要求:输入一个n*n的矩阵,矩阵包括从-1到1的浮点数,将其转化为可视化图像
调库
from PIL import Image import numpy as np import math
载入图像,并使用np.array()转化为矩阵
Image.new()第一个参数是模式,第二个是尺寸,第三个参数如果不填默认是黑色,填3个255就是白色.
img = Image.new('RGB', (256, 256), (255, 255, 255)) matrix=np.array(img) matrix 输出
array([[[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], ..., [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]]], dtype=uint8)
可以利用如下方法修改matrix的值
matrix[0][0]=[0,0,0] matrix
array([[[ 0, 0, 0], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], ..., [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]], [[255, 255, 255], [255, 255, 255], [255, 255, 255], ..., [255, 255, 255], [255, 255, 255], [255, 255, 255]]], dtype=uint8)
制造数据,100*100的矩阵,包含-1到1的浮点数
lis=[[]for i in range(100)] for i in range(100): for j in range(100): lis[i].append(float(math.sin(i+j))) lis[1][1]
0.9092974268256817
将[-1,1]映射到[0,255],负数为绿色系,正数为紫色系.
for i in range(100): for j in range(100): degree=lis[i][j] rgb=int(abs(degree)*255) #print(rgb) if(degree>=0): matrix[i][j]=[rgb,0,rgb] else: matrix[i][j]=[0,rgb,0] img=Image.fromarray(matrix) img
