快乐学AI系列——计算机视觉（5）图像分类和识别

import numpy as np
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score

# 加载手写数字数据集
digits = load_digits()
X, y = digits.data, digits.target

# 将数字标签转换为字母标签
y = np.array(['zero', 'one', 'two', 'three', 'four', 'five', 'six', 'seven', 'eight', 'nine'])[y]

# 划分数据集为训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 初始化KNN分类器并进行训练
knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(X_train, y_train)

# 对测试集进行预测并计算准确率
y_pred = knn.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print("Accuracy:", accuracy)

import cv2
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.svm import LinearSVC
from sklearn.metrics import accuracy_score
from skimage.feature import hog

# 加载数据集
dataset = []
for i in range(10):
    for j in range(50):
        img = cv2.imread(f'digits/{i}_{j}.png', 0)
        img = cv2.resize(img, (20, 20))
        dataset.append((img, i))

# 提取HOG特征和标签
X = []
y = []
for img, label in dataset:
    hog_feature = hog(img, orientations=9, pixels_per_cell=(4, 4), cells_per_block=(2, 2))
    X.append(hog_feature)
    y.append(label)

# 划分数据集为训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 初始化SVM分类器并进行训练
svm = LinearSVC()
svm.fit(X_train, y_train)

# 对测试集进行预测并计算准确率
y_pred = svm.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)
print("Accuracy:", accuracy)

import cv2

# 读取两张图片
img1 = cv2.imread('image1.jpg')
img2 = cv2.imread('image2.jpg')

# 初始化SIFT特征提取器
'''
SIFT算法已经在OpenCV 4中被移除，如果使用OpenCV 4及以上版本，
需要将代码中的 cv2.xfeatures2d.SIFT_create() 替换为 cv2.SIFT_create()。
'''
sift = cv2.xfeatures2d.SIFT_create()

# 提取图片1和图片2中的关键点和特征描述符
kp1, des1 = sift.detectAndCompute(img1, None)
kp2, des2 = sift.detectAndCompute(img2, None)

# 初始化FLANN匹配器
FLANN_INDEX_KDTREE = 0
index_params = dict(algorithm=FLANN_INDEX_KDTREE, trees=5)
search_params = dict(checks=50)
flann = cv2.FlannBasedMatcher(index_params, search_params)

# 对特征描述符进行匹配
matches = flann.knnMatch(des1, des2, k=2)

# 过滤掉不好的匹配
good_matches = []
for m, n in matches:
    if m.distance < 0.7 * n.distance:
        good_matches.append(m)

# 显示匹配结果
result = cv2.drawMatches(img1, kp1, img2, kp2, good_matches, None, flags=2)
cv2.imshow('matches', result)
cv2.waitKey(0)
cv2.destroyAllWindows()