def convolute(img, kernel):
output = oc.conv2(img, kernel, 'same')
# output = cv2.filter2D(img, cv2.CV_64F, kernel)
#plt.imsave(str(i)+str(j),output, cmap = plt.cm.gray)
return output / 18
image,
scaleFactor = 1.1,
minNeighbors = 5,
minSize = (30, 30),
flags = cv2.cv.CV_HAAR_SCALE_IMAGE
)
#Draw a rectangle around the faces
for (x, y, w, h) in face:
roi = image[y:y+h, x:x+w]
roi = cv2.resize(roi,(157,157))
result = []
for k in [kernels]:
output = oc.conv2(roi,k,'same')
result.append(output)
print result[0].shape
total = result[0]
np.savez('tempwavelet',result[0].reshape((1,total.shape[0]*total.shape[1])))
'''
total = np.zeros(result[0].shape)
for i in range(18):
plt1 = plt.gca()
plt1.axes.get_xaxis().set_ticks([])
plt1.axes.get_yaxis().set_ticks([])
plt.subplot(6,3,i+1)
total += result[i]
plt.imshow(result[i],'gray')
'''
face = faceCascade.detectMultiScale(image,
scaleFactor=1.1,
minNeighbors=5,
minSize=(30, 30),
flags=cv2.cv.CV_HAAR_SCALE_IMAGE)
#Draw a rectangle around the faces
for (x, y, w, h) in face:
roi = image[y:y + h, x:x + w]
roi = cv2.resize(roi, (157, 157))
result = []
for k in [kernels]:
output = oc.conv2(roi, k, 'same')
result.append(output)
print result[0].shape
total = result[0]
np.savez('tempwavelet', result[0].reshape(
(1, total.shape[0] * total.shape[1])))
'''
total = np.zeros(result[0].shape)
for i in range(18):
plt1 = plt.gca()
plt1.axes.get_xaxis().set_ticks([])
plt1.axes.get_yaxis().set_ticks([])
plt.subplot(6,3,i+1)
total += result[i]
plt.imshow(result[i],'gray')
def convolute(img,kernel): output = oc.conv2(img,kernel,'same') # output = cv2.filter2D(img, cv2.CV_64F, kernel) #plt.imsave(str(i)+str(j),output, cmap = plt.cm.gray) return output/18
