Python+Tensorflow+CNN实现车牌识别的示例代码

一、项目概述

本次项目目标是实现对自动生成的带有各种噪声的车牌识别。在噪声干扰情况下，车牌字符分割较困难，此次车牌识别是将车牌7个字符同时训练，字符包括31个省份简称、10个阿拉伯数字、24个英文字母（'O'和'I'除外），共有65个类别，7个字符使用单独的loss函数进行训练。
（运行环境：tensorflow1.14.0-GPU版）

二、生成车牌数据集

import os
import cv2 as cv
import numpy as np
from math import *
from PIL import ImageFont
from PIL import Image
from PIL import ImageDraw

index = {"京": 0, "沪": 1, "津": 2, "渝": 3, "冀": 4, "晋": 5, "蒙": 6, "辽": 7, "吉": 8, "黑": 9,
       "苏": 10, "浙": 11, "皖": 12, "闽": 13, "赣": 14, "鲁": 15, "豫": 16, "鄂": 17, "湘": 18, "粤": 19,
       "桂": 20, "琼": 21, "川": 22, "贵": 23, "云": 24, "藏": 25, "陕": 26, "甘": 27, "青": 28, "宁": 29,
       "新": 30, "0": 31, "1": 32, "2": 33, "3": 34, "4": 35, "5": 36, "6": 37, "7": 38, "8": 39,
       "9": 40, "A": 41, "B": 42, "C": 43, "D": 44, "E": 45, "F": 46, "G": 47, "H": 48, "J": 49,
       "K": 50, "L": 51, "M": 52, "N": 53, "P": 54, "Q": 55, "R": 56, "S": 57, "T": 58, "U": 59,
       "V": 60, "W": 61, "X": 62, "Y": 63, "Z": 64}

chars = ["京", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑",
       "苏", "浙", "皖", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤",
       "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁",
       "新", "0", "1", "2", "3", "4", "5", "6", "7", "8",
       "9", "A", "B", "C", "D", "E", "F", "G", "H", "J",
       "K", "L", "M", "N", "P", "Q", "R", "S", "T", "U",
       "V", "W", "X", "Y", "Z"]

def AddSmudginess(img, Smu):
  """
  模糊处理
  :param img: 输入图像
  :param Smu: 模糊图像
  :return: 添加模糊后的图像
  """
  rows = r(Smu.shape[0] - 50)
  cols = r(Smu.shape[1] - 50)
  adder = Smu[rows:rows + 50, cols:cols + 50]
  adder = cv.resize(adder, (50, 50))
  img = cv.resize(img,(50,50))
  img = cv.bitwise_not(img)
  img = cv.bitwise_and(adder, img)
  img = cv.bitwise_not(img)
  return img

def rot(img, angel, shape, max_angel):
  """
  添加透视畸变
  """
  size_o = [shape[1], shape[0]]
  size = (shape[1]+ int(shape[0] * cos((float(max_angel ) / 180) * 3.14)), shape[0])
  interval = abs(int(sin((float(angel) / 180) * 3.14) * shape[0]))
  pts1 = np.float32([[0, 0], [0, size_o[1]], [size_o[0], 0], [size_o[0], size_o[1]]])
  if angel > 0:
    pts2 = np.float32([[interval, 0], [0, size[1]], [size[0], 0], [size[0] - interval, size_o[1]]])
  else:
    pts2 = np.float32([[0, 0], [interval, size[1]], [size[0] - interval, 0], [size[0], size_o[1]]])
  M = cv.getPerspectiveTransform(pts1, pts2)
  dst = cv.warpPerspective(img, M, size)
  return dst

def rotRandrom(img, factor, size):
  """
  添加放射畸变
  :param img: 输入图像
  :param factor: 畸变的参数
  :param size: 图片目标尺寸
  :return: 放射畸变后的图像
  """
  shape = size
  pts1 = np.float32([[0, 0], [0, shape[0]], [shape[1], 0], [shape[1], shape[0]]])
  pts2 = np.float32([[r(factor), r(factor)], [r(factor), shape[0] - r(factor)], [shape[1] - r(factor), r(factor)],
            [shape[1] - r(factor), shape[0] - r(factor)]])
  M = cv.getPerspectiveTransform(pts1, pts2)
  dst = cv.warpPerspective(img, M, size)
  return dst

