Related
I am trying to perform a 2d convolution in python using numpy
I have a 2d array as follows with kernel H_r for the rows and H_c for the columns
data = np.zeros((nr, nc), dtype=np.float32)
#fill array with some data here then convolve
for r in range(nr):
data[r,:] = np.convolve(data[r,:], H_r, 'same')
for c in range(nc):
data[:,c] = np.convolve(data[:,c], H_c, 'same')
data = data.astype(np.uint8);
It does not produce the output that I was expecting, does this code look OK, I think the problem is with the casting from float32 to 8bit. Whats the best way to do this
Thanks
Maybe it is not the most optimized solution, but this is an implementation I used before with numpy library for Python:
def convolution2d(image, kernel, bias):
m, n = kernel.shape
if (m == n):
y, x = image.shape
y = y - m + 1
x = x - m + 1
new_image = np.zeros((y,x))
for i in range(y):
for j in range(x):
new_image[i][j] = np.sum(image[i:i+m, j:j+m]*kernel) + bias
return new_image
I hope this code helps other guys with the same doubt.
Regards.
Edit [Jan 2019]
#Tashus comment bellow is correct, and #dudemeister's answer is thus probably more on the mark. The function he suggested is also more efficient, by avoiding a direct 2D convolution and the number of operations that would entail.
Possible Problem
I believe you are doing two 1d convolutions, the first per columns and the second per rows, and replacing the results from the first with the results of the second.
Notice that numpy.convolve with the 'same' argument returns an array of equal shape to the largest one provided, so when you make the first convolution you already populated the entire data array.
One good way to visualize your arrays during these steps is to use Hinton diagrams, so you can check which elements already have a value.
Possible Solution
You can try to add the results of the two convolutions (use data[:,c] += .. instead of data[:,c] = on the second for loop), if your convolution matrix is the result of using the one dimensional H_r and H_c matrices like so:
Another way to do that would be to use scipy.signal.convolve2d with a 2d convolution array, which is probably what you wanted to do in the first place.
Since you already have your kernel separated you should simply use the sepfir2d function from scipy:
from scipy.signal import sepfir2d
convolved = sepfir2d(data, H_r, H_c)
On the other hand, the code you have there looks all right ...
I checked out many implementations and found none for my purpose, which should be really simple. So here is a dead-simple implementation with for loop
def convolution2d(image, kernel, stride, padding):
image = np.pad(image, [(padding, padding), (padding, padding)], mode='constant', constant_values=0)
kernel_height, kernel_width = kernel.shape
padded_height, padded_width = image.shape
output_height = (padded_height - kernel_height) // stride + 1
output_width = (padded_width - kernel_width) // stride + 1
new_image = np.zeros((output_height, output_width)).astype(np.float32)
for y in range(0, output_height):
for x in range(0, output_width):
new_image[y][x] = np.sum(image[y * stride:y * stride + kernel_height, x * stride:x * stride + kernel_width] * kernel).astype(np.float32)
return new_image
It might not be the most optimized solution either, but it is approximately ten times faster than the one proposed by #omotto and it only uses basic numpy function (as reshape, expand_dims, tile...) and no 'for' loops:
def gen_idx_conv1d(in_size, ker_size):
"""
Generates a list of indices. This indices correspond to the indices
of a 1D input tensor on which we would like to apply a 1D convolution.
For instance, with a 1D input array of size 5 and a kernel of size 3, the
1D convolution product will successively looks at elements of indices [0,1,2],
[1,2,3] and [2,3,4] in the input array. In this case, the function idx_conv1d(5,3)
outputs the following array: array([0,1,2,1,2,3,2,3,4]).
args:
in_size: (type: int) size of the input 1d array.
ker_size: (type: int) kernel size.
return:
idx_list: (type: np.array) list of the successive indices of the 1D input array
access to the 1D convolution algorithm.
