百度PaddlePaddle >>> 8. 用卷積神經網絡輕鬆應付“車牌識別”

前言

繼上一篇的《百度PaddlePaddle >>> 7. 利用深度學習玩轉手勢識別》

這次來試試用卷積神經網絡進行車牌識別，這是所要用到的數據集：車牌識別字符數據集.zip

一、準備數據

1. 解壓數據

首先我們需要先解壓上述數據集的壓縮包，並將其中無關文件刪除（將目錄換成你的即可）：

unzip -q /home/aistudio/data/data23617/characterData.zip

2. 生成數據列表

從解壓的數據生成兩個數據列表：train_data.list 和test_data.list

#導入需要的包
import numpy as np
import paddle as paddle
import paddle.fluid as fluid
from PIL import Image
import cv2
import matplotlib.pyplot as plt
import os
from multiprocessing import cpu_count
from paddle.fluid.dygraph import Pool2D,Conv2D
# from paddle.fluid.dygraph import FC
from paddle.fluid.dygraph import Linear

# 生成車牌字符圖像列表
data_path = '/home/aistudio/data'
character_folders = os.listdir(data_path)
label = 0
LABEL_temp = {}
if(os.path.exists('./train_data.list')):
    os.remove('./train_data.list')
if(os.path.exists('./test_data.list')):
    os.remove('./test_data.list')
for character_folder in character_folders:
    with open('./train_data.list', 'a') as f_train:
        with open('./test_data.list', 'a') as f_test:
            if character_folder == '.DS_Store' or character_folder == '.ipynb_checkpoints' or character_folder == 'data23617':
                continue
            print(character_folder + " " + str(label))
            LABEL_temp[str(label)] = character_folder #存儲一下標籤的對應關係
            character_imgs = os.listdir(os.path.join(data_path, character_folder))
            for i in range(len(character_imgs)):
                if i%10 == 0: 
                    f_test.write(os.path.join(os.path.join(data_path, character_folder), character_imgs[i]) + "\t" + str(label) + '\n')
                else:
                    f_train.write(os.path.join(os.path.join(data_path, character_folder), character_imgs[i]) + "\t" + str(label) + '\n')
    label = label + 1
print('圖像列表已生成')

輸出：

3. 定義數據reader

# 用上一步生成的圖像列表定義車牌字符訓練集和測試集的reader
def data_mapper(sample):
    img, label = sample
    img = paddle.dataset.image.load_image(file=img, is_color=False)
    img = img.flatten().astype('float32') / 255.0
    return img, label
def data_reader(data_list_path):
    def reader():
        with open(data_list_path, 'r') as f:
            lines = f.readlines()
            for line in lines:
                img, label = line.split('\t')
                yield img, int(label)
    return paddle.reader.xmap_readers(data_mapper, reader, cpu_count(), 1024)

# 用於訓練的數據提供器
train_reader = paddle.batch(reader=paddle.reader.shuffle(reader=data_reader('./train_data.list'), buf_size=4000), batch_size=128)
# 用於測試的數據提供器
test_reader = paddle.batch(reader=data_reader('./test_data.list'), batch_size=128)

二、定義網絡

在這裏，我使用了卷積神經網絡：

#定義網絡
class MyLeNet(fluid.dygraph.Layer):
    def __init__(self, training=True):
        super(MyLeNet,self).__init__()
        self.conv1 = Conv2D(num_channels=1,num_filters=32,filter_size=3,act='relu')
        self.pool1 = Pool2D(pool_size=2,pool_stride=2,pool_type='max')
        self.conv2 = Conv2D(num_channels=32, num_filters=32, filter_size=2, act='relu')
        self.pool2 = Pool2D(pool_size=2, pool_stride=2,pool_type='max')
        self.conv3 = Conv2D(num_channels=32, num_filters=64, filter_size=3, act='relu')
        self.fc1 = Linear(input_dim=256,output_dim=1000,act='relu')
        self.drop_ratiol = 0.5 if training else 0.0
        self.fc2 = Linear(input_dim=1000, output_dim=65, act='softmax')
    def forward(self,input1):
        conv1 = self.conv1(input1)
        pool1 = self.pool1(conv1)
        conv2 = self.conv2(pool1)
        pool2 = self.pool2(conv2)
        conv3 = self.conv3(pool2)
        rs_1 = fluid.layers.reshape(conv3,[conv3.shape[0],-1])
        fc1 = self.fc1(rs_1)
        drop1 = fluid.layers.dropout(fc1,self.drop_ratiol)
        y = self.fc2(drop1)
        return y

該網絡一共7層，包括3層卷積層、2層池化層、2層全連接層。

該網絡經過多次訓練後正確率可達90%+

但事實證明，這個網絡還有很大的提升空間，你可以再加一層卷積層和池化層，經過實踐，這樣可以高達98%以上的正確率；
若再對學習率做一個類似分段函數的動態自調節，則最終識別率可達99%！

