图像超分辨（1）

• 参考：

1. 图像插值算法：最近邻插值、双线性插值https://blog.csdn.net/qq_38463737/article/details/119898254?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522170943285416777224452716%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fblog.%2522%257D&request_id=170943285416777224452716&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~blog~first_rank_ecpm_v1~rank_v31_ecpm-1-119898254-null-null.nonecase&utm_term=%E6%9C%80%E8%BF%91%E9%82%BB%E6%8F%92%E5%80%BC%E8%B6%85%E5%88%86%E8%BE%A8&spm=1018.2226.3001.4450

基于插值的图像超分辨算法https://blog.csdn.net/qq_41620742/article/details/124971996?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522170943143716800186576760%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fblog.%2522%257D&request_id=170943143716800186576760&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~blog~first_rank_ecpm_v1~times_rank-7-124971996-null-null.nonecase&utm_term=%E5%8F%8C%E7%BA%BF%E6%80%A7%E6%8F%92%E5%80%BC%E8%B6%85%E5%88%86%E8%BE%A8&spm=1018.2226.3001.4450

图像超分辨——基于插值的方法（个人总结）https://blog.csdn.net/SimonCoder/article/details/107125534?ops_request_misc=&request_id=&biz_id=102&utm_term=%E6%9C%80%E8%BF%91%E9%82%BB%E6%8F%92%E5%80%BC%E8%B6%85%E5%88%86%E8%BE%A8&utm_medium=distribute.pc_search_result.none-task-blog-2~blog~sobaiduweb~default-1-107125534.nonecase&spm=1018.2226.3001.4450

• 传统的超分辨重建方法
• 最近邻插值

import numpy as np
from PIL import Image
import matplotlib.pyplot as plt


def nearest_interpolation(img, dstH, dstW):
    sctH, sctW, _ = img.shape
    retimg = np.zeros((dstH, dstW, 3), dtype=np.uint8)
    for i in range(dstH):
        for j in range(dstW):
            scrx = round(i * (sctH / dstH))
            scry = round(j * (sctW / dstW))

            # Ensure indices are within bounds
            scrx = min(scrx, sctH - 1)
            scry = min(scry, sctW - 1)

            retimg[i, j] = img[int(scrx), int(scry)]
    return retimg


# 读取输入图像
im_path = r"G:\UW\raw_images\3.png"
image = np.array(Image.open(im_path))

# 进行最近邻插值
image1 = nearest_interpolation(image, image.shape[0] * 2, image.shape[1] * 2)

# 将NumPy数组转换为PIL图像
image1 = Image.fromarray(image1.astype('uint8')).convert('RGB')

# 保存输出图像
output_path = 'out.png'
image1.save(output_path)

# 显示输入图像
plt.subplot(1, 2, 1)
plt.title('Input Image')
plt.imshow(image)
plt.axis('off')

# 显示输出图像
plt.subplot(1, 2, 2)
plt.title('Output Image')
plt.imshow(image1)
plt.axis('off')

plt.show()

• 双线性插值

from PIL import Image
import matplotlib.pyplot as plt
import numpy as np
import math
import matplotlib.pyplot as plt


def BiLinear_interpolation(img,dstH,dstW):
    scrH,scrW,_=img.shape
    img=np.pad(img,((0,1),(0,1),(0,0)),'constant')
    retimg=np.zeros((dstH,dstW,3),dtype=np.uint8)
    for i in range(dstH):
        for j in range(dstW):
            scrx=(i+1)*(scrH/dstH)-1
            scry=(j+1)*(scrW/dstW)-1
            x=math.floor(scrx)
            y=math.floor(scry)
            u=scrx-x
            v=scry-y
            retimg[i,j]=(1-u)*(1-v)*img[x,y]+u*(1-v)*img[x+1,y]+(1-u)*v*img[x,y+1]+u*v*img[x+1,y+1]
    return retimg
im_path = r"G:\UW\raw_images\3.png"
image = np.array(Image.open(im_path))
image2 = BiLinear_interpolation(image,image.shape[0]*2,image.shape[1]*2)
image2 = Image.fromarray(image2.astype('uint8')).convert('RGB')
image2.save('out3.png')

plt.subplot(1, 2, 1)
plt.title('Input Image')
plt.imshow(image)
plt.axis('off')

# 显示输出图像
plt.subplot(1, 2, 2)
plt.title('Output Image')
plt.imshow(image2)
plt.axis('off')

plt.show()

• 双三次插值

from PIL import Image
import numpy as np
import math
import matplotlib.pyplot as plt


# 产生16个像素点不同的权重
def BiBubic(x):
    x=abs(x)
    if x<=1:
        return 1-2*(x**2)+(x**3)
    elif x<2:
        return 4-8*x+5*(x**2)-(x**3)
    else:
        return 0

# 双三次插值算法
# dstH为目标图像的高，dstW为目标图像的宽
def BiCubic_interpolation(img,dstH,dstW):
    scrH,scrW,_=img.shape
    #img=np.pad(img,((1,3),(1,3),(0,0)),'constant')
    retimg=np.zeros((dstH,dstW,3),dtype=np.uint8)
    for i in range(dstH):
        for j in range(dstW):
            scrx=i*(scrH/dstH)
            scry=j*(scrW/dstW)
            x=math.floor(scrx)
            y=math.floor(scry)
            u=scrx-x
            v=scry-y
            tmp=0
            for ii in range(-1,2):
                for jj in range(-1,2):
                    if x+ii<0 or y+jj<0 or x+ii>=scrH or y+jj>=scrW:
                        continue
                    tmp+=img[x+ii,y+jj]*BiBubic(ii-u)*BiBubic(jj-v)
            retimg[i,j]=np.clip(tmp,0,255)
    return retimg

im_path=r"G:\UW\raw_images\3.png"
image=np.array(Image.open(im_path))
image2=BiCubic_interpolation(image,image.shape[0]*2,image.shape[1]*2)
image2=Image.fromarray(image2.astype('uint8')).convert('RGB')
image2.save('out4.png')

# 显示输入图像
plt.subplot(1, 2, 1)
plt.title('Input Image')
plt.imshow(image)
plt.axis('off')

# 显示输出图像
plt.subplot(1, 2, 2)
plt.title('Output Image')
plt.imshow(image2)
plt.axis('off')

plt.show()