车辆数据车辆数据车辆数据

import tkinter as tk from tkinter import filedialog from PIL import Image, ImageTk import numpy as np from keras.preprocessing import image as keras_image from keras.applications.vgg16 import preprocess_input, decode_predictions from tensorflow.keras.applications import ResNet50 from keras.preprocessing.image import ImageDataGenerator, load_img, img_to_array, array_to_img from keras.models import Sequential from keras.layers import Conv2D, MaxPooling2D from keras.layers import Activation, Dropout, Flatten, Dense from keras import backend as K import numpy as np import matplotlib.pyplot as plt import glob, os, random # 创建 Tkinter 窗口 root = tk.Tk() root.title("人脸表情识别") # 设置窗口大小和背景颜色 root.geometry("500x400") root.configure(bg='white') # 创建登录函数 def login(): username = username_entry.get() password = password_entry.get() # 这里可以添加验证用户名和密码的逻辑，这里简化为输出 print(f"用户名: {username}, 密码: {password}") # 登录成功后跳转到上传图像页面 show_upload_page() # 显示上传图像页面 def show_upload_page(): # 清空登录部分的控件 username_label.place_forget() username_entry.place_forget() password_label.place_forget() password_entry.place_forget() login_button.place_forget() # 创建上传图像和识别按钮 upload_button.place(x=200, y=300) classify_button.place(x=300, y=300) # 创建用于显示图像的 Label panel.place(x=50, y=50, width=400, height=200) # 上传图像的函数 def upload_image(): file_path = filedialog.askopenfilename() img = Image.open(file_path) img.thumbnail((400, 200)) # 缩小图像以适应界面 img = ImageTk.PhotoImage(img) panel.config(image=img) panel.image = img # 保存图像路径以备识别按钮使用 global uploaded_image_path uploaded_image_path = file_path # 识别图像的函数 def classify_image(): if uploaded_image_path: img = Image.open(uploaded_image_path) img = img.resize((100, 100),3) # 调整图像大小以符合模型输入要求 img = keras_image.img_to_array(img) img = np.expand_dims(img, axis=0) img = preprocess_input(img) #-------------------------------------------------------------------- # 统一定义图像像素的宽度和高度 img_width, img_height = 100, 100 # 定义训练集、验证集的图形路径（文件夹路径即可） train_data_dir = 'augmented/' validation_data_dir = 'augmented/' # 模型训练的参数设置 nb_train_samples = 275 nb_validation_samples = 200 epochs = 10 # 迭代次数 batch_size = 64 # 每个批量观测数 # 图像输入维度设置 if K.image_data_format() == 'channels_first': input_shape = (3, img_width, img_height) else: input_shape = (img_width, img_height, 3) from keras.models import load_model model = Sequential() #-------------------------------------------------------------------- # 输入层：第一层 # 添加第一个卷积层/最大池化层(必选) model.add(Conv2D(filters=32, # 32 个过滤器 kernel_size=(3, 3), # 卷积核大小 3 x 3 input_shape=input_shape, # 图像输入维度 activation='relu')) # 'relu' 激活函数 model.add(MaxPooling2D(pool_size=(2, 2))) # 池化核大小 2 x 2 # ---------------------------------------------------- # 隐藏层：介于第一层和最后一层之间 # 添加第二个卷积层/最大池化层(可选) model.add(Conv2D(filters=32, kernel_size=(3, 3), activation='relu')) model.add(MaxPooling2D(pool_size=(2, 2))) # 添加第三个卷积层/最大池化层(可选) model.add(Conv2D(filters=64, kernel_size=(3, 3), activation='relu')) model.add(MaxPooling2D(pool_size=(2, 2))) # 由于卷积层是 2D 空间，训练时需要将数据展平为 1D 空间 model.add(Flatten()) # 添加展平层(必选) model.add(Dense(units=64, activation='relu')) # 添加全连接层(必选) 64 个神经元 model.add(Dropout(0.5)) # 添加丢弃层，防止过拟合 # --------------------------------------------------- # 输出层：最后一层，神经元控制输出的维度，并指定分类激活函数 model.add(Dense(units=2, activation='sigmoid')) # 指定分类激活函数 model = load_model('model.h5') # 使用模型进行预测 predictions = model.predict(img) # decoded_preds = decode_predictions(preds, top=1)[0] # 获取最高置信度的预测结果 # from keras.preprocessing import image # model = load_model('model.h5') # img_path = '测试1.jpg' # img_width, img_height = 100, 100 # img = image.load_img(img_path, target_size=(100, 100)) # 例如，针对VGG模型，resize图像为224x224 # x = image.img_to_array(img) # x = np.expand_dims(x, axis=0) # x = preprocess_input(x) # 预处理图像数据，例如针对VGG模型 # # 对测试图像进行预测 # predictions = model.predict(x) print(predictions.shape) print('predictions:', predictions) predictions = predictions[0] # 指定要列出子文件和子文件夹的目录路径 folder_path = '车辆数据集' # 替换为你要获取子文件和子文件夹的目录路径 # 获取指定路径下的所有文件和文件夹列表 contents = os.listdir(folder_path) print('contents', contents) max_indices = [i for i, value in enumerate(predictions) if value == max(predictions)][0] print(contents[max_indices]) # 获取分类结果 # label = decoded_preds[0][1] label=contents[max_indices] result_text.delete(1.0, tk.END) # 清空文本框 result_text.insert(tk.END, f"类别: {label}") else: result_text.delete(1.0, tk.END) result_text.insert(tk.END, "请先上传图像") # 创建登录部分 username_label = tk.Label(root, text="用户名:", bg='white') username_label.place(x=50, y=20) username_entry = tk.Entry(root) username_entry.place(x=150, y=20) password_label = tk.Label(root, text="密码:", bg='white') password_label.place(x=50, y=50) password_entry = tk.Entry(root, show="*") password_entry.place(x=150, y=50) login_button = tk.Button(root, text="登录", command=login, bg='lightblue', padx=10, pady=5) login_button.place(x=300, y=35) # 创建上传图像和识别按钮 upload_button = tk.Button(root, text="上传图像", command=upload_image, bg='lightgreen', padx=10, pady=5) classify_button = tk.Button(root, text="识别", command=classify_image, bg='lightyellow', padx=10, pady=5) # 创建用于显示图像的 Label panel = tk.Label(root, bg='white') # 创建用于显示分类结果的文本框 result_text = tk.Text(root, height=2, width=30, bg='lightyellow') result_text.place(x=150, y=350) # 加载预训练模型（这里以 ResNet50 为例） model = ResNet50(weights='imagenet') root.mainloop()