三、模型訓練

with fluid.dygraph.guard():
    model=MyLeNet(True) #模型實例化
    model.train() #訓練模式
    opt=fluid.optimizer.SGDOptimizer(learning_rate=0.01, parameter_list=model.parameters())#優化器選用SGD隨機梯度下降，學習率爲0.001.
    epochs_num= 120#迭代次數爲2
    
    for pass_num in range(epochs_num):
        for batch_id,data in enumerate(train_reader()):
            images=np.array([x[0].reshape(1,20,20) for x in data],np.float32)
            labels = np.array([x[1] for x in data]).astype('int64')
            labels = labels[:, np.newaxis]
            image=fluid.dygraph.to_variable(images)
            label=fluid.dygraph.to_variable(labels)
            
            predict=model(image)#預測

            loss=fluid.layers.cross_entropy(predict,label)
            avg_loss=fluid.layers.mean(loss)#獲取loss值
            
            acc=fluid.layers.accuracy(predict,label)#計算精度
            
            if batch_id!=0 and batch_id%50==0:
                print("train_pass:{},batch_id:{},train_loss:{},train_acc:{}".format(pass_num,batch_id,avg_loss.numpy(),acc.numpy()))
            
            avg_loss.backward()
            opt.minimize(avg_loss)
            model.clear_gradients()            
            
    fluid.save_dygraph(model.state_dict(),'MyLeNet')#保存模型

經過網絡優化後：

四、進行測試

#模型校驗
with fluid.dygraph.guard():
    accs = []
    model=MyLeNet()#模型實例化
    model_dict,_=fluid.load_dygraph('MyLeNet')
    model.load_dict(model_dict)#加載模型參數
    model.eval()#評估模式
    for batch_id,data in enumerate(test_reader()):#測試集
        images=np.array([x[0].reshape(1,20,20) for x in data],np.float32)
        labels = np.array([x[1] for x in data]).astype('int64')
        labels = labels[:, np.newaxis]
            
        image=fluid.dygraph.to_variable(images)
        label=fluid.dygraph.to_variable(labels)
            
        predict=model(image)#預測
        acc=fluid.layers.accuracy(predict,label)
        accs.append(acc.numpy()[0])
        avg_acc = np.mean(accs)
    print(avg_acc)

網絡優化後對測試集進行測試，正確率99%+

五、車牌識別

1. 字符分割

由於上述模型是以單個的車牌字符進行訓練的，而現實情況下這些字符都是一同出現於車牌中，所以我們需要將車牌進行字符分割：

# 對車牌圖片進行處理，分割出車牌中的每一個字符並保存
license_plate = cv2.imread('./車牌.png')
gray_plate = cv2.cvtColor(license_plate, cv2.COLOR_RGB2GRAY)
ret, binary_plate = cv2.threshold(gray_plate, 175, 255, cv2.THRESH_BINARY)
result = []
for col in range(binary_plate.shape[1]):
    result.append(0)
    for row in range(binary_plate.shape[0]):
        result[col] = result[col] + binary_plate[row][col]/255
character_dict = {}
num = 0
i = 0
while i < len(result):
    if result[i] == 0:
        i += 1
    else:
        index = i + 1
        while result[index] != 0:
            index += 1
        character_dict[num] = [i, index-1]
        num += 1
        i = index
for i in range(8):
    if i==2:
        continue
    padding = (170 - (character_dict[i][1] - character_dict[i][0])) / 2
    ndarray = np.pad(binary_plate[:,character_dict[i][0]:character_dict[i][1]], ((0,0), (int(padding), int(padding))), 'constant', constant_values=(0,0))
    ndarray = cv2.resize(ndarray, (20,20))
    cv2.imwrite('./' + str(i) + '.png', ndarray)
    
def load_image(path):
    img = paddle.dataset.image.load_image(file=path, is_color=False)
    img = img.astype('float32')
    img = img[np.newaxis, ] / 255.0
    return img

2. 轉換標籤

我們需要將上述的0-64的label 值中所代表漢字的值與漢字形成對應關係，可以存儲於字典中：

#將標籤進行轉換
print('Label:',LABEL_temp)
match = {'A':'A','B':'B','C':'C','D':'D','E':'E','F':'F','G':'G','H':'H','I':'I','J':'J','K':'K','L':'L','M':'M','N':'N',
        'O':'O','P':'P','Q':'Q','R':'R','S':'S','T':'T','U':'U','V':'V','W':'W','X':'X','Y':'Y','Z':'Z',
        'yun':'雲','cuan':'川','hei':'黑','zhe':'浙','ning':'寧','jin':'津','gan':'贛','hu':'滬','liao':'遼','jl':'吉','qing':'青','zang':'藏',
        'e1':'鄂','meng':'蒙','gan1':'甘','qiong':'瓊','shan':'陝','min':'閩','su':'蘇','xin':'新','wan':'皖','jing':'京','xiang':'湘','gui':'貴',
        'yu1':'渝','yu':'豫','ji':'冀','yue':'粵','gui1':'桂','sx':'晉','lu':'魯',
        '0':'0','1':'1','2':'2','3':'3','4':'4','5':'5','6':'6','7':'7','8':'8','9':'9'}
L = 0
LABEL ={}

for V in LABEL_temp.values():
    LABEL[str(L)] = match[V]
    L += 1
print(LABEL)

輸出：

3. 開始識別

#構建預測動態圖過程
with fluid.dygraph.guard():
    model=MyLeNet()#模型實例化
    model_dict,_=fluid.load_dygraph('MyLeNet')
    model.load_dict(model_dict)#加載模型參數
    model.eval()#評估模式
    lab=[]
    for i in range(8):
        if i==2:
            continue
        infer_imgs = []
        infer_imgs.append(load_image('./' + str(i) + '.png'))
        infer_imgs = np.array(infer_imgs)
        infer_imgs = fluid.dygraph.to_variable(infer_imgs)
        result=model(infer_imgs)
        lab.append(np.argmax(result.numpy()))
# print(lab)


display(Image.open('./車牌.png'))
print('\n車牌識別結果爲：',end='')
for i in range(len(lab)):
    print(LABEL[str(lab[i])],end='')

識別結果：

👇