图像相似度对比

import cv2
from PIL import Image
from numpy import average, dot, linalg
from skimage.metrics import peak_signal_noise_ratio as psnr
from skimage.metrics import structural_similarity as ssim


def calc_similar(image1, image2):
    def hist_similar(lh, rh):
        assert len(lh) == len(rh)
        return sum(1 - (0 if l == r else float(abs(l - r)) / max(l, r)) for l, r in zip(lh, rh)) / len(lh)

    def imgread(image, size=(256, 256)):
        img = Image.open(image)
        return img.resize(size).convert('RGB')
    li, ri = imgread(image1), imgread(image2)
    return hist_similar(li.histogram(), ri.histogram())


def calc_cosin(image1, image2):
    def get_thum(image, size=(64, 64), greyscale=False):
        image = Image.open(image).convert('RGB')
        image = image.resize(size, Image.ANTIALIAS)
        if greyscale:
            image = image.convert('L')
        return image
    image1 = get_thum(image1)
    image2 = get_thum(image2)
    images = [image1, image2]
    vectors = []
    norms = []
    for image in images:
        vector = []
        for pixel_tuple in image.getdata():
            vector.append(average(pixel_tuple))
        vectors.append(vector)
        norms.append(linalg.norm(vector, 2))
    a, b = vectors
    a_norm, b_norm = norms
    res = dot(a / a_norm, b / b_norm)
    return res

def imgread(image):
    image = cv2.imread(image)
    image = cv2.resize(image, (256,256))
    image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    return image


# img1, img2 = './nsfw/2.jpg', './nsfw/3.jpg'
hist = calc_similar(img1, img2)
cosin = calc_cosin(img1, img2)
psnr = psnr(imgread(img1), imgread(img2), data_range=255)
ssim = ssim(imgread(img1), imgread(img2), data_range=255)
print(f'hist:{hist}  cosin:{cosin}  psnr:{psnr} ssim:{ssim}')