROW_SAMPLE, responses)
#字符識(shí)別
def predict(self, samples):
r = self.model.predict(samples)
return r[1].ravel()
class CardPredictor:
def __init__(self):
#車牌識(shí)別的部分參數(shù)保存在js中，便于根據(jù)圖片分辨率做調(diào)整
f = open('config.js')
j = json.load(f)
for c in j["config"]:
if c["open"]:
self.cfg = c.copy()
break
else:
raise RuntimeError('沒有設(shè)置有效配置參數(shù)')
def __del__(self):
self.save_traindata()
def train_svm(self):
#識(shí)別英文字母和數(shù)字
self.model = SVM(C=1, gamma=0.5)
#識(shí)別中文
self.modelchinese = SVM(C=1, gamma=0.5)
if os.path.exists("svm.dat"):
self.model.load("svm.dat")
else:
chars_train = []
chars_label = []
for root, dirs, files in os.walk("train\\chars2"):
if len(os.path.basename(root)) > 1:
continue
root_int = ord(os.path.basename(root))
for filename in files:
filepath = os.path.join(root,filename)
digit_img = cv2.imread(filepath)
digit_img = cv2.cvtColor(digit_img, cv2.COLOR_BGR2GRAY)
chars_train.append(digit_img)
#chars_label.append(1)
chars_label.append(root_int)
chars_train = list(map(deskew, chars_train))
chars_train = preprocess_hog(chars_train)
#chars_train = chars_train.reshape(-1, 20, 20).astype(np.float32)
chars_label = np.array(chars_label)
self.model.train(chars_train, chars_label)
if os.path.exists("svmchinese.dat"):
self.modelchinese.load("svmchinese.dat")
else:
chars_train = []
chars_label = []
for root, dirs, files in os.walk("train\\charsChinese"):
if not os.path.basename(root).startswith("zh_"):
continue
pinyin = os.path.basename(root)
index = provinces.index(pinyin) + PROVINCE_START + 1 #1是拼音對(duì)應(yīng)的漢字
for filename in files:
filepath = os.path.join(root,filename)
digit_img = cv2.imread(filepath)
digit_img = cv2.cvtColor(digit_img, cv2.COLOR_BGR2GRAY)
chars_train.append(digit_img)
#chars_label.append(1)
chars_label.append(index)
chars_train = list(map(deskew, chars_train))
chars_train = preprocess_hog(chars_train)
#chars_train = chars_train.reshape(-1, 20, 20).astype(np.float32)
chars_label = np.array(chars_label)
print(chars_train.shape)
self.modelchinese.train(chars_train, chars_label)
def save_traindata(self):
if not os.path.exists("svm.dat"):
self.model.save("svm.dat")
if not os.path.exists("svmchinese.dat"):
self.modelchinese.save("svmchinese.dat")
def accurate_place(self, card_img_hsv, limit1, limit2, color):
row_num, col_num = card_img_hsv.shape[:2]
xl = col_num
xr = 0
yh = 0
yl = row_num
#col_num_limit = self.cfg["col_num_limit"]
row_num_limit = self.cfg["row_num_limit"]
col_num_limit = col_num * 0.8 if color != "green" else col_num * 0.5#綠色有漸變
for i in range(row_num):
count = 0
for j in range(col_num):
H = card_img_hsv.item(i, j, 0)
S = card_img_hsv.item(i, j, 1)
V = card_img_hsv.item(i, j, 2)
if limit1 < H <= limit2 and 34 < S and 46 < V:
count += 1
if count > col_num_limit:
if yl > i:
yl = i
if yh < i:
yh = i
for j in range(col_num):
count = 0
for i in range(row_num):
H = card_img_hsv.item(i, j, 0)
S = card_img_hsv.item(i, j, 1)
V = card_img_hsv.item(i, j, 2)
if limit1 < H <= limit2 and 34 < S and 46 < V:
count += 1
if count > row_num - row_num_limit:
if xl > j:
xl = j
if xr < j:
xr = j
return xl, xr, yh, yl
def predict(self, car_pic, resize_rate=1):
if type(car_pic) == type(""):
