1
Итак, у меня есть программа обработки изображений, и я работал над ней на рабочем столе Win7. Одной из его функций является изменение контрастности изображения, которое пользователь выбирает. Он отлично работает на рабочем столе Win7, но когда я переношу его на свой ноутбук на Win10, он делает переполнение красного значения. Это результат на ноутбуке Win10:Выполнение python для разных устройств
И это один на Win7 рабочем столе:
Это функция, которая меняет Контраст:
def change_contrast(level):
img = Image.open(a)
img.load()
factor = (259 * (level+255))/(255 * (259-level))
for x in range(img.size[0]):
for y in range(img.size[1]):
color = img.getpixel((x, y))
new_color = tuple(int(factor * (c-128) + 128) for c in color)
img.putpixel((x, y), new_color)
return img
Вот полный код в случае необходимости:
from tkinter import*
import tkinter as Tkinter
from tkinter import filedialog, DISABLED, messagebox as tkMessageBox
import os
import ntpath
from PIL import Image, ImageTk, ImageFilter
import PIL
from collections import Counter
from random import randint
import random
import PIL.ImageOps
def EchelleDeGris():
Ima2=Image.new("RGB",(z[0],z[1]))
px=Ima1.load()
px1=Ima2.load()
for x in range(z[0]):
for y in range(z[1]):
p=px[x,y]
if type(p)==int:
p=(p,p,p)
o=int((p[0]+p[1]+p[2])/3)
px1[x,y]=(o,o,o)
Ima2.save(""+dir_path+"\\Requirements\\ImageMod.png")
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\ImageMod.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
def SupprimerImage():
I2 = Tkinter.Label(main, image=imt)
I2.grid(row=0, column=4, columnspan =4)
def Luminosite():
Ima2=Image.new("RGB",(z[0],z[1]))
px=Ima1.load()
px1=Ima2.load()
for x in range(z[0]):
for y in range(z[1]):
p=px[x,y]
if type(p)==int:
p=(p,p,p)
px1[x,y]=(p[0]+S1.get(),p[1]+S1.get(),p[2]+S1.get())
Ima2.save(""+dir_path+"\\Requirements\\ImageMod.png")
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\ImageMod.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
def AnnulerModifications():
I2 = Tkinter.Label(main, image=im1)
I2.grid(row=0, column=4, columnspan =4)
def get_pixel(pixels, x, y):
try:
return pixels[x, y]
except IndexError:
return None
def get_neighbors(pixels, x, y):
neighbors = list()
neighbors.append(get_pixel(pixels, x, y - 1))
neighbors.append(get_pixel(pixels, x, y + 1))
neighbors.append(get_pixel(pixels, x - 1, y))
neighbors.append(get_pixel(pixels, x + 1, y))
neighbors.append(get_pixel(pixels, x - 1, y - 1))
neighbors.append(get_pixel(pixels, x - 1, y + 1))
neighbors.append(get_pixel(pixels, x + 1, y - 1))
neighbors.append(get_pixel(pixels, x + 1, y + 1))
return neighbors
def filter_art(pixels, size):
indexes = dict()
for x in range(size[0]):
for y in range(size[1]):
color = get_pixel(pixels, x, y)
neighbors = get_neighbors(pixels, x, y)
new_color = Counter(neighbors).most_common()[0][0]
if new_color is not None:
indexes[x, y] = new_color
for x, y in indexes:
pixels[x, y] = indexes[x, y]
def pop_art(path_orig, path_mod, coef):
s=[]
for i in range(9):
r=(randint(0,255), randint(0,255), randint(0,255))
g=(randint(0,255), randint(0,255), randint(0,255))
b=(randint(0,255), randint(0,255), randint(0,255))
image_orig = Image.open(path_orig)
size = image_orig.size
image_mod = Image.new("RGB",(size[0],size[1]))
pixels_orig = image_orig.load()
pixels_mod = image_mod.load()
for x in range(size[0]):
for y in range(size[1]):
p = pixels_orig[x, y]
if isinstance(p, int):
rgb = (p,p,p)
elif isinstance(p, tuple) and len(p) in (3, 4):
rgb = p[:3]
else:
raise TypeError('Unknown pallete')
average_color = sum(rgb)/3
if average_color <= 85:
pixels_mod[x, y] = r
elif 85 < average_color <= 170:
pixels_mod[x, y] = g
elif average_color > 170:
pixels_mod[x, y] = b
for _ in range(coef):
filter_art(pixels_mod, size)
