adding code

proyecto.py

@@ -0,0 +1,162 @@
+import sys
+import tkinter
+import cv2
+
+import numpy as np
+import scipy
+import matplotlib.pyplot as plt
+from scipy.fftpack import fft
+from scipy.io import wavfile # get the api
+import scipy.io.wavfile as waves
+
+import PIL  # MODULO PARA PROCESAR IMAGENES
+from PIL import Image
+import os  # MODULO PARA HACER COSAS EN EL DIRECTORIO
+import fnmatch  # MODULO PARA COMPARAR EXTENSIONES EN EL DIRECTORIO
+import tarfile # MODULO PARA COMPRIMIR
+
+from scipy import misc
+import pywt
+
+#1. TRANSMISOR
+fs, data = wavfile.read(sys.argv[1]) # load the data
+a = data # this is a two channel soundtrack, I get the first track
+result_fft = fft(a)
+result_len = int(len(result_fft)/2)  # you only need half of the fft list (real signal symmetry)
+#solo necesitaria entonces ir hasta result_len -1 para obviar los conjugados
+
+
+#xr=parte real del result_fft
+xr = np.real(result_fft[:result_len])
+#xi=parte imaginaria del result_fft
+xi = np.imag(result_fft[:result_len])
+#Calculando conversion a formato de imagen
+# aux1 = xr.astype(np.uint8)
+# aux2 = xi.astype(np.uint8)
+max_xr = xr[::].max()
+max_xi = xi[::].max()
+min_xr = xr[::].min()
+min_xi = xi[::].min()
+print("max_xr ",max_xr)
+print("max_xi ",max_xi)
+print("min_xr ",min_xr)
+print("min_xi ",min_xi)
+
+if min_xr != max_xr:
+    xrn = ((xr - min_xr) / (max_xr - min_xr))*255        #auxiliar para normalizar parte real de fft_result
+else:
+    xrn = np.zeros(shape = xr.shape)
+if min_xi != max_xi:
+    xri = ((xi - min_xi) / (max_xi - min_xi))*255        #auxiliar para normalizar parte imaginaria de fft_result
+else:
+    xri = np.zeros(shape = xi.shape)
+
+#agregando a cada imagen el minimo y maximo para la hora de descomprimir
+fila = np.zeros((1,xrn.shape[1]))
+
+fila[0] = min_xr
+xrn = np.concatenate((xrn, fila), 0)
+
+fila[0] = min_xi
+xri = np.concatenate((xri,fila), 0)
+
+fila[0] = max_xr
+xrn = np.concatenate((xrn, fila), 0)
+
+fila[0] = max_xi
+xri = np.concatenate((xri,fila), 0)
+
+xrn = xrn.astype(np.uint8)
+xri =xri.astype(np.uint8)
+print(xrn)
+print(xri)
+
+
+#PARA VER LAS IMAGENES
+# plt.rcParams['image.cmap'] = 'gray'
+# plt.imshow(xrn,vmin=0,vmax=1)
+# plt.figure()
+# plt.imshow(xri,vmin=0,vmax=1)
+
+imgr = Image.fromarray(xrn, mode="L")
+imgr.save("real.jpg")
+imgi = Image.fromarray(xri, mode="L")
+imgi.save("imag.jpg")
+
+# plt.show()
+
+
+
+
+
+
+
+ #2.RECEPTOR
+
+ #Algoritmo de descompresion de imagenes en formato .jpg
+real2 = cv2.imread("real.jpg", cv2.IMREAD_GRAYSCALE)
+imag2 = cv2.imread("imag.jpg", cv2.IMREAD_GRAYSCALE)
+
+imag = imag2.astype(np.double)
+real =real2.astype(np.double)
+print("real: ", real.shape)
+
+min_xr = real[real.shape[0] - 2][0]
+min_xi = imag[imag.shape[0] - 2][0]
+max_xr = real[real.shape[0] - 1][0]
+max_xi = imag[imag.shape[0] - 1][0]
+print("max_xr ",max_xr)
+print("max_xi ",max_xi)
+print("min_xr ",min_xr)
+print("min_xi ",min_xi)
+
+#hay que desnormalizar los valores:
+res_imag = np.zeros((imag.shape[0]-2, imag.shape[1]))
+res_real = np.zeros((real.shape[0]-2, real.shape[1]))
+
+for i in range(imag.shape[0] - 2):
+    if min_xi != max_xi:
+        res_imag[i] = (imag[i] * (max_xi - min_xi) / 255) + min_xi
+    else:
+        res_imag[i] = np.full(imag.shape[1], min_xi)
+
+for i in range(real.shape[0] - 2):
+    for j in range(real.shape[1]):
+        if min_xr != max_xr:
+            res_real[i][j] = (real[i][j] * (max_xr - min_xr) / 255) + min_xr
+        else:
+            res_real[i][j] = min_xr
+
+matriz = np.empty(shape = res_real.shape, dtype = complex)
+for i in range(res_real.shape[0]):
+    for j in range(res_real.shape[1]):
+        matriz[i][j] = np.complex(res_real[i][j], res_imag[i][j])
+
+print(matriz.shape)
+print(matriz)
+
+total_matriz = np.empty(shape = (res_real.shape[0]*2, res_real.shape[1]), dtype = complex)
+cong_matriz = matriz.conjugate()
+
+def unir(positivo, negativo, matriz_final):
+    for i in range(positivo.shape[0]):
+        for j in range(positivo.shape[1]):
+            matriz_final[i][j] = positivo[i][j]
+            matriz_final[matriz_final.shape[0] - 1 - i][j] = negativo[i][j]
+
+unir(matriz, cong_matriz, total_matriz)
+
+print("total_matriz ",total_matriz)
+print(total_matriz.shape)
+
+#Aplicar a eso la ifft
+final = np.fft.ifft(total_matriz)
+print("final ", final)
+
+#Aplicar Wavelet para quitar ruido: (funcion wden de matlab, filtro deb3, umbral: minimaxi, opcion de escala: sln, nivel de descoposicin: maximo
+sin_ruido = pywt.wavedec(final,'db3')
+print("SIN RUIDO", sin_ruido)
+
+#Exportar Audio
+sonidofinal=np.array(sin_ruido[0], dtype='int8')
+wavfile.write('salida_' + sys.argv[1], fs, sonidofinal)