img = imreadex(car_pic)
else:
img = car_pic
pic_hight, pic_width = img.shape[:2]
if pic_width > MAX_WIDTH:
pic_rate = MAX_WIDTH / pic_width
img = cv2.resize(img, (MAX_WIDTH, int(pic_hight*pic_rate)), interpolation=cv2.INTER_LANCZOS4)
pic_hight, pic_width = img.shape[:2]
if resize_rate != 1:
img = cv2.resize(img, (int(pic_width*resize_rate), int(pic_hight*resize_rate)), interpolation=cv2.INTER_LANCZOS4)
pic_hight, pic_width = img.shape[:2]
print("h,w:", pic_hight, pic_width)
blur = self.cfg["blur"]
#高斯去噪
if blur > 0:
img = cv2.GaussianBlur(img, (blur, blur), 0)#圖片分辨率調(diào)整
oldimg = img
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
#equ = cv2.equalizeHist(img)
#img = np.hstack((img, equ))
#去掉圖像中不會(huì)是車牌的區(qū)域
kernel = np.ones((20, 20), np.uint8)
img_opening = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)
img_opening = cv2.addWeighted(img, 1, img_opening, -1, 0);
#找到圖像邊緣
ret, img_thresh = cv2.threshold(img_opening, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
img_edge = cv2.Canny(img_thresh, 100, 200)
#使用開運(yùn)算和閉運(yùn)算讓圖像邊緣成為一個(gè)整體
kernel = np.ones((self.cfg["morphologyr"], self.cfg["morphologyc"]), np.uint8)
img_edge1 = cv2.morphologyEx(img_edge, cv2.MORPH_CLOSE, kernel)
img_edge2 = cv2.morphologyEx(img_edge1, cv2.MORPH_OPEN, kernel)
#查找圖像邊緣整體形成的矩形區(qū)域，可能有很多，車牌就在其中一個(gè)矩形區(qū)域中
try:
contours, hierarchy = cv2.findContours(img_edge2, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
except ValueError:
image, contours, hierarchy = cv2.findContours(img_edge2, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
contours = [cnt for cnt in contours if cv2.contourArea(cnt) > Min_Area]
print('len(contours)', len(contours))
#一一排除不是車牌的矩形區(qū)域
car_contours = []
for cnt in contours:
rect = cv2.minAreaRect(cnt)
area_width, area_height = rect[1]
if area_width < area_height:
area_width, area_height = area_height, area_width
wh_ratio = area_width / area_height
#print(wh_ratio)
#要求矩形區(qū)域長(zhǎng)寬比在2到5.5之間，2到5.5是車牌的長(zhǎng)寬比，其余的矩形排除
if wh_ratio > 2 and wh_ratio < 5.5:
car_contours.append(rect)
box = cv2.boxPoints(rect)
box = np.int0(box)
#oldimg = cv2.drawContours(oldimg, [box], 0, (0, 0, 255), 2)
#cv2.imshow("edge4", oldimg)
#cv2.waitKey(0)
print(len(car_contours))
print("精確定位")
card_imgs = []
#矩形區(qū)域可能是傾斜的矩形，需要矯正，以便使用顏色定位
for rect in car_contours:
if rect[2] > -1 and rect[2] < 1:#創(chuàng)造角度，使得左、高、右、低拿到正確的值
angle = 1
else:
angle = rect[2]
rect = (rect[0], (rect[1][0]+5, rect[1][1]+5), angle)#擴(kuò)大范圍，避免車牌邊緣被排除
box = cv2.boxPoints(rect)
heigth_point = right_point = [0, 0]
left_point = low_point = [pic_width, pic_hight]
for point in box:
if left_point[0] > point[0]:
left_point = point
if low_point[1] > point[1]:
low_point = point
if heigth_point[1] < point[1]:
heigth_point = point
if right_point[0] < point[0]:
right_point = point
if left_point[1] <= right_point[1]:#正角度
new_right_point = [right_point[0], heigth_point[1]]
pts2 = np.float32([left_point, heigth_point, new_right_point])#字符只是高度需要改變
pts1 = np.float32([left_point, heigth_point, right_point])
M = cv2.getAffineTransform(pts1, pts2)
dst = cv2.warpAffine(oldimg, M, (pic_width, pic_hight))
point_limit(new_right_point)
