用python实现基于概率分布的超声图像与自然图像性质差异分析这一课题，使用opencv, 两种图像的文件类型都是jpg,超声图像的文件路径是‘D:\zzz\us5’，自然图像的文件路径是‘D:\zzz\na4’，两种图像都有84张，图像的名称都是1到84的顺序数，两种图像的形状大小相同，需要对比两种图像的特征必须有颜色，纹理，形态学特征，差异性分析方法也需要多种，包括完整详细代码

为了实现这个课题，我们需要使用以下模块：

- OpenCV：用于图像处理和分析
- NumPy：用于数值计算和数组处理
- Matplotlib：用于可视化

首先，我们需要导入这些模块：

import cv2
import numpy as np
import matplotlib.pyplot as plt

接下来，我们需要定义函数来加载图像并提取它们的特征。我们将使用颜色直方图、灰度共生矩阵和形态学特征。

def load_images(path):
    images = []
    for i in range(1, 85):
        filename = f"{path}/{i}.jpg"
        img = cv2.imread(filename)
        images.append(img)
    return images

def get_color_histogram(img):
    hist = cv2.calcHist([img], [0, 1, 2], None, [8, 8, 8], [0, 256, 0, 256, 0, 256])
    hist = cv2.normalize(hist, hist).flatten()
    return hist

def get_gray_cooccurrence(img):
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    cooccurrence = np.zeros((256, 256))
    for i in range(gray.shape[0]):
        for j in range(gray.shape[1] - 1):
            cooccurrence[gray[i, j], gray[i, j+1]] += 1
    cooccurrence = cv2.normalize(cooccurrence, cooccurrence).flatten()
    return cooccurrence

def get_morphological_features(img):
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)
    kernel = np.ones((3, 3), np.uint8)
    opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel, iterations=2)
    closing = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel, iterations=2)
    gradient = cv2.morphologyEx(thresh, cv2.MORPH_GRADIENT, kernel)
    features = [np.mean(opening), np.mean(closing), np.mean(gradient), np.std(opening), np.std(closing), np.std(gradient)]
    return features

现在，我们可以使用这些函数来加载图像并提取它们的特征：

us_images = load_images('D:/zzz/us5')
na_images = load_images('D:/zzz/na4')

us_color_histograms = [get_color_histogram(img) for img in us_images]
na_color_histograms = [get_color_histogram(img) for img in na_images]

us_gray_cooc = [get_gray_cooccurrence(img) for img in us_images]
na_gray_cooc = [get_gray_cooccurrence(img) for img in na_images]

us_morphological = [get_morphological_features(img) for img in us_images]
na_morphological = [get_morphological_features(img) for img in na_images]

现在，我们可以使用这些特征来进行两种图像之间的差异性分析。我们将使用Kolmogorov-Smirnov测试来比较颜色直方图和灰度共生矩阵特征的分布，使用t-test来比较形态学特征的均值和标准差。

def compare_features(f1, f2, test):
    scores = []
    for i in range(len(f1)):
        score = test(f1[i], f2[i])[1]
        scores.append(score)
    return scores

def plot_distribution_scores(scores):
    plt.hist(scores, bins=20)
    plt.xlabel('Score')
    plt.ylabel('Frequency')
    plt.show()

color_scores = compare_features(us_color_histograms, na_color_histograms, cv2.compareHist)
gray_scores = compare_features(us_gray_cooc, na_gray_cooc, cv2.compareHist)
morphological_scores = compare_features(us_morphological, na_morphological, ttest_ind)

plot_distribution_scores(color_scores)
plot_distribution_scores(gray_scores)
plot_distribution_scores(morphological_scores)

最后，我们可以将这些特征的分数组合在一起，并使用主成分分析（PCA）来可视化它们之间的差异。

def normalize_scores(scores):
    scores = np.array(scores)
    return (scores - np.min(scores)) / (np.max(scores) - np.min(scores))

color_scores_norm = normalize_scores(color_scores)
gray_scores_norm = normalize_scores(gray_scores)
morphological_scores_norm = normalize_scores(morphological_scores)

feature_vectors = np.column_stack((color_scores_norm, gray_scores_norm, morphological_scores_norm))
mean = np.mean(feature_vectors, axis=0)
centered = feature_vectors - mean
eigenvalues, eigenvectors, _ = np.linalg.svd(centered)
projected = np.dot(centered, eigenvectors[:, :2])

for i in range(len(projected)):
    if i < 84:
        plt.scatter(projected[i, 0], projected[i, 1], c='r', alpha=0.5)
    else:
        plt.scatter(projected[i, 0], projected[i, 1], c='b', alpha=0.5)

plt.xlabel('PC1')
plt.ylabel('PC2')
plt.show()

这将生成一个可视化图，显示两种图像之间的差异：


完整代码如下：