更新接口
This commit is contained in:
		@@ -282,14 +282,15 @@ def LPRNmodel_predict(image_array):
 | 
				
			|||||||
        image_array: numpy数组格式的车牌图像,已经过矫正处理
 | 
					        image_array: numpy数组格式的车牌图像,已经过矫正处理
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    返回:
 | 
					    返回:
 | 
				
			||||||
        list: 包含7个字符的列表,代表车牌号的每个字符
 | 
					        list: 包含最多8个字符的列表,代表车牌号的每个字符
 | 
				
			||||||
              例如: ['京', 'A', '1', '2', '3', '4', '5']
 | 
					              例如: ['京', 'A', '1', '2', '3', '4', '5', ''] (蓝牌7位+占位符)
 | 
				
			||||||
 | 
					                   ['京', 'A', 'D', '1', '2', '3', '4', '5'] (绿牌8位)
 | 
				
			||||||
    """
 | 
					    """
 | 
				
			||||||
    global crnn_model, crnn_decoder, crnn_preprocessor, device
 | 
					    global crnn_model, crnn_decoder, crnn_preprocessor, device
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    if crnn_model is None or crnn_decoder is None or crnn_preprocessor is None:
 | 
					    if crnn_model is None or crnn_decoder is None or crnn_preprocessor is None:
 | 
				
			||||||
        print("CRNN模型未初始化,请先调用initialize_crnn_model()")
 | 
					        print("CRNN模型未初始化,请先调用initialize_crnn_model()")
 | 
				
			||||||
        return ['待', '识', '别', '0', '0', '0', '0']
 | 
					        return ['待', '识', '别', '0', '0', '0', '0', '0']
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    try:
 | 
					    try:
 | 
				
			||||||
        # 预处理图像
 | 
					        # 预处理图像
 | 
				
			||||||
@@ -314,13 +315,17 @@ def LPRNmodel_predict(image_array):
 | 
				
			|||||||
            # 将字符串转换为字符列表
 | 
					            # 将字符串转换为字符列表
 | 
				
			||||||
            char_list = list(predicted_text)
 | 
					            char_list = list(predicted_text)
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
            # 确保返回7个字符(车牌标准长度)
 | 
					            # 确保返回至少7个字符,最多8个字符
 | 
				
			||||||
            if len(char_list) < 7:
 | 
					            if len(char_list) < 7:
 | 
				
			||||||
                # 如果识别结果少于7个字符,用'0'补齐
 | 
					                # 如果识别结果少于7个字符,用'0'补齐到7位
 | 
				
			||||||
                char_list.extend(['0'] * (7 - len(char_list)))
 | 
					                char_list.extend(['0'] * (7 - len(char_list)))
 | 
				
			||||||
            elif len(char_list) > 7:
 | 
					            elif len(char_list) > 8:
 | 
				
			||||||
                # 如果识别结果多于7个字符,截取前7个
 | 
					                # 如果识别结果多于8个字符,截取前8个
 | 
				
			||||||
                char_list = char_list[:7]
 | 
					                char_list = char_list[:8]
 | 
				
			||||||
 | 
					            
 | 
				
			||||||
 | 
					            # 如果是7位,补齐到8位以保持接口一致性(第8位用空字符或占位符)
 | 
				
			||||||
 | 
					            if len(char_list) == 7:
 | 
				
			||||||
 | 
					                char_list.append('')  # 添加空字符作为第8位占位符
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
            return char_list
 | 
					            return char_list
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
@@ -328,4 +333,4 @@ def LPRNmodel_predict(image_array):
 | 
				
			|||||||
        print(f"CRNN识别失败: {e}")
 | 
					        print(f"CRNN识别失败: {e}")
 | 
				
			||||||
        import traceback
 | 
					        import traceback
 | 
				
			||||||
        traceback.print_exc()
 | 
					        traceback.print_exc()
 | 
				
			||||||
        return ['识', '别', '失', '败', '0', '0', '0']
 | 
					        return ['识', '别', '失', '败', '0', '0', '0', '0']
 | 
				
			||||||
 
 | 
				
			|||||||
@@ -1,28 +1,27 @@
 | 
				
			|||||||
import torch
 | 
					import torch
 | 
				
			||||||
import torch.nn as nn
 | 
					import torch.nn as nn
 | 
				
			||||||
import numpy as np
 | 
					 | 
				
			||||||
import cv2
 | 
					import cv2
 | 
				
			||||||
from torch.autograd import Variable
 | 
					import numpy as np
 | 
				
			||||||
import os
 | 
					import os
 | 
				
			||||||
 | 
					import sys
 | 
				
			||||||
 | 
					from torch.autograd import Variable
 | 
				
			||||||
 | 
					from PIL import Image
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					# 添加父目录到路径,以便导入模型和数据加载器
 | 
				
			||||||
 | 
					sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 | 
				
			||||||
 | 
					
 | 
				
			||||||
# 字符集定义
 | 
					# LPRNet字符集定义(与训练时保持一致)
 | 
				
			||||||
CHARS = ['京', '沪', '津', '渝', '冀', '晋', '蒙', '辽', '吉', '黑',
 | 
					CHARS = ['京', '沪', '津', '渝', '冀', '晋', '蒙', '辽', '吉', '黑',
 | 
				
			||||||
         '苏', '浙', '皖', '闽', '赣', '鲁', '豫', '鄂', '湘', '粤',
 | 
					         '苏', '浙', '皖', '闽', '赣', '鲁', '豫', '鄂', '湘', '粤',
 | 
				
			||||||
         '桂', '琼', '川', '贵', '云', '藏', '陕', '甘', '青', '宁',
 | 
					         '桂', '琼', '川', '贵', '云', '藏', '陕', '甘', '青', '宁', '新',
 | 
				