def tfactor(img):
  """
  添加饱和度光照的噪声
  """
  hsv = cv.cvtColor(img,cv.COLOR_BGR2HSV)
  hsv[:, :, 0] = hsv[:, :, 0] * (0.8 + np.random.random() * 0.2)
  hsv[:, :, 1] = hsv[:, :, 1] * (0.3 + np.random.random() * 0.7)
  hsv[:, :, 2] = hsv[:, :, 2] * (0.2 + np.random.random() * 0.8)
  img = cv.cvtColor(hsv, cv.COLOR_HSV2BGR)
  return img

def random_envirment(img, noplate_bg):
  """
  添加自然环境的噪声, noplate_bg为不含车牌的背景图
  """
  bg_index = r(len(noplate_bg))
  env = cv.imread(noplate_bg[bg_index])
  env = cv.resize(env, (img.shape[1], img.shape[0]))
  bak = (img == 0)
  bak = bak.astype(np.uint8) * 255
  inv = cv.bitwise_and(bak, env)
  img = cv.bitwise_or(inv, img)
  return img

def GenCh(f, val):
  """
  生成中文字符
  """
  img = Image.new("RGB", (45, 70), (255, 255, 255))
  draw = ImageDraw.Draw(img)
  draw.text((0, 3), val, (0, 0, 0), font=f)
  img = img.resize((23, 70))
  A = np.array(img)
  return A

def GenCh1(f, val):
  """
  生成英文字符
  """
  img =Image.new("RGB", (23, 70), (255, 255, 255))
  draw = ImageDraw.Draw(img)
  draw.text((0, 2), val, (0, 0, 0), font=f)  # val.decode('utf-8')
  A = np.array(img)
  return A

def AddGauss(img, level):
  """
  添加高斯模糊
  """
  return cv.blur(img, (level * 2 + 1, level * 2 + 1))

def r(val):
  return int(np.random.random() * val)

def AddNoiseSingleChannel(single):
  """
  添加高斯噪声
  """
  diff = 255 - single.max()
  noise = np.random.normal(0, 1 + r(6), single.shape)
  noise = (noise - noise.min()) / (noise.max() - noise.min())
  noise *= diff
  # noise= noise.astype(np.uint8)
  dst = single + noise
  return dst

def addNoise(img):  # sdev = 0.5,avg=10
  img[:, :, 0] = AddNoiseSingleChannel(img[:, :, 0])
  img[:, :, 1] = AddNoiseSingleChannel(img[:, :, 1])
  img[:, :, 2] = AddNoiseSingleChannel(img[:, :, 2])
  return img

class GenPlate:
  def __init__(self, fontCh, fontEng, NoPlates):
    self.fontC = ImageFont.truetype(fontCh, 43, 0)
    self.fontE = ImageFont.truetype(fontEng, 60, 0)
    self.img = np.array(Image.new("RGB", (226, 70),(255, 255, 255)))
    self.bg = cv.resize(cv.imread("data\\images\\template.bmp"), (226, 70))  # template.bmp:车牌背景图
    self.smu = cv.imread("data\\images\\smu2.jpg")  # smu2.jpg:模糊图像
    self.noplates_path = []
    for parent, parent_folder, filenames in os.walk(NoPlates):
      for filename in filenames:
        path = parent + "\\" + filename
        self.noplates_path.append(path)

  def draw(self, val):
    offset = 2
    self.img[0:70, offset+8:offset+8+23] = GenCh(self.fontC, val[0])
    self.img[0:70, offset+8+23+6:offset+8+23+6+23] = GenCh1(self.fontE, val[1])
    for i in range(5):
      base = offset + 8 + 23 + 6 + 23 + 17 + i * 23 + i * 6
      self.img[0:70, base:base+23] = GenCh1(self.fontE, val[i+2])
    return self.img

  def generate(self, text):
    if len(text) == 7:
      fg = self.draw(text)  # decode(encoding="utf-8")
      fg = cv.bitwise_not(fg)
      com = cv.bitwise_or(fg, self.bg)
      com = rot(com, r(60)-30, com.shape,30)
      com = rotRandrom(com, 10, (com.shape[1], com.shape[0]))
      com = tfactor(com)
      com = random_envirment(com, self.noplates_path)
      com = AddGauss(com, 1+r(4))
      com = addNoise(com)
      return com

