用pycharm，opencv4.7.0将以上实验完成，必须有界面设计

要在PyCharm中使用OpenCV 4.7.0完成上述人脸检测与识别实验，并且包括界面设计，你可以按照以下步骤进行： ### 实验环境与工具 - **操作系统**: Windows 10 - **开发环境**: PyCharm - **编程语言**: Python - **库**: OpenCV 4.7.0, tkinter (用于界面设计) ### 实验步骤 #### 1. 安装必要的库 首先，确保你已经安装了OpenCV和tkinter。可以在PyCharm的终端中运行以下命令： ```bash pip install opencv-python ``` #### 2. 数据收集 ##### 2.1 下载标准人脸数据集 下载并解压ORL人脸数据库： - URL: https://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html - 解压后的文件结构应为每个人一个文件夹，每个文件夹下有10张PGM格式的图片。 ##### 2.2 建立自己的人脸数据集 编写代码调用摄像头捕获图像，并保存人脸样本。以下是示例代码： ```python import cv2 import os def collect_faces(): face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml') cap = cv2.VideoCapture(0) if not cap.isOpened(): raise IOError("Cannot open webcam") if not os.path.exists("my_faces"): os.makedirs("my_faces") count = 0 while True: ret, frame = cap.read() gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30)) for (x, y, w, h) in faces: cv2.rectangle(frame, (x, y), (x+w, y+h), (255, 0, 0), 2) face_image = gray[y:y+h, x:x+w] face_resized = cv2.resize(face_image, (92, 112)) cv2.imwrite(f"my_faces/{count}.jpg", face_resized) count += 1 cv2.imshow('Video', frame) if cv2.waitKey(1) & 0xFF == ord('q'): break cap.release() cv2.destroyAllWindows() collect_faces() ``` #### 3. 训练模型 生成CSV文件并训练模型。以下是示例代码： ```python import os import cv2 import numpy as np from sklearn.model_selection import train_test_split from sklearn.metrics import accuracy_score from sklearn.decomposition import PCA from sklearn.discriminant_analysis import LinearDiscriminantAnalysis from sklearn.neighbors import KNeighborsClassifier def generate_csv(base_path, output_file): with open(output_file, 'w') as f: for root, dirs, files in os.walk(base_path): for file in files: if file.endswith('.pgm') or file.endswith('.jpg'): path = os.path.join(root, file) label = os.path.basename(os.path.dirname(path)) f.write(f"{path};{label}\n") generate_csv("path_to_dataset", "dataset.csv") def load_data(csv_file): images = [] labels = [] with open(csv_file, 'r') as f: for line in f: parts = line.strip().split(';') image_path = parts[0] label = int(parts[1]) image = cv2.imread(image_path, 0) images.append(image.flatten()) labels.append(label) return np.array(images), np.array(labels) images, labels = load_data("dataset.csv") X_train, X_test, y_train, y_test = train_test_split(images, labels, test_size=0.2, random_state=42) # Train models pca = PCA(n_components=100) lda = LinearDiscriminantAnalysis(n_components=100) knn = KNeighborsClassifier(n_neighbors=3) X_train_pca = pca.fit_transform(X_train) X_test_pca = pca.transform(X_test) knn.fit(X_train_pca, y_train) y_pred_pca = knn.predict(X_test_pca) print(f"PCA Accuracy: {accuracy_score(y_test, y_pred_pca)}") X_train_lda = lda.fit_transform(X_train, y_train) X_test_lda = lda.transform(X_test) knn.fit(X_train_lda, y_train) y_pred_lda = knn.predict(X_test_lda) print(f"LDA Accuracy: {accuracy_score(y_test, y_pred_lda)}") # Save models joblib.dump(pca, 'pca_model.pkl') joblib.dump(lda, 'lda_model.pkl') joblib.dump(knn, 'knn_model.pkl') ``` #### 4. 测试模型 创建一个简单的GUI界面来测试模型。以下是示例代码： ```python import tkinter as tk from tkinter import filedialog import joblib import cv2 import numpy as np class FaceRecognitionApp: def __init__(self, master): self.master = master self.master.title("Face Recognition App") self.image_label = tk.Label(master) self.image_label.pack() self.result_label = tk.Label(master, text="") self.result_label.pack() self.upload_button = tk.Button(master, text="Upload Image", command=self.upload_image) self.upload_button.pack() self.capture_button = tk.Button(master, text="Capture Image", command=self.capture_image) self.capture_button.pack() self.recognize_button = tk.Button(master, text="Recognize", command=self.recognize_face) self.recognize_button.pack() self.image_path = None self.face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml') self.pca = joblib.load('pca_model.pkl') self.knn = joblib.load('knn_model.pkl') def upload_image(self): self.image_path = filedialog.askopenfilename() self.display_image() def capture_image(self): cap = cv2.VideoCapture(0) ret, frame = cap.read() cv2.imwrite("captured_image.jpg", frame) cap.release() self.image_path = "captured_image.jpg" self.display_image() def display_image(self): if self.image_path: img = cv2.imread(self.image_path) img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) img = cv2.resize(img, (300, 300)) img_tk = ImageTk.PhotoImage(image=Image.fromarray(img)) self.image_label.configure(image=img_tk) self.image_label.image = img_tk def recognize_face(self): if self.image_path: img = cv2.imread(self.image_path, 0) faces = self.face_cascade.detectMultiScale(img, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30)) if len(faces) > 0: (x, y, w, h) = faces[0] face_image = img[y:y+h, x:x+w] face_resized = cv2.resize(face_image, (92, 112)).flatten() face_pca = self.pca.transform([face_resized]) prediction = self.knn.predict(face_pca)[0] self.result_label.config(text=f"Predicted Label: {prediction}") else: self.result_label.config(text="No face detected") root = tk.Tk() app = FaceRecognitionApp(root) root.mainloop() ``` ### 总结 1. **数据收集**：下载标准人脸数据集，调用摄像头捕获并保存人脸样本。 2. **训练模型**：生成CSV文件，加载数据，训练PCA和KNN模型，并保存模型。 3. **测试模型**：创建一个简单的GUI界面，允许用户上传或捕获图像，并进行人脸识别。 希望这些步骤能帮助你在PyCharm中完成人脸检测与识别实验。如果有任何问题，请随时提问。