Skip to content
TheChatArchiveThe Chat Archive
gpt-3.5-turbo-0301· OpenAIAi2 WildChat-4.8M (AI2)

python不用noise库生成2d柏林噪声 需要可以传参octaves,persistence ,seed,frequence

  1. User

    python不用noise库生成2d柏林噪声 需要可以传参octaves,persistence ,seed,frequence

  2. Assistant

    以下是可以生成2D柏林噪声的Python代码,它不需要使用任何外部库,使用Perlin Noise算法来生成噪声。代码可以接受octaves(八度数)、persistence(持续度)、seed(种子)、frequency(频率)等参数:

    import random
    import math
    
    def perlin(x, y, octaves, persistence, freq, seed):
        # 生成噪声图像的像素值
        def noise(x, y):
            n = int(x) + int(y) * 57
            n = (n << 13) ^ n
            t = (n * (n * n * seed + 1993) + 1013) & 0x7fffffff
            return 1.0 - float(t) * 0.931322574615478515625e-9
        
        total = 0.0
        for i in range(octaves):
            f = 2 ** i
            p = persistence ** i
            total += interp_noise(x * f * freq, y * f * freq, noise) * p
        return total
    
    # 插值函数
    def interp_noise(x, y, noise):
        x0 = int(x)
        y0 = int(y)
        x1 = x0 + 1
        y1 = y0 + 1
    
        sx = x - x0
        sy = y - y0
    
        n0 = noise(x0, y0)
        n1 = noise(x1, y0)
        ix0 = lerp(n0, n1, sx)
    
        n0 = noise(x0, y1)
        n1 = noise(x1, y1)
        ix1 = lerp(n0, n1, sx)
    
        return lerp(ix0, ix1, sy)
    
    # 线性差值
    def lerp(a, b, x):
        return a * (1.0 - x) + b * x
    
    # 生成2D噪声
    def gen_noise(width, height, octaves, persistence, freq, seed):
        values = []
        for y in range(height):
            row = []
            for x in range(width):
                row.append(perlin(x, y, octaves, persistence, freq, seed))
            values.append(row)
        return values
    

    使用上述代码生成示例模拟地形图像的方式如下:

    import matplotlib.pyplot as plt
    import numpy as np
    
    width = 512
    height = 512
    octaves = 8
    persistence = 0.5
    freq = 0.02
    seed = 123
    
    # 生成2D噪声图像数据
    image_data = gen_noise(width, height, octaves, persistence, freq, seed)
    
    # 显示噪声图像
    plt.figure()
    plt.imshow(np.array(image_data))
    plt.show()
    

    该代码在Python3.6执行没有问题,在较老版本中,可能需要调整代码中的部分Python语法。