  @staticmethod
  def genPlateString(pos, val):
    """
	  生成车牌string，存为图片
    生成车牌list，存为label
    """
    plateStr = ""
    plateList=[]
    box = [0, 0, 0, 0, 0, 0, 0]
    if pos != -1:
      box[pos] = 1
    for unit, cpos in zip(box, range(len(box))):
      if unit == 1:
        plateStr += val
        plateList.append(val)
      else:
        if cpos == 0:
          plateStr += chars[r(31)]
          plateList.append(plateStr)
        elif cpos == 1:
          plateStr += chars[41 + r(24)]
          plateList.append(plateStr)
        else:
          plateStr += chars[31 + r(34)]
          plateList.append(plateStr)
    plate = [plateList[0]]
    b = [plateList[i][-1] for i in range(len(plateList))]
    plate.extend(b[1:7])
    return plateStr, plate

  @staticmethod
  def genBatch(batchsize, outputPath, size):
    """
    将生成的车牌图片写入文件夹，对应的label写入label.txt
    :param batchsize: 批次大小
    :param outputPath: 输出图像的保存路径
    :param size: 输出图像的尺寸
    :return: None
    """
    if not os.path.exists(outputPath):
      os.mkdir(outputPath)
    outfile = open('data\\plate\\label.txt', 'w', encoding='utf-8')
    for i in range(batchsize):
      plateStr, plate = G.genPlateString(-1, -1)
      # print(plateStr, plate)
      img = G.generate(plateStr)
      img = cv.resize(img, size)
      cv.imwrite(outputPath + "\\" + str(i).zfill(2) + ".jpg", img)
      outfile.write(str(plate) + "\n")

if __name__ == '__main__':
  G = GenPlate("data\\font\\platech.ttf", 'data\\font\\platechar.ttf', "data\\NoPlates")
  G.genBatch(101, 'data\\plate', (272, 72))

生成的车牌图像尺寸尽量不要超过300，本次尺寸选取：272 * 72

生成车牌所需文件：

• 字体文件：中文‘platech.ttf'，英文及数字‘platechar.ttf'
• 背景图：来源于不含车牌的车辆裁剪图片
• 车牌（蓝底）：template.bmp
• 噪声图像：smu2.jpg

车牌生成后保存至plate文件夹，示例如下：

三、数据导入

from genplate import *
import matplotlib.pyplot as plt

# 产生用于训练的数据
class OCRIter:
  def __init__(self, batch_size, width, height):
    super(OCRIter, self).__init__()
    self.genplate = GenPlate("data\\font\\platech.ttf", 'data\\font\\platechar.ttf', "data\\NoPlates")
    self.batch_size = batch_size
    self.height = height
    self.width = width

  def iter(self):
    data = []
    label = []
    for i in range(self.batch_size):
      img, num = self.gen_sample(self.genplate, self.width, self.height)
      data.append(img)
      label.append(num)
    return np.array(data), np.array(label)

  @staticmethod
  def rand_range(lo, hi):
    return lo + r(hi - lo)

  def gen_rand(self):
    name = ""
    label = list([])
    label.append(self.rand_range(0, 31))  #产生车牌开头32个省的标签
    label.append(self.rand_range(41, 65))  #产生车牌第二个字母的标签
    for i in range(5):
      label.append(self.rand_range(31, 65))  #产生车牌后续5个字母的标签
    name += chars[label[0]]
    name += chars[label[1]]
    for i in range(5):
      name += chars[label[i+2]]
    return name, label

  def gen_sample(self, genplate, width, height):
    num, label = self.gen_rand()
    img = genplate.generate(num)
    img = cv.resize(img, (height, width))
    img = np.multiply(img, 1/255.0)
    return img, label    #返回的label为标签，img为车牌图像

