深入Python计算机视觉世界：图像处理与分析的终极指南

import cv2import numpy as np# 读取图像image = cv2.imread("image.jpg")# 图像增强：调整对比度和亮度image_enhanced = cv2.addWeighted(image, 1.5, np.zeros(image.shape, image.dtype), 0, 20)# 图像滤波：高斯滤波image_filtered = cv2.GaussianBlur(image_enhanced, (5, 5), 0)# 图像分割：阈值分割image_segmented = cv2.threshold(image_filtered, 127, 255, cv2.THRESH_BINARY)[1]# 显示图像cv2.imshow("Enhanced Image", image_enhanced)cv2.imshow("Filtered Image", image_filtered)cv2.imshow("Segmented Image", image_segmented)cv2.waiTKEy(0)cv2.destroyAllwindows()

import cv2import numpy as npfrom sklearn.svm import SVC# 读取图像image = cv2.imread("image.jpg")# 目标检测：使用Haar级联分类器检测人脸face_cascade = cv2.CascadeClassifier("haarcascade_frontalface_default.xml")faces = face_cascade.detectMultiScale(image, 1.3, 5)# 对象识别：使用SVM分类器识别猫和狗model = SVC()model.fit(X_train, y_train)predictions = model.predict(X_test)# 图像分类：使用CNN分类器对图像进行分类model = keras.models.load_model("model.h5")prediction = model.predict(image)# 显示结果for (x, y, w, h) in faces:cv2.rectangle(image, (x, y), (x+w, y+h), (0, 255, 0), 2)for i, pred in enumerate(predictions):print(f"Class {i+1}: {pred:.2f}")print(f"Predicted Class: {np.argmax(prediction)}")cv2.imshow("Detected Faces", image)cv2.waitKey(0)cv2.destroyAllWindows()