python实现图像最近邻插值

ratio_H = tar_H/src_H = tar_x/src_x = 3
ratio_W = tar_W/src_W = tar_y/src_y = 4

tar_x = src_x/ratio_H
tar_y = src_y/ratio_W

def nearest(image, target_size):
    """
    Nearest Neighbour interpolate for RGB  image
    
    :param image: rgb image
    :param target_size: tuple = (height, width)
    :return: None
    """
    if target_size[0] < image.shape[0] or target_size[1] < image.shape[1]:
        raise ValueError("target image must bigger than input image")
    # 1：按照尺寸创建目标图像
    target_image = np.zeros(shape=(*target_size, 3))
    # 2:计算height和width的缩放因子
    alpha_h = target_size[0]/image.shape[0]
    alpha_w = target_size[1]/image.shape[1]

    for tar_x in range(target_image.shape[0]-1):
        for tar_y in range(target_image.shape[1]-1):
            # 3:计算目标图像人任一像素点
            # target_image[tar_x,tar_y]需要从原始图像
            # 的哪个确定的像素点image[src_x, xrc_y]取值
            # 也就是计算坐标的映射关系
            src_x = round(tar_x/alpha_h)
            src_y = round(tar_y/alpha_w)

            # 4：对目标图像的任一像素点赋值
            target_image[tar_x, tar_y] = image[src_x, src_y]

    return target_image