example:
>>> gen_idx_conv1d(in_size=5, ker_size=3)
array([0, 1, 2, 1, 2, 3, 2, 3, 4])
"""
f = lambda dim1, dim2, axis: np.reshape(np.tile(np.expand_dims(np.arange(dim1),axis),dim2),-1)
out_size = in_size-ker_size+1
return f(ker_size, out_size, 0)+f(out_size, ker_size, 1)
def repeat_idx_2d(idx_list, nbof_rep, axis):
"""
Repeats an array of indices (idx_list) a number of time (nbof_rep) "along" an axis
(axis). This function helps to browse through a 2d array of size
(len(idx_list),nbof_rep).
args:
idx_list: (type: np.array or list) a 1D array of indices.
nbof_rep: (type: int) number of repetition.
axis: (type: int) axis "along" which the repetition will be applied.
return
idx_list: (type: np.array) a 1D array of indices of size len(idx_list)*nbof_rep.
example:
>>> a = np.array([0, 1, 2])
>>> repeat_idx_2d(a, 3, 0) # repeats array 'a' 3 times along 'axis' 0
array([0, 0, 0, 1, 1, 1, 2, 2, 2])
>>> repeat_idx_2d(a, 3, 1) # repeats array 'a' 3 times along 'axis' 1
array([0, 1, 2, 0, 1, 2, 0, 1, 2])
>>> b = np.reshape(np.arange(3*4), (3,4))
>>> b[repeat_idx_2d(np.arange(3), 4, 0), repeat_idx_2d(np.arange(4), 3, 1)]
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11])
"""
assert axis in [0,1], "Axis should be equal to 0 or 1."
tile_axis = (nbof_rep,1) if axis else (1,nbof_rep)
return np.reshape(np.tile(np.expand_dims(idx_list, 1),tile_axis),-1)
def conv2d(im, ker):
"""
Performs a 'valid' 2D convolution on an image. The input image may be
a 2D or a 3D array.
The output image first two dimensions will be reduced depending on the
convolution size.
The kernel may be a 2D or 3D array. If 2D, it will be applied on every
channel of the input image. If 3D, its last dimension must match the
image one.
args:
im: (type: np.array) image (2D or 3D).
ker: (type: np.array) convolution kernel (2D or 3D).
returns:
im: (type: np.array) convolved image.
example:
>>> im = np.reshape(np.arange(10*10*3),(10,10,3))/(10*10*3) # 3D image
>>> ker = np.array([[0,1,0],[-1,0,1],[0,-1,0]]) # 2D kernel
>>> conv2d(im, ker) # 3D array of shape (8,8,3)
"""
if len(im.shape)==2: # if the image is a 2D array, it is reshaped by expanding the last dimension
im = np.expand_dims(im,-1)
im_x, im_y, im_w = im.shape
if len(ker.shape)==2: # if the kernel is a 2D array, it is reshaped so it will be applied to all of the image channels
ker = np.tile(np.expand_dims(ker,-1),[1,1,im_w]) # the same kernel will be applied to all of the channels
assert ker.shape[-1]==im.shape[-1], "Kernel and image last dimension must match."
ker_x = ker.shape[0]
ker_y = ker.shape[1]
# shape of the output image
out_x = im_x - ker_x + 1
out_y = im_y - ker_y + 1
# reshapes the image to (out_x, ker_x, out_y, ker_y, im_w)
idx_list_x = gen_idx_conv1d(im_x, ker_x) # computes the indices of a 1D conv (cf. idx_conv1d doc)
idx_list_y = gen_idx_conv1d(im_y, ker_y)
idx_reshaped_x = repeat_idx_2d(idx_list_x, len(idx_list_y), 0) # repeats the previous indices to be used in 2D (cf. repeat_idx_2d doc)
idx_reshaped_y = repeat_idx_2d(idx_list_y, len(idx_list_x), 1)
im_reshaped = np.reshape(im[idx_reshaped_x, idx_reshaped_y, :], [out_x, ker_x, out_y, ker_y, im_w]) # reshapes
# reshapes the 2D kernel
ker = np.reshape(ker,[1, ker_x, 1, ker_y, im_w])
# applies the kernel to the image and reduces the dimension back to the one of original input image
return np.squeeze(np.sum(im_reshaped*ker, axis=(1,3)))
I tried to add a lot of comments to explain the method but the global idea is to reshape the 3D input image to a 5D one of shape (output_image_height, kernel_height, output_image_width, kernel_width, output_image_channel) and then to apply the kernel directly using the basic array multiplication. Of course, this methods is then using more memory (during the execution the size of the image is thus multiply by kernel_height*kernel_width) but it is faster.