			||||||
         '新',
 | 
					 | 
				
			||||||
         '0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
 | 
					         '0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
 | 
				
			||||||
         'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K',
 | 
					         'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K',
 | 
				
			||||||
         'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V',
 | 
					         'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V',
 | 
				
			||||||
         'W', 'X', 'Y', 'Z', 'I', 'O', '-'
 | 
					         'W', 'X', 'Y', 'Z', 'I', 'O', '-']
 | 
				
			||||||
         ]
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
CHARS_DICT = {char: i for i, char in enumerate(CHARS)}
 | 
					CHARS_DICT = {char: i for i, char in enumerate(CHARS)}
 | 
				
			||||||
 | 
					
 | 
				
			||||||
# 全局变量
 | 
					# 简化的LPRNet模型定义
 | 
				
			||||||
lprnet_model = None
 | 
					 | 
				
			||||||
device = None
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
class small_basic_block(nn.Module):
 | 
					class small_basic_block(nn.Module):
 | 
				
			||||||
    def __init__(self, ch_in, ch_out):
 | 
					    def __init__(self, ch_in, ch_out):
 | 
				
			||||||
        super(small_basic_block, self).__init__()
 | 
					        super(small_basic_block, self).__init__()
 | 
				
			||||||
@@ -58,8 +57,8 @@ class LPRNet(nn.Module):
 | 
				
			|||||||
            nn.BatchNorm2d(num_features=256),
 | 
					            nn.BatchNorm2d(num_features=256),
 | 
				
			||||||
            nn.ReLU(),  # 10
 | 
					            nn.ReLU(),  # 10
 | 
				
			||||||
            small_basic_block(ch_in=256, ch_out=256),   # *** 11 ***
 | 
					            small_basic_block(ch_in=256, ch_out=256),   # *** 11 ***
 | 
				
			||||||
            nn.BatchNorm2d(num_features=256),   # 12
 | 
					            nn.BatchNorm2d(num_features=256),
 | 
				
			||||||
            nn.ReLU(),
 | 
					            nn.ReLU(),  # 13
 | 
				
			||||||
            nn.MaxPool3d(kernel_size=(1, 3, 3), stride=(4, 1, 2)),  # 14
 | 
					            nn.MaxPool3d(kernel_size=(1, 3, 3), stride=(4, 1, 2)),  # 14
 | 
				
			||||||
            nn.Dropout(dropout_rate),
 | 
					            nn.Dropout(dropout_rate),
 | 
				
			||||||
            nn.Conv2d(in_channels=64, out_channels=256, kernel_size=(1, 4), stride=1), # 16
 | 
					            nn.Conv2d(in_channels=64, out_channels=256, kernel_size=(1, 4), stride=1), # 16
 | 
				
			||||||
@@ -68,7 +67,7 @@ class LPRNet(nn.Module):
 | 
				
			|||||||
            nn.Dropout(dropout_rate),
 | 
					            nn.Dropout(dropout_rate),
 | 
				
			||||||
            nn.Conv2d(in_channels=256, out_channels=class_num, kernel_size=(13, 1), stride=1), # 20
 | 
					            nn.Conv2d(in_channels=256, out_channels=class_num, kernel_size=(13, 1), stride=1), # 20
 | 
				
			||||||
            nn.BatchNorm2d(num_features=class_num),
 | 
					            nn.BatchNorm2d(num_features=class_num),
 | 
				
			||||||
            nn.ReLU(),  # *** 22 ***
 | 
					            nn.ReLU(),  # 22
 | 
				
			||||||
        )
 | 
					        )
 | 
				
			||||||
        self.container = nn.Sequential(
 | 
					        self.container = nn.Sequential(
 | 
				
			||||||
            nn.Conv2d(in_channels=448+self.class_num, out_channels=self.class_num, kernel_size=(1,1), stride=(1,1)),
 | 
					            nn.Conv2d(in_channels=448+self.class_num, out_channels=self.class_num, kernel_size=(1,1), stride=(1,1)),
 | 
				
			||||||
@@ -98,101 +97,177 @@ class LPRNet(nn.Module):
 | 
				
			|||||||
 | 
					
 | 
				
			||||||
        return logits
 | 
					        return logits
 | 
				
			||||||
 | 
					
 | 
				
			||||||
def build_lprnet(lpr_max_len=8, phase=False, class_num=66, dropout_rate=0.5):
 | 
					class LPRNetInference:
 | 
				
			||||||
    """构建LPRNet模型"""
 | 
					    def __init__(self, model_path=None, img_size=[94, 24], lpr_max_len=8, dropout_rate=0.5):
 | 
				
			||||||
    Net = LPRNet(lpr_max_len, phase, class_num, dropout_rate)
 | 
					        """
 | 
				
			||||||
 | 
					        初始化LPRNet推理类
 | 
				
			||||||
 | 
					        Args:
 | 
				
			||||||
 | 
					            model_path: 训练好的模型权重文件路径
 | 
				
			||||||
 | 
					            img_size: 输入图像尺寸 [width, height]
 | 
				
			||||||
 | 
					            lpr_max_len: 车牌最大长度
 | 
				
			||||||
 | 
					            dropout_rate: dropout率
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        self.img_size = img_size
 | 
				
			||||||
 | 
					        self.lpr_max_len = lpr_max_len
 | 
				
			||||||
 | 
					        self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    if phase == "train":
 | 
					        # 设置默认模型路径
 | 
				
			||||||
        return Net.train()
 | 
					        if model_path is None:
 | 
				
			||||||
 | 
					            current_dir = os.path.dirname(os.path.abspath(__file__))
 | 
				
			||||||
 | 
					            model_path = os.path.join(current_dir, 'LPRNet__iteration_74000.pth')
 | 
				
