from tkinter import Tk, Label
root = Tk()
root.title('ROOT')
# set text at runtime
label_name = Label(root)
label_name.__setitem__('text', 'abc')
label_name.pack()
# "text" can not be omitted
label_phonetic = Label(root, text='phonetic:')
label_phonetic.pack()
label_meaning = Label(root, text='meaning:')
label_meaning.pack()
label_example = Label(root, text='example:')
label_example.pack()
root.mainloop()
class window(Frame):
def __init__(self, master=None):
self.frame=Frame.__init__(self,master)
self.master=master
self.initwindow(self.frame)
#inisiasi frame
def initwindow(self,master):
self.master.title("UAS PENGOLAHAN CITRA DIGITAL HILMAN")
self.master.geometry("1350x670+0+0") #ukuran size 1350 x 570 posisi awal di 0,0
self.master.minsize(1350,670) #ukuran terkecil (tidak bisa diperkecil lagi)
self.initmenu(master)
self.left_window(master)
self.right_window(master)
self.array=[]
#pembuatan menu pada frame
def initmenu(self,master):
self.menubar = Menu(master)
self.filemenu = Menu(self.menubar, tearoff=0) #tearoff = menghilangkan garis putus-putus
self.filemenu.add_command(label="Browse Gambar",command=self.openimage)
self.filemenu.add_command(label="Simpan Hasil Proses",command=self.saveimage,state="disable")
self.menubar.add_cascade(label="File", menu=self.filemenu) #menambahkan file menu ke menubar
warna = Menu(self.menubar, tearoff=0)
warna.add_command(label="Grayscale",command=self.grayscale)
warna.add_command(label="Black&White",command=self.blackwhite)
warna.add_separator() #tambah garis pemisah
warna.add_command(label="Brightening",command=lambda : self.brigness_darkening("brightening"))
warna.add_command(label="Darkening",command=lambda : self.brigness_darkening("darkening")) #lambda biar henteu langsung di run commandna
self.menubar.add_cascade(label="Warna",menu=warna,state="disable") # default menu disabled
morfology = Menu(self.menubar, tearoff=0)
morfology.add_command(label="Erosi",command=lambda : self.morphology("erosi"))
morfology.add_command(label="Dilasi",command=lambda : self.morphology("dilasi"))
morfology.add_command(label="Opening",command=lambda : self.morphology("opening"))
morfology.add_command(label="Closing",command=lambda : self.morphology("closing"))
self.menubar.add_cascade(label="Morfology",menu=morfology,state="disable")
filtering = Menu(self.menubar,tearoff=0)
filtering.add_command(label="Mean Filtering",command=lambda : self.filtering("mean"))
filtering.add_command(label="Median Filtering",command=lambda : self.filtering("median"))
filtering.add_command(label="Gaussian Filtering",command=lambda : self.filtering("gaussian"))
self.menubar.add_cascade(label="Filtering",menu=filtering,state="disable")
edge_detection=Menu(self.menubar,tearoff=0)
edge_detection.add_command(label="Roberts",command=lambda : self.edge_detection("robert"))
edge_detection.add_command(label="Prewitt",command=lambda : self.edge_detection("prewitt"))
edge_detection.add_command(label="Sobel",command=lambda : self.edge_detection("sobel"))
edge_detection.add_command(label="Canny",command=lambda : self.edge_detection("canny"))
edge_detection.add_command(label="Laplacian",command=lambda : self.edge_detection("laplacian"))
