def showCortexImg(pV, nV):
"""Summary
This function encapsulates the routine to generate rectified cortical views of
all opponent retinal ganglion cells
Args:
pV (vector): Positive rectified imagevector
nV (vector): Negative rectified imagevector
Returns:
mergecort: Return a merged image of all cortical opponent cells as a numpy image array
"""
# object arrays of the positive and negative images
pos_cort_img = np.empty(8, dtype=object)
neg_cort_img = np.empty(8, dtype=object)
for t in range(8):
# cortical mapping functions
lpos, rpos = cortex.cort_img(pV[:, t, :], L, L_loc, R, R_loc,
cort_size, G)
lneg, rneg = cortex.cort_img(nV[:, t, :], L, L_loc, R, R_loc,
cort_size, G)
pos_cort_img[t] = np.concatenate((np.rot90(lpos), np.rot90(rpos, k=3)),
axis=1)
neg_cort_img[t] = np.concatenate((np.rot90(lneg), np.rot90(rneg, k=3)),
axis=1)
# stack all images into a grid
posRGcort = np.vstack((pos_cort_img[:4]))
negRGcort = np.vstack((neg_cort_img[:4]))
posYBcort = np.vstack((pos_cort_img[4:]))
negYBcort = np.vstack((neg_cort_img[4:]))
mergecort = np.concatenate((posRGcort, negRGcort, posYBcort, negYBcort),
axis=1)
return mergecort
def showNonOpponency(C, theta):
"""Summary
This function encapsulates the routine to generate backprojected and cortical views for
the magnocellular pathway retinal ganglion cells
Args:
C (vector): The sharp retina is passed to the function
theta (float): A threshold value is passed to the function
Returns:
merged: Return a merged image of the backprojected view as a numpy image array
mergecort: Return a merged image of the cortical view as a numpy image array
"""
GI = retina.gauss_norm_img(x,
y,
dcoeff[i],
dloc[i],
imsize=imgsize,
rgb=False)
# Sample using the other recepetive field, note there is no temporal response with still images
S = retina.sample(img, x, y, dcoeff[i], dloc[i], rgb=True)
#backproject the imagevectors
ncentreV, nsurrV = rgc.nonopponency(C, S, theta)
ninverse = retina.inverse(ncentreV,
x,
y,
dcoeff[i],
dloc[i],
GI,
imsize=imgsize,
rgb=True)
ninv_crop = retina.crop(ninverse, x, y, dloc[i])
ninverse2 = retina.inverse(nsurrV,
x,
y,
dcoeff[i],
dloc[i],
GI,
imsize=imgsize,
rgb=True)
ninv_crop2 = retina.crop(ninverse2, x, y, dloc[i])
# place descriptive text onto generated images
cv2.putText(ninv_crop, "R+G + ", (xx, yy), font, 1, (255, 255, 255), 2)
cv2.putText(ninv_crop2, "R+G - ", (xx, yy), font, 1, (255, 255, 255), 2)
merged = np.concatenate((ninv_crop, ninv_crop2), axis=1)
# create cortical maps of the imagevectors
lposnon, rposnon = cortex.cort_img(ncentreV, L, L_loc, R, R_loc, cort_size,
G)
lnegnon, rnegnon = cortex.cort_img(nsurrV, L, L_loc, R, R_loc, cort_size,
G)
pos_cort_img = np.concatenate((np.rot90(lposnon), np.rot90(rposnon, k=3)),
axis=1)
neg_cort_img = np.concatenate((np.rot90(lnegnon), np.rot90(rnegnon, k=3)),
axis=1)
mergecort = np.concatenate((pos_cort_img, neg_cort_img), axis=1)
return merged, mergecort
def showNonOpponency(C, theta):
"""Summary
This function encapsulates the routine to generate backprojected and cortical views for
the magnocellular pathway retinal ganglion cells
Args:
C (vector): The sharp retina is passed to the function
theta (float): A threshold value is passed to the function
Returns:
merged: Return a merged image of the backprojected view as a numpy image array
mergecort: Return a merged image of the cortical view as a numpy image array
"""
# Sample using the other recepetive field, but with a temporally different image, lateimg
S = retina.sample(lateimg, x, y, dcoeff[i], dloc[i], rgb=True)
ncentreV, nsurrV = rgc.nonopponency(C, S, theta)
ninverse = retina.inverse(ncentreV,
x,
y,
dcoeff[i],
dloc[i],
GI,
imsize=imgsize,
rgb=False)
ninv_crop = retina.crop(ninverse, x, y, dloc[i])
ninverse2 = retina.inverse(nsurrV,
x,
y,
dcoeff[i],
dloc[i],
GI,
imsize=imgsize,
rgb=False)
ninv_crop2 = retina.crop(ninverse2, x, y, dloc[i])
merged = np.concatenate((ninv_crop, ninv_crop2), axis=1)
lposnon, rposnon = cortex.cort_img(ncentreV, L, L_loc, R, R_loc, cort_size,
G)
lnegnon, rnegnon = cortex.cort_img(nsurrV, L, L_loc, R, R_loc, cort_size,
G)
pos_cort_img = np.concatenate((np.rot90(lposnon), np.rot90(rposnon, k=3)),
axis=1)
neg_cort_img = np.concatenate((np.rot90(lnegnon), np.rot90(rnegnon, k=3)),
axis=1)
mergecort = np.concatenate((pos_cort_img, neg_cort_img), axis=1)
return merged, mergecort
def speedup(loc, coeff, img, rgb, show_res):
'''
This test measures the performance of the two implementation
from initialisation to the end of the cortical transform
'''
init_p = time.time()
GI = retina.gauss_norm_img(int(img.shape[1] / 2), int(img.shape[0] / 2),
coeff, loc, img.shape, rgb)
init_c = time.time()
ret = retina_cuda.create_retina(
loc, coeff, img.shape, (int(img.shape[1] / 2), int(img.shape[0] / 2)))
sample_p = time.time()
V_p = retina.sample(img, img.shape[1] / 2, img.shape[0] / 2, coeff, loc,
rgb)
sample_c = time.time()
V_c = ret.sample(img)
invert_p = time.time()
inv_p = retina.inverse(V_p, img.shape[1] / 2, img.shape[0] / 2, coeff, loc,
GI, img.shape, rgb)
invert_c = time.time()
inv_c = ret.inverse(V_c)
retina_end = time.time()
cort_init_p = time.time()
L, R = cortex.LRsplit(loc)
L_loc, R_loc = cortex.cort_map(L, R)
L_loc, R_loc, G, cort_size = cortex.cort_prepare(L_loc, R_loc)
cort_init_c = time.time()
cort = cortex_cuda.create_cortex_from_fields(loc, rgb=rgb)
cort_img_p = time.time()
l_p, r_p = cortex.cort_img(V_p, L, L_loc, R, R_loc, cort_size, G)
cort_img_c = time.time()
l_c = cort.cort_image_left(V_c)
r_c = cort.cort_image_right(V_c)
cort_end = time.time()
print '%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,' % (init_c - init_p, sample_p - init_c, sample_c - sample_p, \
invert_p - sample_c, invert_c - invert_p, retina_end - invert_c,\
cort_init_c - cort_init_p, cort_img_p - cort_init_c, cort_img_c - cort_img_p, cort_end - cort_img_c)
if show_res:
cv2.namedWindow("inverse CUDA", cv2.WINDOW_NORMAL)
cv2.imshow("inverse CUDA", inv_c)
cv2.namedWindow("inverse Piotr", cv2.WINDOW_NORMAL)
cv2.imshow("inverse Piotr", inv_p)
c_c = np.concatenate((np.rot90(l_c), np.rot90(r_c, k=3)), axis=1)
c_p = np.concatenate((np.rot90(l_p), np.rot90(r_p, k=3)), axis=1)
cv2.namedWindow("cortex CUDA", cv2.WINDOW_NORMAL)
cv2.imshow("cortex CUDA", c_c)
cv2.namedWindow("cortex Piotr", cv2.WINDOW_NORMAL)
cv2.imshow("cortex Piotr", c_p)
def showCortexImg(pV, nV, t):
"""Summary
This function encapsulates the routine to generate rectified cortical views of