			||||||
 | 
					        
 | 
				
			||||||
 | 
					        # 初始化模型
 | 
				
			||||||
 | 
					        self.model = LPRNet(lpr_max_len=lpr_max_len, phase=False, class_num=len(CHARS), dropout_rate=dropout_rate)
 | 
				
			||||||
 | 
					        
 | 
				
			||||||
 | 
					        # 加载模型权重
 | 
				
			||||||
 | 
					        if model_path and os.path.exists(model_path):
 | 
				
			||||||
 | 
					            print(f"Loading LPRNet model from {model_path}")
 | 
				
			||||||
 | 
					            try:
 | 
				
			||||||
 | 
					                self.model.load_state_dict(torch.load(model_path, map_location=self.device))
 | 
				
			||||||
 | 
					                print("LPRNet模型权重加载成功")
 | 
				
			||||||
 | 
					            except Exception as e:
 | 
				
			||||||
 | 
					                print(f"Warning: 加载模型权重失败: {e}. 使用随机权重.")
 | 
				
			||||||
        else:
 | 
					        else:
 | 
				
			||||||
        return Net.eval()
 | 
					            print(f"Warning: 模型文件不存在或未指定: {model_path}. 使用随机权重.")
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
def preprocess_image(image_array, img_size=(94, 24)):
 | 
					        self.model.to(self.device)
 | 
				
			||||||
    """图像预处理"""
 | 
					        self.model.eval()
 | 
				
			||||||
    # 确保输入是numpy数组
 | 
					        
 | 
				
			||||||
 | 
					        print(f"LPRNet模型加载完成,设备: {self.device}")
 | 
				
			||||||
 | 
					        print(f"模型参数数量: {sum(p.numel() for p in self.model.parameters()):,}")
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    def preprocess_image(self, image_array):
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        预处理图像数组 - 使用与训练时相同的预处理方式
 | 
				
			||||||
 | 
					        Args:
 | 
				
			||||||
 | 
					            image_array: numpy数组格式的图像 (H, W, C)
 | 
				
			||||||
 | 
					        Returns:
 | 
				
			||||||
 | 
					            preprocessed_image: 预处理后的图像tensor
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        if image_array is None:
 | 
				
			||||||
 | 
					            raise ValueError("Input image is None")
 | 
				
			||||||
 | 
					        
 | 
				
			||||||
 | 
					        # 确保图像是numpy数组
 | 
				
			||||||
        if not isinstance(image_array, np.ndarray):
 | 
					        if not isinstance(image_array, np.ndarray):
 | 
				
			||||||
        raise ValueError("输入必须是numpy数组")
 | 
					            raise ValueError("Input must be numpy array")
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    # 调整图像尺寸
 | 
					        # 检查图像维度
 | 
				
			||||||
    height, width = image_array.shape[:2]
 | 
					        if len(image_array.shape) != 3:
 | 
				
			||||||
    if height != img_size[1] or width != img_size[0]:
 | 
					            raise ValueError(f"Expected 3D image array, got {len(image_array.shape)}D")
 | 
				
			||||||
        image_array = cv2.resize(image_array, img_size)
 | 
					 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    # 归一化到[0,1]
 | 
					        height, width, channels = image_array.shape
 | 
				
			||||||
    image_array = image_array.astype(np.float32) / 255.0
 | 
					        if channels != 3:
 | 
				
			||||||
 | 
					            raise ValueError(f"Expected 3 channels, got {channels}")
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    # 转换为CHW格式
 | 
					        # 调整图像尺寸到模型要求的尺寸
 | 
				
			||||||
    if len(image_array.shape) == 3:
 | 
					        if height != self.img_size[1] or width != self.img_size[0]:
 | 
				
			||||||
        image_array = np.transpose(image_array, (2, 0, 1))
 | 
					            image_array = cv2.resize(image_array, tuple(self.img_size))
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    # 添加batch维度
 | 
					        # 使用与训练时相同的预处理方式
 | 
				
			||||||
    image_array = np.expand_dims(image_array, axis=0)
 | 
					        image_array = image_array.astype('float32')
 | 
				
			||||||
 | 
					        image_array -= 127.5
 | 
				
			||||||
 | 
					        image_array *= 0.0078125
 | 
				
			||||||
 | 
					        image_array = np.transpose(image_array, (2, 0, 1))  # HWC -> CHW
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    return image_array
 | 
					        # 转换为tensor并添加batch维度
 | 
				
			||||||
 | 
					        image_tensor = torch.from_numpy(image_array).unsqueeze(0)
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
def greedy_decode(prebs):
 | 
					        return image_tensor
 | 
				
			||||||
    """贪婪解码"""
 | 
					    
 | 
				
			||||||
    preb_labels = list()
 | 
					    def decode_prediction(self, logits):
 | 
				
			||||||
    for i in range(prebs.shape[0]):
 | 
					        """
 | 
				
			||||||
        preb = prebs[i, :, :]
 | 
					        解码模型预测结果 - 使用正确的CTC贪婪解码
 | 
				
			||||||
        preb_label = list()
 | 
					        Args:
 | 
				
			||||||
        for j in range(preb.shape[1]):
 | 
					            logits: 模型输出的logits [batch_size, num_classes, sequence_length]
 | 
				
			||||||
 | 
					        Returns:
 | 
				
			||||||
 | 
					            predicted_text: 预测的车牌号码
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        # 转换为numpy进行处理
 | 
				
			||||||
 | 
					        prebs = logits.cpu().detach().numpy()
 | 
				
			||||||
 | 
					        preb = prebs[0, :, :]  # 取第一个batch [num_classes, sequence_length]
 | 
				