self.menubar.add_cascade(label="Deteksi Garis",menu=edge_detection,state="disable")
noise=Menu(self.menubar,tearoff=0)
noise.add_command(label="Salt & pepper",command=self.salt_pepper)
self.menubar.add_cascade(label="Tambah Noise",menu=noise,state="disable")
self.psnr=Menu(self.menubar,tearoff=0)
self.psnr.add_command(label="Hitung PSNR",command=self.hitung_psnr)
self.psnr.add_command(label="Lihat Tabel PSNR",state="disable",command=self.show_psnr)
self.menubar.add_cascade(label="PSNR",menu=self.psnr,state="disable")
self.master.config(menu=self.menubar) # mengatur menu utama dengan menubar
Label(master, text="Proses Pengolahan Citra",pady=20,font=("Times New Roman",19)).grid(column=1) #.grid ditampilkan pada colom 1 row 0
#inisiasi citra awal (kiri)
def left_window(self,master):
Label(master, text="Gambar Asli",pady=10,font=("Times New Roman",16,"bold"),height=1).grid(column=0,row=1,sticky=W+E)
self.left_gambar = Label(master, text="Gambar Kosong",width=78,height=37,relief=GROOVE) #set width (caracter length)
self.left_gambar.grid(row=3)
#inisiasi label citra hasil pengolahan (kanan)
def right_window(self,master):
Label(master, text="Gambar Hasil Pemrosesan",pady=10,font=("Times New Roman",16,"bold"),height=1).grid(column=2,row=1,sticky=W+E)
self.right_gambar = Label(master, text="Gambar Kosong",width=78,height=37,relief=GROOVE)
self.right_gambar.grid(row=3,column=2)
#menampilkan citra hasil pengolahan
def set_right_gambar(self,hasil):
if self.filemenu.entrycget(1,"state")=="disabled": #enable menu "Simpan Gambar Hasil" ketika citra di bagian kanan sudah muncul
self.filemenu.entryconfig("Simpan Hasil Proses",state="normal")
self.psnr.entryconfig("Lihat Tabel PSNR",state="normal")
hasil=Image.fromarray(hasil) #array to image
self.proses=hasil
hasil = hasil.resize((550,550),Image.ANTIALIAS) #resize gambar ukuran 550x550
hasil=ImageTk.PhotoImage(hasil)
self.right_gambar.__setitem__("image",hasil) #set value image
self.right_gambar.image=hasil
if self.right_gambar.__getitem__("text")=="Gambar Kosong" : #ubah ukuran label dari character width jadi pixel
self.right_gambar.__setitem__("width",550) #set width (pixel)
self.right_gambar.__setitem__("height",550)
self.right_gambar.__setitem__("text","") # kosongkan text
#manipulasi citra (RGB2GRAYSCALE)
def grayscale(self):
hasil=numpy.array(self.proses) #image ke array
hasil=cv2.cvtColor(hasil,cv2.COLOR_BGR2GRAY) #format image dalam BGR bukan RGB
self.right_image=hasil
self.set_right_gambar(hasil)
#manipulasi citra (RGB2BINER)
def blackwhite(self):
hasil=numpy.array(self.proses)
hasil=cv2.cvtColor(hasil,cv2.COLOR_BGR2GRAY)
ret,hasil=cv2.threshold(hasil,127,255,cv2.THRESH_BINARY)
self.right_image=hasil
self.set_right_gambar(hasil)
#load citra
def openimage(self):
self.src = filedialog.askopenfilename() #openfile
if len(self.src)>0:
self.original =Image.open(self.src)
self.proses=self.original.resize((550,550),Image.ANTIALIAS)
hasil = self.proses
hasil = ImageTk.PhotoImage(hasil)
if self.left_gambar.__getitem__("text")=="Gambar Kosong": #atur ukuran label menjadi ukuran gambar (pixel)
self.left_gambar.__setitem__("width",550)
self.left_gambar.__setitem__("height",550)