'''
# 测试代码
O = OCRIter(2, 272, 72)
img, lbl = O.iter()
for im in img:
  plt.imshow(im, cmap='gray')
  plt.show()
print(img.shape)
print(lbl)
'''

四、CNN模型构建

import tensorflow as tf

def cnn_inference(images, keep_prob):
  W_conv = {
    'conv1': tf.Variable(tf.random.truncated_normal([3, 3, 3, 32],
                            stddev=0.1)),
    'conv2': tf.Variable(tf.random.truncated_normal([3, 3, 32, 32],
                            stddev=0.1)),
    'conv3': tf.Variable(tf.random.truncated_normal([3, 3, 32, 64],
                            stddev=0.1)),
    'conv4': tf.Variable(tf.random.truncated_normal([3, 3, 64, 64],
                            stddev=0.1)),
    'conv5': tf.Variable(tf.random.truncated_normal([3, 3, 64, 128],
                            stddev=0.1)),
    'conv6': tf.Variable(tf.random.truncated_normal([3, 3, 128, 128],
                            stddev=0.1)),
    'fc1_1': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_2': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_3': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_4': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_5': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_6': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    'fc1_7': tf.Variable(tf.random.truncated_normal([5*30*128, 65],
                            stddev=0.01)),
    } 

  b_conv = {
    'conv1': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[32])),
    'conv2': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[32])),
    'conv3': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[64])),
    'conv4': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[64])),
    'conv5': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[128])),
    'conv6': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[128])),
    'fc1_1': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_2': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_3': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_4': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_5': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_6': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    'fc1_7': tf.Variable(tf.constant(0.1, dtype=tf.float32,
                     shape=[65])),
    } 

  # 第1层卷积层
  conv1 = tf.nn.conv2d(images, W_conv['conv1'], strides=[1,1,1,1], padding='VALID')
  conv1 = tf.nn.bias_add(conv1, b_conv['conv1'])
  conv1 = tf.nn.relu(conv1)

  # 第2层卷积层
  conv2 = tf.nn.conv2d(conv1, W_conv['conv2'], strides=[1,1,1,1], padding='VALID')
  conv2 = tf.nn.bias_add(conv2, b_conv['conv2'])
  conv2 = tf.nn.relu(conv2)
  # 第1层池化层
  pool1 = tf.nn.max_pool2d(conv2, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')

  # 第3层卷积层
  conv3 = tf.nn.conv2d(pool1, W_conv['conv3'], strides=[1,1,1,1], padding='VALID')
  conv3 = tf.nn.bias_add(conv3, b_conv['conv3'])
  conv3 = tf.nn.relu(conv3)

  # 第4层卷积层
  conv4 = tf.nn.conv2d(conv3, W_conv['conv4'], strides=[1,1,1,1], padding='VALID')
  conv4 = tf.nn.bias_add(conv4, b_conv['conv4'])
  conv4 = tf.nn.relu(conv4)
  # 第2层池化层
  pool2 = tf.nn.max_pool2d(conv4, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')

  # 第5层卷积层
  conv5 = tf.nn.conv2d(pool2, W_conv['conv5'], strides=[1,1,1,1], padding='VALID')
  conv5 = tf.nn.bias_add(conv5, b_conv['conv5'])
  conv5 = tf.nn.relu(conv5)

  # 第4层卷积层
  conv6 = tf.nn.conv2d(conv5, W_conv['conv6'], strides=[1,1,1,1], padding='VALID')
  conv6 = tf.nn.bias_add(conv6, b_conv['conv6'])
  conv6 = tf.nn.relu(conv6)
  # 第3层池化层
  pool3 = tf.nn.max_pool2d(conv6, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')

  #第1_1层全连接层
  # print(pool3.shape)
  reshape = tf.reshape(pool3, [-1, 5 * 30 * 128])
  fc1 = tf.nn.dropout(reshape, keep_prob)
  fc1_1 = tf.add(tf.matmul(fc1, W_conv['fc1_1']), b_conv['fc1_1'])