			||||||
 | 
					        
 | 
				
			||||||
 | 
					        # 贪婪解码:对每个时间步选择最大概率的字符
 | 
				
			||||||
 | 
					        preb_label = []
 | 
				
			||||||
 | 
					        for j in range(preb.shape[1]):  # 遍历每个时间步
 | 
				
			||||||
            preb_label.append(np.argmax(preb[:, j], axis=0))
 | 
					            preb_label.append(np.argmax(preb[:, j], axis=0))
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        no_repeat_blank_label = list()
 | 
					        # CTC解码:去除重复字符和空白字符
 | 
				
			||||||
 | 
					        no_repeat_blank_label = []
 | 
				
			||||||
        pre_c = preb_label[0]
 | 
					        pre_c = preb_label[0]
 | 
				
			||||||
        if pre_c != len(CHARS) - 1:
 | 
					        
 | 
				
			||||||
 | 
					        # 处理第一个字符
 | 
				
			||||||
 | 
					        if pre_c != len(CHARS) - 1:  # 不是空白字符
 | 
				
			||||||
            no_repeat_blank_label.append(pre_c)
 | 
					            no_repeat_blank_label.append(pre_c)
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        for c in preb_label:  # 去除重复标签和空白标签
 | 
					        # 处理后续字符
 | 
				
			||||||
            if (pre_c == c) or (c == len(CHARS) - 1):
 | 
					        for c in preb_label:
 | 
				
			||||||
 | 
					            if (pre_c == c) or (c == len(CHARS) - 1):  # 重复字符或空白字符
 | 
				
			||||||
                if c == len(CHARS) - 1:
 | 
					                if c == len(CHARS) - 1:
 | 
				
			||||||
                    pre_c = c
 | 
					                    pre_c = c
 | 
				
			||||||
                continue
 | 
					                continue
 | 
				
			||||||
            no_repeat_blank_label.append(c)
 | 
					            no_repeat_blank_label.append(c)
 | 
				
			||||||
            pre_c = c
 | 
					            pre_c = c
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        preb_labels.append(no_repeat_blank_label)
 | 
					        # 转换为字符
 | 
				
			||||||
 | 
					        decoded_chars = [CHARS[idx] for idx in no_repeat_blank_label]
 | 
				
			||||||
 | 
					        return ''.join(decoded_chars)
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    return preb_labels
 | 
					    def predict(self, image_array):
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        预测单张图像的车牌号码
 | 
				
			||||||
 | 
					        Args:
 | 
				
			||||||
 | 
					            image_array: numpy数组格式的图像
 | 
				
			||||||
 | 
					        Returns:
 | 
				
			||||||
 | 
					            prediction: 预测的车牌号码
 | 
				
			||||||
 | 
					            confidence: 预测置信度
 | 
				
			||||||
 | 
					        """
 | 
				
			||||||
 | 
					        try:
 | 
				
			||||||
 | 
					            # 预处理图像
 | 
				
			||||||
 | 
					            image = self.preprocess_image(image_array)
 | 
				
			||||||
 | 
					            if image is None:
 | 
				
			||||||
 | 
					                return None, 0.0
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
def LPRNinitialize_model(model_path=None):
 | 
					            image = image.to(self.device)
 | 
				
			||||||
    """初始化LPRNet模型"""
 | 
					            
 | 
				
			||||||
    global lprnet_model, device
 | 
					            # 模型推理
 | 
				
			||||||
 | 
					            with torch.no_grad():
 | 
				
			||||||
 | 
					                logits = self.model(image)
 | 
				
			||||||
 | 
					                # logits shape: [batch_size, class_num, sequence_length]
 | 
				
			||||||
 | 
					                
 | 
				
			||||||
 | 
					                # 计算置信度(使用softmax后的最大概率平均值)
 | 
				
			||||||
 | 
					                probs = torch.softmax(logits, dim=1)
 | 
				
			||||||
 | 
					                max_probs = torch.max(probs, dim=1)[0]
 | 
				
			||||||
 | 
					                confidence = torch.mean(max_probs).item()
 | 
				
			||||||
 | 
					                
 | 
				
			||||||
 | 
					                # 解码预测结果
 | 
				
			||||||
 | 
					                prediction = self.decode_prediction(logits)
 | 
				
			||||||
 | 
					            
 | 
				
			||||||
 | 
					            return prediction, confidence
 | 
				
			||||||
 | 
					            
 | 
				
			||||||
 | 
					        except Exception as e:
 | 
				
			||||||
 | 
					            print(f"预测图像失败: {e}")
 | 
				
			||||||
 | 
					            return None, 0.0
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					# 全局变量
 | 
				
			||||||
 | 
					lpr_model = None
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					def LPRNinitialize_model():
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
 | 
					    初始化LPRNet模型
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    返回:
 | 
				
			||||||
 | 
					        bool: 初始化是否成功
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
 | 
					    global lpr_model
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    try:
 | 
					    try:
 | 
				
			||||||
        # 设置设备
 | 
					        # 模型权重文件路径
 | 
				
			||||||
        device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 | 
					        model_path = os.path.join(os.path.dirname(__file__), 'LPRNet__iteration_74000.pth')
 | 
				
			||||||
        print(f"使用设备: {device}")
 | 
					 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        # 构建模型
 | 
					        # 创建推理对象
 | 
				
			||||||
        lprnet_model = build_lprnet(
 | 
					        lpr_model = LPRNetInference(model_path)
 | 
				
			||||||
            lpr_max_len=8, 
 | 
					 | 
				
			||||||
            phase=False, 
 | 
					 | 
				
			||||||
            class_num=len(CHARS), 
 | 
					 | 
				
			||||||
            dropout_rate=0.5
 | 
					 | 
				
			||||||
        )
 | 
					 | 
				
			||||||
        
 | 
					 | 
				
			||||||
        # 加载预训练权重
 | 
					 | 
				
			||||||
        if model_path is None:
 | 
					 | 
				
			||||||
            model_path = os.path.join(os.path.dirname(__file__), "Final_LPRNet_model.pth")
 | 
					 | 
				