image_mod.save(''+dir_path+'\\PopArt\\Modified Images\\result'+str(i)+'.png')
Img=[None]*9
for i in range(9):
Img[i]=Image.open(""+dir_path+"\\PopArt\\Modified Images\\result"+str(i)+".png")
basewidth = int(Img[i].size[1]/3)
wpercent = (basewidth/float(Img[i].size[0]))
hsize = int((float(Img[i].size[1]) * float(wpercent)))
Img[i] = Img[i].resize((basewidth , hsize), PIL.Image.ANTIALIAS)
Img[i].save(''+dir_path+'\\PopArt\\Resized Images\\resized_image'+str(i)+'.png')
Img1=[None]*9
pixels1=[None]*9
Imaz=Image.new("RGB",(basewidth*3,hsize*3))
pixels=Imaz.load()
for i in range(9):
Img1[i]=Image.open(''+dir_path+'\\PopArt\\Resized Images\\resized_image'+str(i)+'.png')
pixels1[i]=Img1[i].load()
for x in range(0,basewidth):
for y in range(0,hsize):
pixels[x,y]=pixels1[0][x,y]
for y in range(hsize,hsize*2):
pixels[x,y]=pixels1[1][x,y-hsize]
for y in range(hsize*2,hsize*3):
pixels[x,y]=pixels1[2][x,y-hsize*2]
for x in range(basewidth,basewidth*2):
for y in range(0,hsize):
pixels[x,y]=pixels1[3][x-basewidth,y]
for y in range(hsize,hsize*2):
pixels[x,y]=pixels1[4][x-basewidth,y-hsize]
for y in range(hsize*2,hsize*3):
pixels[x,y]=pixels1[5][x-basewidth,y-hsize*2]
for x in range(basewidth*2,basewidth*3):
for y in range(0,hsize):
pixels[x,y]=pixels1[6][x-basewidth*2,y]
for y in range(hsize,hsize*2):
pixels[x,y]=pixels1[7][x-basewidth*2,y-hsize]
for y in range(hsize*2,hsize*3):
pixels[x,y]=pixels1[8][x-basewidth*2,y-hsize*2]
Imaz = Imaz.resize((size[0] , size[1]), PIL.Image.ANTIALIAS)
Imaz.save(""+dir_path+"\\PopArt\\Result Image\\result.png")
def usepop():
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\traitement.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
I2.update_idletasks()
pop_art(a, None, coef=4)
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\PopArt\\Result Image\\result.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
def change_contrast(level):
img = Image.open(a)
img.load()
factor = (259 * (level+255))/(255 * (259-level))
for x in range(img.size[0]):
for y in range(img.size[1]):
color = img.getpixel((x, y))
new_color = tuple(int(factor * (c-128) + 128) for c in color)
img.putpixel((x, y), new_color)
return img
def use_contrast():
result = change_contrast(S2.get())
result.save(""+dir_path+"\\Requirements\\ImageMod.png")
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\ImageMod.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
def recherche_contours():
Ima2=Image.new("RGB",(z[0],z[1]))
px=Ima1.load()
px1=Ima2.load()
for x in range(z[0]):
for y in range(z[1]):
p=px[x,y]
if type(p)==int:
p=(p,p,p)
o=int((p[0]+p[1]+p[2])/3)
px1[x,y]=(o,o,o)
Ima2 = Ima2.filter(ImageFilter.FIND_EDGES)
image = Ima2
if image.mode == 'RGBA':
r,g,b,a = image.split()
rgb_image = Image.merge('RGB', (r,g,b))
inverted_image = PIL.ImageOps.invert(rgb_image)
r2,g2,b2 = inverted_image.split()
final_transparent_image = Image.merge('RGBA', (r2,g2,b2,a))
final_transparent_image.save(""+dir_path+"\\Requirements\\ImageMod.png")
else:
inverted_image = PIL.ImageOps.invert(image)
inverted_image.save(""+dir_path+"\\Requirements\\ImageMod.png")
im2 = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\ImageMod.png")
main.image = im2
I2 = Tkinter.Label(main, image=im2)
I2.grid(row=0, column=4, columnspan =4)
main=Tk()
main.withdraw()
a = filedialog.askopenfilename()
main.deiconify()
dir_path = os.path.dirname(os.path.realpath("Test2.py"))
main.configure(background="#a1dbcd")
main.title("Photoshop Version.Megzari")
Ima1=Image.open(a)
z=Ima1.size
nux=Image.new("RGB",(z[0],z[1]))
nuxy=nux.load()
for x in range(z[0]):
for y in range(z[1]):
nuxy[x,y]=(255,255,255)
nux.save(""+dir_path+"\\Requirements\\Blank.png")
if z>(400,400):
main.withdraw()
tkMessageBox.showinfo("Resolution Error", "The image is too big, please select a smaller one.")