  #第1_2层全连接层
  fc1_2 = tf.add(tf.matmul(fc1, W_conv['fc1_2']), b_conv['fc1_2'])

  #第1_3层全连接层
  fc1_3 = tf.add(tf.matmul(fc1, W_conv['fc1_3']), b_conv['fc1_3'])

  #第1_4层全连接层
  fc1_4 = tf.add(tf.matmul(fc1, W_conv['fc1_4']), b_conv['fc1_4'])

  #第1_5层全连接层
  fc1_5 = tf.add(tf.matmul(fc1, W_conv['fc1_5']), b_conv['fc1_5'])

  #第1_6层全连接层
  fc1_6 = tf.add(tf.matmul(fc1, W_conv['fc1_6']), b_conv['fc1_6'])

  #第1_7层全连接层
  fc1_7 = tf.add(tf.matmul(fc1, W_conv['fc1_7']), b_conv['fc1_7'])

  return fc1_1, fc1_2, fc1_3, fc1_4, fc1_5, fc1_6, fc1_7

def calc_loss(logit1, logit2, logit3, logit4, logit5, logit6, logit7, labels):
  labels = tf.convert_to_tensor(labels, tf.int32)

  loss1 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit1, labels=labels[:, 0]))
  tf.compat.v1.summary.scalar('loss1', loss1)

  loss2 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit2, labels=labels[:, 1]))
  tf.compat.v1.summary.scalar('loss2', loss2)

  loss3 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit3, labels=labels[:, 2]))
  tf.compat.v1.summary.scalar('loss3', loss3)

  loss4 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit4, labels=labels[:, 3]))
  tf.compat.v1.summary.scalar('loss4', loss4)

  loss5 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit5, labels=labels[:, 4]))
  tf.compat.v1.summary.scalar('loss5', loss5)

  loss6 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit6, labels=labels[:, 5]))
  tf.compat.v1.summary.scalar('loss6', loss6)

  loss7 = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
    logits=logit7, labels=labels[:, 6]))
  tf.compat.v1.summary.scalar('loss7', loss7)

  return loss1, loss2, loss3, loss4, loss5, loss6, loss7

def train_step(loss1, loss2, loss3, loss4, loss5, loss6, loss7, learning_rate):
  optimizer1 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op1 = optimizer1.minimize(loss1)

  optimizer2 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op2 = optimizer2.minimize(loss2)

  optimizer3 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op3 = optimizer3.minimize(loss3)

  optimizer4 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op4 = optimizer4.minimize(loss4)

  optimizer5 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op5 = optimizer5.minimize(loss5)

  optimizer6 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op6 = optimizer6.minimize(loss6)

  optimizer7 = tf.compat.v1.train.AdamOptimizer(learning_rate=learning_rate)
  train_op7 = optimizer7.minimize(loss7)

  return train_op1, train_op2, train_op3, train_op4, train_op5, train_op6, train_op7

def pred_model(logit1, logit2, logit3, logit4, logit5, logit6, logit7, labels):
  labels = tf.convert_to_tensor(labels, tf.int32)
  labels = tf.reshape(tf.transpose(labels), [-1])
  logits = tf.concat([logit1, logit2, logit3, logit4, logit5, logit6, logit7], 0)
  prediction = tf.nn.in_top_k(logits, labels, 1)
  accuracy = tf.reduce_mean(tf.cast(prediction, tf.float32))
  tf.compat.v1.summary.scalar('accuracy', accuracy)
  return accuracy