			||||||
        
 | 
					 | 
				
			||||||
        if os.path.exists(model_path):
 | 
					 | 
				
			||||||
            checkpoint = torch.load(model_path, map_location=device)
 | 
					 | 
				
			||||||
            lprnet_model.load_state_dict(checkpoint)
 | 
					 | 
				
			||||||
            print(f"成功加载预训练模型: {model_path}")
 | 
					 | 
				
			||||||
        else:
 | 
					 | 
				
			||||||
            print(f"警告: 未找到预训练模型文件 {model_path},使用随机初始化权重")
 | 
					 | 
				
			||||||
        
 | 
					 | 
				
			||||||
        lprnet_model.to(device)
 | 
					 | 
				
			||||||
        lprnet_model.eval()
 | 
					 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        print("LPRNet模型初始化完成")
 | 
					        print("LPRNet模型初始化完成")
 | 
				
			||||||
        
 | 
					 | 
				
			||||||
        # 统计模型参数
 | 
					 | 
				
			||||||
        total_params = sum(p.numel() for p in lprnet_model.parameters())
 | 
					 | 
				
			||||||
        print(f"LPRNet模型参数数量: {total_params:,}")
 | 
					 | 
				
			||||||
        
 | 
					 | 
				
			||||||
        return True
 | 
					        return True
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    except Exception as e:
 | 
					    except Exception as e:
 | 
				
			||||||
@@ -209,76 +284,45 @@ def LPRNmodel_predict(image_array):
 | 
				
			|||||||
        image_array: numpy数组格式的车牌图像,已经过矫正处理
 | 
					        image_array: numpy数组格式的车牌图像,已经过矫正处理
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    返回:
 | 
					    返回:
 | 
				
			||||||
        list: 包含7个字符的列表,代表车牌号的每个字符
 | 
					        list: 包含最多8个字符的列表,代表车牌号的每个字符
 | 
				
			||||||
              例如: ['京', 'A', '1', '2', '3', '4', '5']
 | 
					              例如: ['京', 'A', '1', '2', '3', '4', '5'] (蓝牌7位)
 | 
				
			||||||
 | 
					                   ['京', 'A', 'D', '1', '2', '3', '4', '5'] (绿牌8位)
 | 
				
			||||||
    """
 | 
					    """
 | 
				
			||||||
    global lprnet_model, device
 | 
					    global lpr_model
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    if lprnet_model is None:
 | 
					    if lpr_model is None:
 | 
				
			||||||
        print("LPRNet模型未初始化,请先调用LPRNinitialize_model()")
 | 
					        print("LPRNet模型未初始化,请先调用LPRNinitialize_model()")
 | 
				
			||||||
        return ['待', '识', '别', '0', '0', '0', '0']
 | 
					        return ['待', '识', '别', '0', '0', '0', '0', '0']
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    try:
 | 
					    try:
 | 
				
			||||||
        # 预处理图像
 | 
					        # 预测车牌号
 | 
				
			||||||
        processed_image = preprocess_image(image_array)
 | 
					        predicted_text, confidence = lpr_model.predict(image_array)
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        # 转换为tensor
 | 
					        if predicted_text is None:
 | 
				
			||||||
        input_tensor = torch.from_numpy(processed_image).float()
 | 
					            print("LPRNet识别失败")
 | 
				
			||||||
        input_tensor = input_tensor.to(device)
 | 
					            return ['识', '别', '失', '败', '0', '0', '0', '0']
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        # 模型推理
 | 
					        print(f"LPRNet识别结果: {predicted_text}, 置信度: {confidence:.3f}")
 | 
				
			||||||
        with torch.no_grad():
 | 
					 | 
				
			||||||
            prebs = lprnet_model(input_tensor)
 | 
					 | 
				
			||||||
            prebs = prebs.cpu().detach().numpy()
 | 
					 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        # 贪婪解码
 | 
					        # 将字符串转换为字符列表
 | 
				
			||||||
        preb_labels = greedy_decode(prebs)
 | 
					        char_list = list(predicted_text)
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
        if len(preb_labels) > 0 and len(preb_labels[0]) > 0:
 | 
					        # 确保返回至少7个字符,最多8个字符
 | 
				
			||||||
            # 将索引转换为字符
 | 
					        if len(char_list) < 7:
 | 
				
			||||||
            predicted_chars = [CHARS[idx] for idx in preb_labels[0] if idx < len(CHARS)]
 | 
					            # 如果识别结果少于7个字符,用'0'补齐到7位
 | 
				
			||||||
 | 
					            char_list.extend(['0'] * (7 - len(char_list)))
 | 
				
			||||||
 | 
					        elif len(char_list) > 8:
 | 
				
			||||||
 | 
					            # 如果识别结果多于8个字符,截取前8个
 | 
				
			||||||
 | 
					            char_list = char_list[:8]
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
            print(f"LPRNet识别结果: {''.join(predicted_chars)}")
 | 
					        # 如果是7位,补齐到8位以保持接口一致性(第8位用空字符或占位符)
 | 
				
			||||||
 | 
					        if len(char_list) == 7:
 | 
				
			||||||
 | 
					            char_list.append('')  # 添加空字符作为第8位占位符
 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
            # 确保返回7个字符(车牌标准长度)
 | 
					        return char_list
 | 
				
			||||||
            if len(predicted_chars) < 7:
 | 
					 | 
				
			||||||
                # 如果识别结果少于7个字符,用'0'补齐
 | 
					 | 
				
			||||||
                predicted_chars.extend(['0'] * (7 - len(predicted_chars)))
 | 
					 | 
				