To do this reshape step, I 'over-used' the indexing methods of numpy arrays, especially, the possibility of giving a numpy array as indices into a numpy array.
This methods could also be used to re-code the 2D convolution product in Pytorch or Tensorflow using the base math functions but I have no doubt in saying that it will be slower than the existing nn.conv2d operator...
I really enjoyed coding this method by only using the numpy basic tools.
One of the most obvious is to hard code the kernel.
img = img.convert('L')
a = np.array(img)
out = np.zeros([a.shape[0]-2, a.shape[1]-2], dtype='float')
out += a[:-2, :-2]
out += a[1:-1, :-2]
out += a[2:, :-2]
out += a[:-2, 1:-1]
out += a[1:-1,1:-1]
out += a[2:, 1:-1]
out += a[:-2, 2:]
out += a[1:-1, 2:]
out += a[2:, 2:]
out /= 9.0
out = out.astype('uint8')
img = Image.fromarray(out)
This example does a box blur 3x3 completely unrolled. You can multiply the values where you have a different value and divide them by a different amount. But, if you honestly want the quickest and dirtiest method this is it. I think it beats Guillaume Mougeot's method by a factor of like 5. His method beating the others by a factor of 10.
It may lose a few steps if you're doing something like a gaussian blur. and need to multiply some stuff.
Try to first round and then cast to uint8:
data = data.round().astype(np.uint8);
I wrote this convolve_stride which uses numpy.lib.stride_tricks.as_strided. Moreover it supports both strides and dilation. It is also compatible to tensor with order > 2.
import numpy as np
from numpy.lib.stride_tricks import as_strided
from im2col import im2col
def conv_view(X, F_s, dr, std):
X_s = np.array(X.shape)
F_s = np.array(F_s)
dr = np.array(dr)
Fd_s = (F_s - 1) * dr + 1
if np.any(Fd_s > X_s):
raise ValueError('(Dilated) filter size must be smaller than X')
std = np.array(std)
X_ss = np.array(X.strides)
Xn_s = (X_s - Fd_s) // std + 1
Xv_s = np.append(Xn_s, F_s)
Xv_ss = np.tile(X_ss, 2) * np.append(std, dr)
return as_strided(X, Xv_s, Xv_ss, writeable=False)
def convolve_stride(X, F, dr=None, std=None):
if dr is None:
dr = np.ones(X.ndim, dtype=int)
if std is None:
std = np.ones(X.ndim, dtype=int)
if not (X.ndim == F.ndim == len(dr) == len(std)):
raise ValueError('X.ndim, F.ndim, len(dr), len(std) must be the same')
Xv = conv_view(X, F.shape, dr, std)
return np.tensordot(Xv, F, axes=X.ndim)
%timeit -n 100 -r 10 convolve_stride(A, F)
#31.2 ms ± 1.31 ms per loop (mean ± std. dev. of 10 runs, 100 loops each)
Super simple and fast convolution using only basic numpy:
import numpy as np
def conv2d(image, kernel):
# apply kernel to image, return image of the same shape
# assume both image and kernel are 2D arrays
# kernel = np.flipud(np.fliplr(kernel)) # optionally flip the kernel
k = kernel.shape[0]
width = k//2
# place the image inside a frame to compensate for the kernel overlap
a = framed(image, width)
b = np.zeros(image.shape) # fill the output array with zeros; do not use np.empty()
# shift the image around each pixel, multiply by the corresponding kernel value and accumulate the results
for p, dp, r, dr in [(i, i + image.shape[0], j, j + image.shape[1]) for i in range(k) for j in range(k)]:
b += a[p:dp, r:dr] * kernel[p, r]
# or just write two nested for loops if you prefer
# np.clip(b, 0, 255, out=b) # optionally clip values exceeding the limits
return b
def framed(image, width):
a = np.zeros((image.shape[0]+2*width, image.shape[1]+2*width))
a[width:-width, width:-width] = image
# alternatively fill the frame with ones or copy border pixels
return a
Run it:
Image.fromarray(conv2d(image, kernel).astype('uint8'))
Instead of sliding the kernel along the image and computing the transformation pixel by pixel, create a series of shifted versions of the image corresponding to each element in the kernel and apply the corresponding kernel value to each of the shifted image versions.