self.left_gambar.__setitem__("text","")
self.left_gambar.__setitem__("image",hasil) #tampilkan gambar ke label kiri
self.left_gambar.image = hasil #menjaga referensi (menghindari garbage collection)
if self.menubar.entrycget(2,"state")=="disabled": #mengaktifkan menu yang disable sebelumnya
self.menubar.entryconfig("Deteksi Garis",state="normal")
self.menubar.entryconfig("Filtering",state="normal")
self.menubar.entryconfig("Morfology",state="normal")
self.menubar.entryconfig("Warna",state="normal")
self.menubar.entryconfig("Tambah Noise",state="normal")
self.menubar.entryconfig("PSNR",state="normal")
#edge detection canny
def edge_detection(self,jenis):
hasil=numpy.array(self.proses)
hasil=cv2.cvtColor(hasil,cv2.COLOR_BGR2GRAY)
hasil = cv2.GaussianBlur(hasil, (5, 5), 0) #blurring (gaussian filter)
if jenis == "canny" :
hasil = cv2.Canny(hasil, self.input_value("Nilai Threshold", "Batas Threshold Bawah"), self.input_value("Nilai Threshold", "Batas Threshold Atas"))
elif jenis == "sobel":
img_sobelx = cv2.Sobel(hasil,cv2.CV_8U,1,0,ksize=3)
img_sobely = cv2.Sobel(hasil,cv2.CV_8U,0,1,ksize=3)
hasil = img_sobelx + img_sobely #menggabungkan hasil sobel vertical + horizontal
elif jenis =="robert":
kernelx = numpy.array([[-1,0],[0,1]])
kernely = numpy.array([[0,-1],[1,0]])
img_robertx = cv2.filter2D(hasil, -1, kernelx)
img_roberty = cv2.filter2D(hasil, -1, kernely)
hasil = img_robertx+img_roberty
elif jenis =="prewitt":
kernelx = numpy.array([[1,1,1],[0,0,0],[-1,-1,-1]])
kernely = numpy.array([[-1,0,1],[-1,0,1],[-1,0,1]])
img_prewittx = cv2.filter2D(hasil, -1, kernelx)
img_prewitty = cv2.filter2D(hasil, -1, kernely)
hasil = img_prewittx+img_prewitty
elif jenis =="laplacian":
hasil =cv2.Laplacian(hasil,cv2.CV_64F,3)
self.right_image=hasil # self.right_image untuk penyimpanan file hasil, tidak menggunakan hasil karena kalo berwarna format yang harus disimpan BGR
self.set_right_gambar(hasil) #deteksi garis dalam GRAYSCALE tidak usah di ke BGR kan
#simpan citra hasil pengolahan (kanan)
def saveimage(self):
path=filedialog.asksaveasfilename(defaultextension=".png") #simpan extensi png
if path != "":
cv2.imwrite(path,self.right_image)
#manipulasi citra morpology
def morphology(self,bentuk):
hasil=self.proses.convert('RGB')
hasil=numpy.array(hasil)
kernel = self.input_kernel("Pemilihan Ukuran Kernel","Ukuran Kernel ?")
if bentuk == "erosi":
hasil = cv2.erode(hasil,kernel,iterations = 1)
elif bentuk == "dilasi":
hasil = cv2.dilate(hasil,kernel,iterations = 1)
elif bentuk == "opening":
hasil = cv2.morphologyEx(hasil, cv2.MORPH_OPEN, kernel)
elif bentuk == "closing":
hasil = cv2.morphologyEx(hasil, cv2.MORPH_CLOSE, kernel)
self.right_image=cv2.cvtColor(hasil,cv2.COLOR_RGB2BGR)
self.set_right_gambar(hasil)
#manipulasi citra filtering
def filtering(self,bentuk):
hasil=self.proses.convert('RGB')
hasil=numpy.array(hasil)
koefisien = self.input_value("Pemilihan Kernel","Ukuran Kernel")
if bentuk == "mean" :
hasil = cv2.blur(hasil,(koefisien,koefisien))
elif bentuk =="gaussian":
hasil = cv2.GaussianBlur(hasil,(koefisien,koefisien),0)
elif bentuk =="median":
hasil = cv2.medianBlur(hasil,koefisien)
self.right_image=cv2.cvtColor(hasil,cv2.COLOR_RGB2BGR)