			||||||
            elif len(predicted_chars) > 7:
 | 
					 | 
				
			||||||
                # 如果识别结果多于7个字符,截取前7个
 | 
					 | 
				
			||||||
                predicted_chars = predicted_chars[:7]
 | 
					 | 
				
			||||||
            
 | 
					 | 
				
			||||||
            return predicted_chars
 | 
					 | 
				
			||||||
        else:
 | 
					 | 
				
			||||||
            print("LPRNet识别结果为空")
 | 
					 | 
				
			||||||
            return ['识', '别', '为', '空', '0', '0', '0']
 | 
					 | 
				
			||||||
        
 | 
					        
 | 
				
			||||||
    except Exception as e:
 | 
					    except Exception as e:
 | 
				
			||||||
        print(f"LPRNet识别失败: {e}")
 | 
					        print(f"LPRNet识别失败: {e}")
 | 
				
			||||||
        import traceback
 | 
					        import traceback
 | 
				
			||||||
        traceback.print_exc()
 | 
					        traceback.print_exc()
 | 
				
			||||||
        return ['识', '别', '失', '败', '0', '0', '0']
 | 
					        return ['识', '别', '失', '败', '0', '0', '0', '0']
 | 
				
			||||||
 | 
					 | 
				
			||||||
# 为了保持与其他模块的一致性,提供一个处理器类
 | 
					 | 
				
			||||||
class LPRProcessor:
 | 
					 | 
				
			||||||
    def __init__(self):
 | 
					 | 
				
			||||||
        self.initialized = False
 | 
					 | 
				
			||||||
    
 | 
					 | 
				
			||||||
    def initialize(self, model_path=None):
 | 
					 | 
				
			||||||
        """初始化模型"""
 | 
					 | 
				
			||||||
        self.initialized = LPRNinitialize_model(model_path)
 | 
					 | 
				
			||||||
        return self.initialized
 | 
					 | 
				
			||||||
    
 | 
					 | 
				
			||||||
    def predict(self, image_array):
 | 
					 | 
				
			||||||
        """预测接口"""
 | 
					 | 
				
			||||||
        if not self.initialized:
 | 
					 | 
				
			||||||
            print("模型未初始化")
 | 
					 | 
				
			||||||
            return ['未', '初', '始', '化', '0', '0', '0']
 | 
					 | 
				
			||||||
        return LPRNmodel_predict(image_array)
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
# 创建全局处理器实例
 | 
					 | 
				
			||||||
_processor = LPRProcessor()
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
def get_lpr_processor():
 | 
					 | 
				
			||||||
    """获取LPR处理器实例"""
 | 
					 | 
				
			||||||
    return _processor
 | 
					 | 
				
			||||||
@@ -22,6 +22,17 @@ def LPRNinitialize_model():
 | 
				
			|||||||
    return _processor
 | 
					    return _processor
 | 
				
			||||||
 | 
					
 | 
				
			||||||
def LPRNmodel_predict(image_array):
 | 
					def LPRNmodel_predict(image_array):
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
 | 
					    OCR车牌号识别接口函数
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    参数:
 | 
				
			||||||
 | 
					        image_array: numpy数组格式的车牌图像,已经过矫正处理
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    返回:
 | 
				
			||||||
 | 
					        list: 包含最多8个字符的列表,代表车牌号的每个字符
 | 
				
			||||||
 | 
					              例如: ['京', 'A', '1', '2', '3', '4', '5', ''] (蓝牌7位+占位符)
 | 
				
			||||||
 | 
					                   ['京', 'A', 'D', '1', '2', '3', '4', '5'] (绿牌8位)
 | 
				
			||||||
 | 
					    """
 | 
				
			||||||
    # 获取原始预测结果
 | 
					    # 获取原始预测结果
 | 
				
			||||||
    raw_result = _processor.predict(image_array)
 | 
					    raw_result = _processor.predict(image_array)
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
@@ -37,13 +48,19 @@ def LPRNmodel_predict(image_array):
 | 
				
			|||||||
    # 转换为字符列表
 | 
					    # 转换为字符列表
 | 
				
			||||||
    char_list = list(filtered_str)
 | 
					    char_list = list(filtered_str)
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    # 确保返回长度为7的列表
 | 
					    # 确保返回至少7个字符,最多8个字符
 | 
				
			||||||
    if len(char_list) >= 7:
 | 
					    if len(char_list) < 7:
 | 
				
			||||||
        # 如果长度大于等于7,取前7个字符
 | 
					        # 如果识别结果少于7个字符,用'0'补齐到7位
 | 
				
			||||||
        return char_list[:7]
 | 
					        char_list.extend(['0'] * (7 - len(char_list)))
 | 
				
			||||||
    else:
 | 
					    elif len(char_list) > 8:
 | 
				
			||||||
        # 如果长度小于7,用空字符串补齐到7位
 | 
					        # 如果识别结果多于8个字符,截取前8个
 | 
				
			||||||
        return char_list + [''] * (7 - len(char_list))
 | 
					        char_list = char_list[:8]
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    # 如果是7位,补齐到8位以保持接口一致性(第8位用空字符或占位符)
 | 
				
			||||||
 | 
					    if len(char_list) == 7:
 | 
				
			||||||
 | 
					        char_list.append('')  # 添加空字符作为第8位占位符
 | 
				
			||||||
 | 
					    
 | 
				
			||||||
 | 
					    return char_list
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 
 | 
				
			|||||||
							
								
								
									
										53
									
								
								README.md
									
									
									
									
									
								
							
							
						
						
									
										53
									
								
								README.md
									
									
									
									
									