This is probably the fastest you can get using just basic numpy; the speed is already comparable to C implementation of scipy convolve2d and better than fftconvolve. The idea is similar to #Tatarize. This example works only for one color component; for RGB just repeat for each (or modify the algorithm accordingly).
Typically, Convolution 2D is a misnomer. Ideally, under the hood,
whats being done is a correlation of 2 matrices.
pad == same
returns the output as the same as input dimension
It can also take asymmetric images. In order to perform correlation(convolution in deep learning lingo) on a batch of 2d matrices, one can iterate over all the channels, calculate the correlation for each of the channel slices with the respective filter slice.
For example: If image is (28,28,3) and filter size is (5,5,3) then take each of the 3 slices from the image channel and perform the cross correlation using the custom function above and stack the resulting matrix in the respective dimension of the output.
def get_cross_corr_2d(W, X, pad = 'valid'):
if(pad == 'same'):
pr = int((W.shape[0] - 1)/2)
pc = int((W.shape[1] - 1)/2)
conv_2d = np.zeros((X.shape[0], X.shape[1]))
X_pad = np.zeros((X.shape[0] + 2*pr, X.shape[1] + 2*pc))
X_pad[pr:pr+X.shape[0], pc:pc+X.shape[1]] = X
for r in range(conv_2d.shape[0]):
for c in range(conv_2d.shape[1]):
conv_2d[r,c] = np.sum(np.inner(W, X_pad[r:r+W.shape[0], c:c+W.shape[1]]))
return conv_2d
else:
pr = W.shape[0] - 1
pc = W.shape[1] - 1
conv_2d = np.zeros((X.shape[0] - W.shape[0] + 2*pr + 1,
X.shape[1] - W.shape[1] + 2*pc + 1))
X_pad = np.zeros((X.shape[0] + 2*pr, X.shape[1] + 2*pc))
X_pad[pr:pr+X.shape[0], pc:pc+X.shape[1]] = X
for r in range(conv_2d.shape[0]):
for c in range(conv_2d.shape[1]):
conv_2d[r,c] = np.sum(np.multiply(W, X_pad[r:r+W.shape[0], c:c+W.shape[1]]))
return conv_2d
This code incorrect:
for r in range(nr):
data[r,:] = np.convolve(data[r,:], H_r, 'same')
for c in range(nc):
data[:,c] = np.convolve(data[:,c], H_c, 'same')
See Nussbaumer transformation from multidimentional convolution to one dimentional.
I need to calculate the covariance matrix for RGB values across an image dataset, and then apply Cholesky decomposition to the final result.
The covariance matrix for RGB values is a 3x3 matrix M, where M_(i, i) is the variance of channel i and M_(i, j) is the covariance between channels i and j.
The end result should be something like this:
([[0.26, 0.09, 0.02],
[0.27, 0.00, -0.05],
[0.27, -0.09, 0.03]])
I'd prefer to stick to PyTorch functions even though Numpy has a Cov function.
I attempted to recreate the numpy Cov function in PyTorch here based on other cov implementations and clones:
def pytorch_cov(tensor, tensor2=None, rowvar=True):
if tensor2 is not None:
tensor = torch.cat((tensor, tensor2), dim=0)
tensor = tensor.view(1, -1) if tensor.dim() < 2 else tensor
tensor = tensor.t() if not rowvar and tensor.size(0) != 1 else tensor
tensor = tensor - torch.mean(tensor, dim=1, keepdim=True)
return 1 / (tensor.size(1) - 1) * tensor.mm(tensor.t())
def cov_vec(x):
c = x.size(0)
m1 = x - torch.sum(x, dim=[1],keepdims=True)/ c
out = torch.einsum('ijk,ilk->ijl',m1,m1) / (c - 1)
return out
The dataset loading would be like this:
dataset = torchvision.datasets.ImageFolder(data_path)
loader = torch.utils.data.DataLoader(dataset)
for images, _ in loader:
batch_size = images.size(0)
...