五、模型训练

import os
import time
import datetime
import numpy as np
import tensorflow as tf
from input_data import OCRIter
import model

os.environ["TF_CPP_MIN_LOG_LEVEL"] = '3'

img_h = 72
img_w = 272
num_label = 7
batch_size = 32
epoch = 10000
learning_rate = 0.0001

logs_path = 'logs\\1005'
model_path = 'saved_model\\1005'

image_holder = tf.compat.v1.placeholder(tf.float32, [batch_size, img_h, img_w, 3])
label_holder = tf.compat.v1.placeholder(tf.int32, [batch_size, 7])
keep_prob = tf.compat.v1.placeholder(tf.float32)

def get_batch():
  data_batch = OCRIter(batch_size, img_h, img_w)
  image_batch, label_batch = data_batch.iter()
  return np.array(image_batch), np.array(label_batch)

logit1, logit2, logit3, logit4, logit5, logit6, logit7 = model.cnn_inference(
  image_holder, keep_prob)

loss1, loss2, loss3, loss4, loss5, loss6, loss7 = model.calc_loss(
  logit1, logit2, logit3, logit4, logit5, logit6, logit7, label_holder)

train_op1, train_op2, train_op3, train_op4, train_op5, train_op6, train_op7 = model.train_step(
  loss1, loss2, loss3, loss4, loss5, loss6, loss7, learning_rate)

accuracy = model.pred_model(logit1, logit2, logit3, logit4, logit5, logit6, logit7, label_holder)

input_image=tf.compat.v1.summary.image('input', image_holder)

summary_op = tf.compat.v1.summary.merge(tf.compat.v1.get_collection(tf.compat.v1.GraphKeys.SUMMARIES))

init_op = tf.compat.v1.global_variables_initializer()

with tf.compat.v1.Session() as sess:
  sess.run(init_op)

  train_writer = tf.compat.v1.summary.FileWriter(logs_path, sess.graph)
  saver = tf.compat.v1.train.Saver()

  start_time1 = time.time()
  for step in range(epoch):
    # 生成车牌图像以及标签数据
    img_batch, lbl_batch = get_batch()

    start_time2 = time.time()
    time_str = datetime.datetime.now().isoformat()

    feed_dict = {image_holder:img_batch, label_holder:lbl_batch, keep_prob:0.6}
    _1, _2, _3, _4, _5, _6, _7, ls1, ls2, ls3, ls4, ls5, ls6, ls7, acc = sess.run(
      [train_op1, train_op2, train_op3, train_op4, train_op5, train_op6, train_op7,
       loss1, loss2, loss3, loss4, loss5, loss6, loss7, accuracy], feed_dict)
    summary_str = sess.run(summary_op, feed_dict)
    train_writer.add_summary(summary_str,step)
    duration = time.time() - start_time2
    loss_total = ls1 + ls2 + ls3 + ls4 + ls5 + ls6 + ls7
    if step % 10 == 0:
      sec_per_batch = float(duration)
      print('%s: Step %d, loss_total = %.2f, acc = %.2f%%, sec/batch = %.2f' %
        (time_str, step, loss_total, acc * 100, sec_per_batch))
    if step % 5000 == 0 or (step + 1) == epoch:
      checkpoint_path = os.path.join(model_path,'model.ckpt')
      saver.save(sess, checkpoint_path, global_step=step)
  end_time = time.time()
  print("Training over. It costs {:.2f} minutes".format((end_time - start_time1) / 60))

六、训练结果展示

训练参数：
batch_size = 32
epoch = 10000
learning_rate = 0.0001
在tensorboard中查看训练过程
accuracy :

accuracy

曲线在epoch = 10000左右时达到收敛，最终精确度在94%左右

loss :

以上三张分别是loss1，loss2， loss7的曲线图像，一号位字符是省份简称，识别相对字母数字较难，loss1=0.08左右，二号位字符是字母，loss2稳定在0.001左右，但是随着字符往后，loss值也将越来越大，7号位字符loss7稳定在0.6左右。