point_limit(heigth_point)
point_limit(left_point)
card_img = dst[int(left_point[1]):int(heigth_point[1]), int(left_point[0]):int(new_right_point[0])]
card_imgs.append(card_img)
#cv2.imshow("card", card_img)
#cv2.waitKey(0)
elif left_point[1] > right_point[1]:#負(fù)角度
new_left_point = [left_point[0], heigth_point[1]]
pts2 = np.float32([new_left_point, heigth_point, right_point])#字符只是高度需要改變
pts1 = np.float32([left_point, heigth_point, right_point])
M = cv2.getAffineTransform(pts1, pts2)
dst = cv2.warpAffine(oldimg, M, (pic_width, pic_hight))
point_limit(right_point)
point_limit(heigth_point)
point_limit(new_left_point)
card_img = dst[int(right_point[1]):int(heigth_point[1]), int(new_left_point[0]):int(right_point[0])]
card_imgs.append(card_img)
#cv2.imshow("card", card_img)
#cv2.waitKey(0)
#開始使用顏色定位，排除不是車牌的矩形，目前只識(shí)別藍(lán)、綠、黃車牌
colors = []
for card_index,card_img in enumerate(card_imgs):
green = yello = blue = black = white = 0
card_img_hsv = cv2.cvtColor(card_img, cv2.COLOR_BGR2HSV)
#有轉(zhuǎn)換失敗的可能，原因來自于上面矯正矩形出錯(cuò)
if card_img_hsv is None:
continue
row_num, col_num= card_img_hsv.shape[:2]
card_img_count = row_num * col_num
for i in range(row_num):
for j in range(col_num):
H = card_img_hsv.item(i, j, 0)
S = card_img_hsv.item(i, j, 1)
V = card_img_hsv.item(i, j, 2)
if 11 < H <= 34 and S > 34:#圖片分辨率調(diào)整
yello += 1
elif 35 < H <= 99 and S > 34:#圖片分辨率調(diào)整
green += 1
elif 99 < H <= 124 and S > 34:#圖片分辨率調(diào)整
blue += 1
if 0 < H <180 and 0 < S < 255 and 0 < V < 46:
black += 1
elif 0 < H <180 and 0 < S < 43 and 221 < V < 225:
white += 1
color = "no"
limit1 = limit2 = 0
if yello*2 >= card_img_count:
color = "yello"
limit1 = 11
limit2 = 34#有的圖片有色偏偏綠
elif green*2 >= card_img_count:
color = "green"
limit1 = 35
limit2 = 99
elif blue*2 >= card_img_count:
color = "blue"
limit1 = 100
limit2 = 124#有的圖片有色偏偏紫
elif black + white >= card_img_count*0.7:#TODO
color = "bw"
print(color)
colors.append(color)
print(blue, green, yello, black, white, card_img_count)
#cv2.imshow("color", card_img)
#cv2.waitKey(0)
if limit1 == 0:
continue
#以上為確定車牌顏色
#以下為根據(jù)車牌顏色再定位，縮小邊緣非車牌邊界
xl, xr, yh, yl = self.accurate_place(card_img_hsv, limit1, limit2, color)
if yl == yh and xl == xr:
continue
need_accurate = False
if yl >= yh:
yl = 0
yh = row_num
need_accurate = True
if xl >= xr:
xl = 0
xr = col_num
need_accurate = True
card_imgs[card_index] = card_img[yl:yh, xl:xr] if color != "green" or yl < (yh-yl)//4 else card_img[yl-(yh-yl)//4:yh, xl:xr]
if need_accurate:#可能x或y方向未縮小，需要再試一次
card_img = card_imgs[card_index]
card_img_hsv = cv2.cvtColor(card_img, cv2.COLOR_BGR2HSV)
xl, xr, yh, yl = self.accurate_place(card_img_hsv, limit1, limit2, color)
if yl == yh and xl == xr:
continue
if yl >= yh:
yl = 0
yh = row_num
if xl >= xr:
xl = 0
xr = col_num
card_imgs[card_index] = card_img[yl:yh, xl:xr] if color != "green" or yl < (yh-yl)//4 else card_img[yl-(yh-yl)//4:yh, xl:xr]
#以上為車牌定位
#以下為識(shí)別車牌中的字符
predict_result = []
roi = None
card_color = None
for i, color in enumerate(colors):
if color in ("blue", "yello", "green"):
card_img = card_imgs[i]
gray_img = cv2.cvtColor(card_img, cv2.COLOR_BGR2GRAY)
#黃、綠車牌字符比背景暗、與藍(lán)車牌剛好相反，所以黃、綠車牌需要反向