one opponent retinal ganglion cell type
Args:
pV (vector): Positive rectified imagevector
nV (vector): Negative rectified imagevector
t (int): Index position of opponent cell species
Returns:
mergecort: Return a merged image of all cortical opponent cells as a numpy image array
"""
lpos, rpos = cortex.cort_img(pV[:, t, :], L, L_loc, R, R_loc, cort_size, G)
lneg, rneg = cortex.cort_img(nV[:, t, :], L, L_loc, R, R_loc, cort_size, G)
pos_cort_img = np.concatenate((np.rot90(lpos), np.rot90(rpos, k=3)),
axis=1)
neg_cort_img = np.concatenate((np.rot90(lneg), np.rot90(rneg, k=3)),
axis=1)
mergecort = np.concatenate((pos_cort_img, neg_cort_img), axis=1)
return mergecort
def correctness_test(loc, coeff, cap, rgb=False):
'''
CUDA code uses the minimal initialisation from the host,
all tracatable values are computed on the GPU
Get an image from the camera, generate inverse and cortical image
with both implementation and subtract the results
'''
r, img = cap.read()
if not rgb: img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# create CUDA objects to pass to evaluation
ret = retina_cuda.create_retina(
loc, coeff, img.shape, (int(img.shape[1] / 2), int(img.shape[0] / 2)),
None)
cort = cortex_cuda.create_cortex_from_fields(loc, rgb=rgb)
while ord('q') != cv2.waitKey(10):
r, img = cap.read()
if not rgb: img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
if r:
'''
Sample the image img with CUDA retina ret, inverse transform it with ret and
create the cortical image with CUDA cortex cort
Sample and generate retina and cortical images from img with Piotrs's code
Visually compare the results by showing the subtraction of the generatd images
'''
V_c = ret.sample(img) # sample with CUDA
inv_c = ret.inverse(V_c) # inverse with CUDA
l_c = cort.cort_image_left(V_c) # left cortical image CUDA
r_c = cort.cort_image_right(V_c) # right cortical image CUDA
c_c = np.concatenate(
(np.rot90(l_c), np.rot90(r_c, k=3)),
axis=1) #concatenate the results into one image
# create Piotr's retina and cortical images
GI = retina.gauss_norm_img(int(img.shape[1] / 2),
int(img.shape[0] / 2), coeff, loc,
img.shape, rgb)
L, R = cortex.LRsplit(loc)
L_loc, R_loc = cortex.cort_map(L, R)
L_loc, R_loc, G, cort_size = cortex.cort_prepare(L_loc, R_loc)
V_p = retina.sample(img, img.shape[1] / 2, img.shape[0] / 2, coeff,
loc, rgb)
inv_p = retina.inverse(V_p, img.shape[1] / 2, img.shape[0] / 2,
coeff, loc, GI, img.shape, rgb)
l_p, r_p = cortex.cort_img(V_p, L, L_loc, R, R_loc, cort_size, G)
c_p = np.concatenate((np.rot90(
l_p[:l_c.shape[0], :]), np.rot90(r_p[:r_c.shape[0], :], k=3)),
axis=1)
# show CUDA results
cv2.namedWindow("inverse CUDA", cv2.WINDOW_NORMAL)
cv2.imshow("inverse CUDA", inv_c)
cv2.namedWindow("cortex CUDA", cv2.WINDOW_NORMAL)
cv2.imshow("cortex CUDA", c_c)
# show Piotr's results
cv2.namedWindow("inverse Piotr", cv2.WINDOW_NORMAL)
cv2.imshow("inverse Piotr", inv_p)
cv2.namedWindow("cortex Piotr", cv2.WINDOW_NORMAL)
cv2.imshow("cortex Piotr", c_p)
# show the difference of the images
cv2.namedWindow("inverse diff", cv2.WINDOW_NORMAL)
cv2.imshow("inverse diff", np.power((inv_c - inv_p), 2) * 255)
cv2.namedWindow("cortex diff", cv2.WINDOW_NORMAL)
cv2.imshow("cortex diff", np.power((c_c - c_p), 2) * 255)