sys.exit()
elif z<(400,400):
im1 = ImageTk.PhotoImage(file=a)
I1 = Tkinter.Label(main, image=im1)
I1.grid(row=0, column=1, columnspan =3)
imt = ImageTk.PhotoImage(file=""+dir_path+"\\Requirements\\Blank.png")
T1 = Tkinter.Label(main, image=imt)
T1.grid(row=0, column=4, columnspan =4)
B1 = Tkinter.Button(main, text ="Echelle de gris", command = EchelleDeGris, fg="#a1dbcd", bg="#383a39", state=NORMAL)
B1.grid(padx=20, pady=20, row=1, column=0)
B3 = Tkinter.Button(main, text ="Appliquer Luminosité", command = Luminosite, fg="#a1dbcd", bg="#383a39")
B3.grid(padx=20, pady=20, row=1, column=1)
S1 = Scale(main, from_=0, to=254, orient=HORIZONTAL, fg="#a1dbcd", bg="#383a39", length = 200)
S1.grid(row=2, column=1)
B2 = Tkinter.Button(main, text ="Supprimer Image", command = SupprimerImage, fg="#a1dbcd", bg="#383a39")
B2.grid(padx=20, pady=20, row=1, column=7)
B3 = Tkinter.Button(main, text ="Annuler Modifications", command = AnnulerModifications, fg="#a1dbcd", bg="#383a39")
B3.grid(padx=20, pady=20, row=1, column=6)
B4 = Tkinter.Button(main, text ="Pop Art", command = usepop, fg="#a1dbcd", bg="#383a39")
B4.grid(padx=20, pady=20, row=1, column=3)
S2 = Scale(main, from_=-258, to=258, orient=HORIZONTAL, fg="#a1dbcd", bg="#383a39", length = 200)
S2.grid(row=2, column=4)
B4 = Tkinter.Button(main, text ="Appliquer Contraste", command = use_contrast, fg="#a1dbcd", bg="#383a39")
B4.grid(padx=20, pady=20, row=1, column=4)
B5 = Tkinter.Button(main, text ="Trouver Contours", command = recherche_contours, fg="#a1dbcd", bg="#383a39")
B5.grid(padx=20, pady=20, row=1, column=5)
s=S1.get()
s2=S2.get()
main.mainloop()
Я не знаю, почему это дает мне другой результат в зависимости от платформы, с которой я ее запускаю.
EDIT: Код, который использует PIL.ImageEnhance.Contrast:
from PIL import Image, ImageEnhance
image = Image.open('E:\Projet Complet\Images\obama.png')
contrast = ImageEnhance.Contrast(image)
contrast_applied=contrast.enhance(1)
image.show()
На Win7 у меня есть Python 3.2.5 и 1.1.7 PIL,
А на Win10 У меня есть Python 3.4.5 и PIL 1,1 .7
какой-либо причине вы не используете [ 'PIL.ImageEnhance.Contrast'] (http://pillow.readthedocs.io/en/3.1.x/reference/ImageEnhance .html # PIL.ImageEnhance.Contrast)? –
В случае, если вы не заметили, мой код получает значение шкалы ('s2 = S2.get()') между '-258' и' 258' и создает контрастное изображение в зависимости от этого значения. С 'PIL.ImageEnhance.Contrast', я не могу настроить уровень контраста, он просто дает мне один улучшенный образ. –
Он принимает число с плавающей запятой от '0.0' до' 1.0'. Это не обеспечивает выравнивание? –