								
							@@ -15,11 +15,10 @@ License_plate_recognition/
 | 
				
			|||||||
├── OCR_part/                 # OCR识别模块
 | 
					├── OCR_part/                 # OCR识别模块
 | 
				
			||||||
│   └── ocr_interface.py      # OCR接口(占位)
 | 
					│   └── ocr_interface.py      # OCR接口(占位)
 | 
				
			||||||
├── CRNN_part/                # CRNN识别模块
 | 
					├── CRNN_part/                # CRNN识别模块
 | 
				
			||||||
│   └── crnn_interface.py     # CRNN接口
 | 
					│   └── crnn_interface.py     # CRNN接口(占位)
 | 
				
			||||||
└── LPRNET_part/              # LPRNet识别模块
 | 
					└── LPRNET_part/              # LPRNet识别模块
 | 
				
			||||||
    ├── lpr_interface.py      # LPRNet接口
 | 
					    ├── lpr_interface.py      # LPRNet接口(已完成)
 | 
				
			||||||
    ├── Final_LPRNet_model.pth # 预训练模型文件
 | 
					    └── LPRNet__iteration_74000.pth  # LPRNet模型权重文件
 | 
				
			||||||
    └── will_delete/          # 参考资料(可删除)
 | 
					 | 
				
			||||||
```
 | 
					```
 | 
				
			||||||
 | 
					
 | 
				
			||||||
## 功能特性
 | 
					## 功能特性
 | 
				
			||||||
@@ -39,19 +38,22 @@ License_plate_recognition/
 | 
				
			|||||||
- 将倾斜的车牌矫正为标准矩形
 | 
					- 将倾斜的车牌矫正为标准矩形
 | 
				
			||||||
- 输出标准尺寸的车牌图像供识别使用
 | 
					- 输出标准尺寸的车牌图像供识别使用
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### 4. PyQt界面
 | 
					### 4. 多种识别方案
 | 
				
			||||||
 | 
					- 支持OCR、CRNN和LPRNet三种车牌识别方法
 | 
				
			||||||
 | 
					- LPRNet模型准确率高达98%
 | 
				
			||||||
 | 
					- 模块化接口设计,便于切换不同识别算法
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					### 5. PyQt界面
 | 
				
			||||||
- 左侧:实时摄像头画面显示
 | 
					- 左侧:实时摄像头画面显示
 | 
				
			||||||
- 右侧:检测结果展示区域
 | 
					- 右侧:检测结果展示区域
 | 
				
			||||||
  - 顶部显示识别到的车牌数量
 | 
					  - 顶部显示识别到的车牌数量
 | 
				
			||||||
  - 每行显示:车牌类型、矫正后图像、车牌号
 | 
					  - 每行显示:车牌类型、矫正后图像、车牌号
 | 
				
			||||||
- 美观的现代化界面设计
 | 
					- 美观的现代化界面设计
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### 5. 模块化设计
 | 
					### 6. 模块化设计
 | 
				
			||||||
- yolopart:负责车牌定位和矫正
 | 
					- yolopart:负责车牌定位和矫正
 | 
				
			||||||
- OCR_part:基于PaddleOCR的车牌号识别模块
 | 
					- OCR_part/CRNN_part/LPRNET_part:负责车牌号识别
 | 
				
			||||||
- CRNN_part:基于CRNN网络的车牌号识别模块
 | 
					- 各模块独立,便于维护和扩展
 | 
				
			||||||
- LPRNET_part:基于LPRNet网络的车牌号识别模块
 | 
					 | 
				
			||||||
- 各模块独立,便于维护和扩展,可通过修改main.py中的导入语句切换识别模块
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
## 安装和使用
 | 
					## 安装和使用
 | 
				
			||||||
 | 
					
 | 
				
			||||||
@@ -73,7 +75,21 @@ pip install -r requirements.txt
 | 
				
			|||||||
python main.py
 | 
					python main.py
 | 
				
			||||||
```
 | 
					```
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### 5. 使用说明
 | 
					### 5. 选择识别模块
 | 
				
			||||||
 | 
					在 `main.py` 中修改导入语句来选择不同的识别方案:
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					```python
 | 
				
			||||||
 | 
					# 使用LPRNet(推荐,准确率98%)
 | 
				
			||||||
 | 
					from LPRNET_part.lpr_interface import LPRNmodel_predict, LPRNinitialize_model
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					# 使用OCR
 | 
				
			||||||
 | 
					from OCR_part.ocr_interface import LPRNmodel_predict, LPRNinitialize_model
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					# 使用CRNN
 | 
				
			||||||
 | 
					from CRNN_part.crnn_interface import LPRNmodel_predict, LPRNinitialize_model
 | 
				
			||||||
 | 
					```
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					### 6. 使用说明
 | 
				
			||||||
1. 点击"启动摄像头"按钮开始检测
 | 
					1. 点击"启动摄像头"按钮开始检测
 | 
				
			||||||
2. 将车牌对准摄像头
 | 
					2. 将车牌对准摄像头
 | 
				
			||||||
3. 系统会自动检测车牌并显示:
 | 
					3. 系统会自动检测车牌并显示:
 | 
				
			||||||
@@ -95,8 +111,9 @@ YOLO Pose模型输出包含:
 | 
				
			|||||||
 | 
					
 | 
				
			||||||
## 接口说明
 | 
					## 接口说明
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### OCR/CRNN接口
 | 
					### 车牌识别接口
 | 
				