七、预测单张车牌

import os
import cv2 as cv
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
import model

os.environ["TF_CPP_MIN_LOG_LEVEL"] = '3' # 只显示 Error

index = {"京": 0, "沪": 1, "津": 2, "渝": 3, "冀": 4, "晋": 5, "蒙": 6, "辽": 7, "吉": 8, "黑": 9,
       "苏": 10, "浙": 11, "皖": 12, "闽": 13, "赣": 14, "鲁": 15, "豫": 16, "鄂": 17, "湘": 18, "粤": 19,
       "桂": 20, "琼": 21, "川": 22, "贵": 23, "云": 24, "藏": 25, "陕": 26, "甘": 27, "青": 28, "宁": 29,
       "新": 30, "0": 31, "1": 32, "2": 33, "3": 34, "4": 35, "5": 36, "6": 37, "7": 38, "8": 39,
       "9": 40, "A": 41, "B": 42, "C": 43, "D": 44, "E": 45, "F": 46, "G": 47, "H": 48, "J": 49,
       "K": 50, "L": 51, "M": 52, "N": 53, "P": 54, "Q": 55, "R": 56, "S": 57, "T": 58, "U": 59,
       "V": 60, "W": 61, "X": 62, "Y": 63, "Z": 64}

chars = ["京", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑",
       "苏", "浙", "皖", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤",
       "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁",
       "新", "0", "1", "2", "3", "4", "5", "6", "7", "8",
       "9", "A", "B", "C", "D", "E", "F", "G", "H", "J",
       "K", "L", "M", "N", "P", "Q", "R", "S", "T", "U",
       "V", "W", "X", "Y", "Z"]

def get_one_image(test):
  """ 随机获取单张车牌图像 """
  n = len(test)
  rand_num =np.random.randint(0,n)
  img_dir = test[rand_num]
  image_show = Image.open(img_dir)
  plt.imshow(image_show)  # 显示车牌图片
  image = cv.imread(img_dir)
  image = image.reshape(72, 272, 3)
  image = np.multiply(image, 1 / 255.0)
  return image

batch_size = 1
x = tf.compat.v1.placeholder(tf.float32, [batch_size, 72, 272, 3])
keep_prob = tf.compat.v1.placeholder(tf.float32)

test_dir = 'data\\plate\\'
test_image = []
for file in os.listdir(test_dir):
  test_image.append(test_dir + file)
test_image = list(test_image)

image_array = get_one_image(test_image)

logit1, logit2, logit3, logit4, logit5, logit6, logit7 = model.cnn_inference(x, keep_prob)

model_path = 'saved_model\\1005'

saver = tf.compat.v1.train.Saver()

with tf.compat.v1.Session() as sess:
  print ("Reading checkpoint...")
  ckpt = tf.train.get_checkpoint_state(model_path)
  if ckpt and ckpt.model_checkpoint_path:
    global_step = ckpt.model_checkpoint_path.split('/')[-1].split('-')[-1]
    saver.restore(sess, ckpt.model_checkpoint_path)
    print('Loading success, global_step is %s' % global_step)
  else:
    print('No checkpoint file found')

  pre1, pre2, pre3, pre4, pre5, pre6, pre7 = sess.run(
    [logit1, logit2, logit3, logit4, logit5, logit6, logit7],
    feed_dict={x:image_array, keep_prob:1.0})
  prediction = np.reshape(np.array([pre1, pre2, pre3, pre4, pre5, pre6, pre7]), [-1, 65])

  max_index = np.argmax(prediction, axis=1)
  print(max_index)
  line = ''
  result = np.array([])
  for i in range(prediction.shape[0]):
    if i == 0:
      result = np.argmax(prediction[i][0:31])
    if i == 1:
      result = np.argmax(prediction[i][41:65]) + 41
    if i > 1:
      result = np.argmax(prediction[i][31:65]) + 31
    line += chars[result]+" "
  print ('predicted: ' + line)
plt.show()

随机测试20张车牌，18张预测正确，2张预测错误，从最后两幅预测错误的图片可以看出，模型对相似字符以及遮挡字符识别成功率仍有待提高。测试结果部分展示如下：

八、总结

本次构建的CNN模型较为简单，只有6卷积层+3池化层+1全连接层，可以通过增加模型深度以及每层之间的神经元数量来优化模型，提高识别的准确率。此次训练数据集来源于自动生成的车牌，由于真实的车牌图像与生成的车牌图像在噪声干扰上有所区分，所以识别率上会有所出入。如果使用真实的车牌数据集，需要对车牌进行滤波、均衡化、腐蚀、矢量量化等预处理方法。

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持我们。