if color == "green" or color == "yello":
gray_img = cv2.bitwise_not(gray_img)
ret, gray_img = cv2.threshold(gray_img, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
#查找水平直方圖波峰
x_histogram = np.sum(gray_img, axis=1)
x_min = np.min(x_histogram)
x_average = np.sum(x_histogram)/x_histogram.shape[0]
x_threshold = (x_min + x_average)/2
wave_peaks = find_waves(x_threshold, x_histogram)
if len(wave_peaks) == 0:
print("peak less 0:")
continue
#認(rèn)為水平方向，最大的波峰為車牌區(qū)域
wave = max(wave_peaks, key=lambda x:x[1]-x[0])
gray_img = gray_img[wave[0]:wave[1]]
#查找垂直直方圖波峰
row_num, col_num= gray_img.shape[:2]
#去掉車牌上下邊緣1個(gè)像素，避免白邊影響閾值判斷
gray_img = gray_img[1:row_num-1]
y_histogram = np.sum(gray_img, axis=0)
y_min = np.min(y_histogram)
y_average = np.sum(y_histogram)/y_histogram.shape[0]
y_threshold = (y_min + y_average)/5#U和0要求閾值偏小，否則U和0會(huì)被分成兩半
wave_peaks = find_waves(y_threshold, y_histogram)
#for wave in wave_peaks:
# cv2.line(card_img, pt1=(wave[0], 5), pt2=(wave[1], 5), color=(0, 0, 255), thickness=2)
#車牌字符數(shù)應(yīng)大于6
if len(wave_peaks) <= 6:
print("peak less 1:", len(wave_peaks))
continue
wave = max(wave_peaks, key=lambda x:x[1]-x[0])
max_wave_dis = wave[1] - wave[0]
#判斷是否是左側(cè)車牌邊緣
if wave_peaks[0][1] - wave_peaks[0][0] < max_wave_dis/3 and wave_peaks[0][0] == 0:
wave_peaks.pop(0)
#組合分離漢字
cur_dis = 0
for i,wave in enumerate(wave_peaks):
if wave[1] - wave[0] + cur_dis > max_wave_dis * 0.6:
break
else:
cur_dis += wave[1] - wave[0]
if i > 0:
wave = (wave_peaks[0][0], wave_peaks[i][1])
wave_peaks = wave_peaks[i+1:]
wave_peaks.insert(0, wave)
#去除車牌上的分隔點(diǎn)
point = wave_peaks[2]
if point[1] - point[0] < max_wave_dis/3:
point_img = gray_img[:,point[0]:point[1]]
if np.mean(point_img) < 255/5:
wave_peaks.pop(2)
if len(wave_peaks) <= 6:
print("peak less 2:", len(wave_peaks))
continue
part_cards = seperate_card(gray_img, wave_peaks)
for i, part_card in enumerate(part_cards):
#可能是固定車牌的鉚釘
if np.mean(part_card) < 255/5:
print("a point")
continue
part_card_old = part_card
#w = abs(part_card.shape[1] - SZ)//2
w = part_card.shape[1] // 3
part_card = cv2.copyMakeBorder(part_card, 0, 0, w, w, cv2.BORDER_CONSTANT, value = [0,0,0])
part_card = cv2.resize(part_card, (SZ, SZ), interpolation=cv2.INTER_AREA)
#cv2.imshow("part", part_card_old)
#cv2.waitKey(0)
#cv2.imwrite("u.jpg", part_card)
#part_card = deskew(part_card)
part_card = preprocess_hog([part_card])
if i == 0:
resp = self.modelchinese.predict(part_card)
charactor = provinces[int(resp[0]) - PROVINCE_START]
else:
resp = self.model.predict(part_card)
charactor = chr(resp[0])
#判斷最后一個(gè)數(shù)是否是車牌邊緣，假設(shè)車牌邊緣被認(rèn)為是1
if charactor == "1" and i == len(part_cards)-1:
if part_card_old.shape[0]/part_card_old.shape[1] >= 8:#1太細(xì)，認(rèn)為是邊緣
print(part_card_old.shape)
continue
predict_result.append(charactor)
roi = card_img
card_color = color
break
return predict_result, roi, card_color#識(shí)別到的字符、定位的車牌圖像、車牌顏色
if __name__ == '__main__':
c = CardPredictor()
c.train_svm()
r, roi, color = c.predict("2.jpg")
print(r)
到了這里，關(guān)于車牌識(shí)別系統(tǒng) opencv的文章就介紹完了。如果您還想了解更多內(nèi)容，請(qǐng)?jiān)谟疑辖撬阉鱐OY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章，希望大家以后多多支持TOY模板網(wǎng)！