			||||||
车牌号识别部分使用统一接口:
 | 
					
 | 
				
			||||||
 | 
					项目为OCR、CRNN和LPRNet识别模块提供了标准接口:
 | 
				
			||||||
 | 
					
 | 
				
			||||||
```python
 | 
					```python
 | 
				
			||||||
# 接口函数名(导入所需模块,每个模块统一函数名)
 | 
					# 接口函数名(导入所需模块,每个模块统一函数名)
 | 
				
			||||||
@@ -108,7 +125,7 @@ LPRNinitialize_model()
 | 
				
			|||||||
# 预测主函数
 | 
					# 预测主函数
 | 
				
			||||||
from 对应模块 import LPRNmodel_predict
 | 
					from 对应模块 import LPRNmodel_predict
 | 
				
			||||||
result = LPRNmodel_predict(corrected_image)  # 返回7个字符的列表
 | 
					result = LPRNmodel_predict(corrected_image)  # 返回7个字符的列表
 | 
				
			||||||
 | 
					```
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### 输入参数
 | 
					### 输入参数
 | 
				
			||||||
- `corrected_image`:numpy数组格式的矫正后车牌图像
 | 
					- `corrected_image`:numpy数组格式的矫正后车牌图像
 | 
				
			||||||
@@ -117,6 +134,14 @@ result = LPRNmodel_predict(corrected_image)  # 返回7个字符的列表
 | 
				
			|||||||
- 长度为7的字符列表,包含车牌号的每个字符
 | 
					- 长度为7的字符列表,包含车牌号的每个字符
 | 
				
			||||||
- 例如:`['京', 'A', '1', '2', '3', '4', '5']`
 | 
					- 例如:`['京', 'A', '1', '2', '3', '4', '5']`
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					### LPRNet模块特性
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					- **高准确率**: 模型准确率高达98%
 | 
				
			||||||
 | 
					- **快速推理**: 基于深度学习的端到端识别
 | 
				
			||||||
 | 
					- **CTC解码**: 使用CTC(Connectionist Temporal Classification)解码算法
 | 
				
			||||||
 | 
					- **支持中文**: 完整支持中文省份简称和字母数字组合
 | 
				
			||||||
 | 
					- **模型权重**: 使用预训练的LPRNet__iteration_74000.pth权重文件
 | 
				
			||||||
 | 
					
 | 
				
			||||||
## 开发说明
 | 
					## 开发说明
 | 
				
			||||||
 | 
					
 | 
				
			||||||
### 添加新的识别算法
 | 
					### 添加新的识别算法
 | 
				
			||||||
 
 | 
				
			|||||||
							
								
								
									
										22
									
								
								main.py
									
									
									
									
									
								
							
							
						
						
									
										22
									
								
								main.py
									
									
									
									
									
								
							@@ -361,8 +361,8 @@ class MainWindow(QMainWindow):
 | 
				
			|||||||
            # 矫正车牌图像
 | 
					            # 矫正车牌图像
 | 
				
			||||||
            corrected_image = self.correct_license_plate(detection)
 | 
					            corrected_image = self.correct_license_plate(detection)
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
            # 获取车牌号(占位)
 | 
					            # 获取车牌号,传入车牌类型信息
 | 
				
			||||||
            plate_number = self.recognize_plate_number(corrected_image)
 | 
					            plate_number = self.recognize_plate_number(corrected_image, detection['class_name'])
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
            # 创建车牌显示组件
 | 
					            # 创建车牌显示组件
 | 
				
			||||||
            plate_widget = LicensePlateWidget(
 | 
					            plate_widget = LicensePlateWidget(
 | 
				
			||||||
@@ -389,7 +389,7 @@ class MainWindow(QMainWindow):
 | 
				
			|||||||
            detection['keypoints']
 | 
					            detection['keypoints']
 | 
				
			||||||
        )
 | 
					        )
 | 
				
			||||||
    
 | 
					    
 | 
				
			||||||
    def recognize_plate_number(self, corrected_image):
 | 
					    def recognize_plate_number(self, corrected_image, class_name):
 | 
				
			||||||
        """识别车牌号"""
 | 
					        """识别车牌号"""
 | 
				
			||||||
        if corrected_image is None:
 | 
					        if corrected_image is None:
 | 
				
			||||||
            return "识别失败"
 | 
					            return "识别失败"
 | 
				
			||||||
@@ -399,9 +399,21 @@ class MainWindow(QMainWindow):
 | 
				
			|||||||
            # 函数名改成一样的了,所以不要修改这里了,想用哪个模块直接导入
 | 
					            # 函数名改成一样的了,所以不要修改这里了,想用哪个模块直接导入
 | 
				
			||||||
            result = LPRNmodel_predict(corrected_image)
 | 
					            result = LPRNmodel_predict(corrected_image)
 | 
				
			||||||
            
 | 
					            
 | 
				
			||||||
            # 将字符列表转换为字符串
 | 
					            # 将字符列表转换为字符串,支持8位车牌号
 | 
				
			||||||
            if isinstance(result, list) and len(result) >= 7:
 | 
					            if isinstance(result, list) and len(result) >= 7:
 | 
				
			||||||
                return ''.join(result[:7])
 | 
					                # 根据车牌类型决定显示位数
 | 
				
			||||||
 | 
					                if class_name == '绿牌' and len(result) >= 8:
 | 
				
			||||||
 | 
					                    # 绿牌显示8位,过滤掉空字符占位符
 | 
				
			||||||
 | 
					                    plate_chars = [char for char in result[:8] if char != '']
 | 
				
			||||||
 | 
					                    # 如果过滤后确实有8位,显示8位;否则显示7位
 | 
				
			||||||
 | 
					                    if len(plate_chars) == 8:
 | 
				
			||||||
 | 
					                        return ''.join(plate_chars)
 | 
				
			||||||
 | 
					                    else:
 | 
				
			||||||
 | 
					                        return ''.join(plate_chars[:7])
 | 
				
			||||||
 | 
					                else:
 | 
				
			||||||
 | 
					                    # 蓝牌或其他类型显示前7位,过滤掉空字符
 | 
				
			||||||
 | 
					                    plate_chars = [char for char in result[:7] if char != '']
 | 
				
			||||||
 | 
					                    return ''.join(plate_chars)
 | 
				
			||||||
            else:
 | 
					            else:
 | 
				
			||||||
                return "识别失败"
 | 
					                return "识别失败"
 | 
				
			||||||
        except Exception as e:
 | 
					        except Exception as e:
 | 
				
			||||||
 
 | 
				
			|||||||
		Reference in New Issue
	
	Block a user