self.set_right_gambar(hasil)
def input_kernel(self,judul,pesan):
answer=self.input_value(judul,pesan)
if answer != 0:
answer = numpy.ones((answer,answer),numpy.uint8) # matriks nxn semua elemen 1
else:
answer = numpy.ones((1,1),numpy.uint8)
return answer
def input_value(self,nama_window,pesan):
answer = simpledialog.askinteger(nama_window,pesan,minvalue=0, maxvalue=255,initialvalue=0)
if answer is None:
answer = 0
return answer
#manipulasi citra brigness darkening
def brigness_darkening(self,param):
hasil=self.proses.convert('RGB') #jikapun citra dalam bentuk grayscale, tetap dibaca sebagai BGR
hasil=numpy.array(hasil)
koefisien = self.input_value("Pemilihan Konstanta","Nilai Konstanta")
if koefisien != 0 :
for r in range(len(hasil)):
for s in range(len(hasil[r])):
if param == "brightening":
hasil[r][s][0] = numpy.clip(hasil[r][s][0] + koefisien, 0, 255) #numpy.clip prinsip conservative filtering
hasil[r][s][1] = numpy.clip(hasil[r][s][1] + koefisien, 0, 255)
hasil[r][s][2] = numpy.clip(hasil[r][s][2] + koefisien, 0, 255)
else :
hasil[r][s][0] = numpy.clip(hasil[r][s][0] - koefisien, 0, 255)
hasil[r][s][1] = numpy.clip(hasil[r][s][1] - koefisien, 0, 255)
hasil[r][s][2] = numpy.clip(hasil[r][s][2] - koefisien, 0, 255)
self.right_image=cv2.cvtColor(hasil,cv2.COLOR_RGB2BGR)
self.set_right_gambar(hasil)
def salt_pepper(self):
hasil=self.proses.convert('RGB') #jikapun citra dalam bentuk grayscale, tetap dibaca sebagai BGR
hasil=numpy.array(hasil)
intensitas = simpledialog.askfloat("Nilai Intesitas Noise","Intensitas Noise",minvalue=0, maxvalue=1,initialvalue=0)
if intensitas is None:
intensitas = 0
intensitas*=0.5
thres = 1 - intensitas
for i in range(hasil.shape[0]):
for j in range(hasil.shape[1]):
rdn = random.random()
if rdn < intensitas:
hasil[i][j][0]=0
hasil[i][j][1]=0
hasil[i][j][2]=0
elif rdn > thres:
hasil[i][j][0]=255
hasil[i][j][1]=255
hasil[i][j][2]=255
self.right_image=cv2.cvtColor(hasil,cv2.COLOR_RGB2BGR)
self.set_right_gambar(hasil)
def show_psnr(self):
top = Toplevel(width=300,height=300)
top.title("Nilai MSE dan PSNR")
top.minsize(300,300)
tree = Treeview(top, columns=('MSE', 'PSNR'))
tree.heading('#0', text='File')
tree.heading('#1', text='MSE')
tree.heading('#2', text='PSNR (db)')
tree.column('#0',anchor='center',width=200,stretch="true")
tree.column('#1',anchor='center',width=100,stretch="true")
tree.column('#2',anchor='center',width=200,stretch="true")
for i in self.array :
tree.insert('', 'end', text=i[0], values=(i[1], i[2]))
tree.grid(padx=10,pady=10)
close = Button(top,text="Close",pady=10,command=top.destroy,relief="groove",width=10)
close.grid(row=1,sticky="E",padx=10,pady=10)
top.mainloop()
def hitung_psnr(self):
proses = self.proses.resize((550,550),Image.ANTIALIAS)
original = self.original.resize((550,550),Image.ANTIALIAS)
original=numpy.array(original)
proses=numpy.array(proses)
original=cv2.cvtColor(original,cv2.COLOR_BGR2GRAY)
proses=cv2.cvtColor(proses,cv2.COLOR_BGR2GRAY)
mse = numpy.mean(numpy.subtract(original,proses)**2)
if mse == 0:
nilai_psnr="tak hingga"
else :
nilai_psnr=20 * math.log10(255.0 / math.sqrt(mse))
self.array.append([self.src.split("/")[-1],mse,nilai_psnr])
self.show_psnr()