For the moment I'm just experimenting with images created with torch.randn(batch_size, 3, height, width).
Edit:
I'm attempting to replicate the matrix from Tensorflow's Lucid here, and somewhat explained on distill.pub here.
Second Edit:
In order to make the output resemble the example one, you have to do this instead of using Cholesky:
rgb_cov_tensor = rgb_cov_tensor / len(loader.dataset)
U,S,V = torch.svd(rgb_cov_tensor)
epsilon = 1e-10
svd_sqrt = U # torch.diag(torch.sqrt(S + epsilon))
The resulting matrix can then be used to perform color decorrelation, which is useful for visualizing features (DeepDream). I've implemented it in my project here.
Here is a function for computing the (unbiased) sample covariance matrix on a 3 channel image, named rgb_cov. Cholesky decomposition is straightforward with torch.cholesky:
import torch
def rgb_cov(im):
'''
Assuming im a torch.Tensor of shape (H,W,3):
'''
im_re = im.reshape(-1, 3)
im_re -= im_re.mean(0, keepdim=True)
return 1/(im_re.shape[0]-1) * im_re.T # im_re
#Test:
im = torch.randn(50,50,3)
cov = rgb_cov(im)
L_cholesky = torch.cholesky(cov)
I'd like to compute the cross correlation using de Fast Fourier Transform, for cloud motion tracking following the steps of the image below.
def roi_image(image):
image = cv.imread(image, 0)
roi = image[700:900, 1900:2100]
return roi
def FouTransf(image):
img_f32 = np.float32(image)
d_ft = cv.dft(img_f32, flags = cv.DFT_COMPLEX_OUTPUT)
d_ft_shift = np.fft.fftshift(d_ft)
rows, cols = image.shape
opt_rows = cv.getOptimalDFTSize(rows)
opt_cols = cv.getOptimalDFTSize(cols)
opt_img = np.zeros((opt_rows, opt_cols))
opt_img[:rows, :cols] = image
crow, ccol = opt_rows / 2 , opt_cols / 2
mask = np.zeros((opt_rows, opt_cols, 2), np.uint8)
mask[int(crow-50):int(crow+50), int(ccol-50):int(ccol+50)] = 1
f_mask = d_ft_shift*mask
return f_mask
def inv_FouTransf(image):
f_ishift = np.fft.ifftshift(image)
img_back = cv.idft(f_ishift)
img_back = cv.magnitude(img_back[:, :, 0], img_back[:, :, 1])
return img_back
def rms(sigma):
rms = np.std(sigma)
return rms
# Step 1: Import images
a = roi_image(path_a)
b = roi_image(path_b)
# Step 2: Convert the image to frequency domain
G_t0 = FouTransf(a)
G_t0_conj = G_t0.conj()
G_t1 = FouTransf(b)
# Step 3: Compute C(m, v)
C = G_t0_conj * G_t1
# Step 4: Convert the image to space domain to obtain Cov (p, q)
c_w = inv_FouTransf(C)
# Step 5: Compute Cross correlation
R_pq = c_w / (rms(a) * rms(b))
I'm a little confused because I've never use that technique. ¿The application es accurate?
HINT: eq (1) is : R(p,q) = Cov(p,q) / (sigma_t0 * sigma_t1). If more information is required the paper is: "An Automated Techinique or Obtaining Cloud Motion from Geostatiory Satellite Data Using Cross Correlation".
I found this source but I don't know if does something I'm trying.
If you are trying to do something similar to cv2.matchTemplate(), a working python implementation of the Normalized Cross-Correlation (NCC) method can be found in this repository:
########################################################################################
# Author: Ujash Joshi, University of Toronto, 2017 #
# Based on Octave implementation by: Benjamin Eltzner, 2014 <b.eltzner#gmx.de> #
# Octave/Matlab normxcorr2 implementation in python 3.5 #
# Details: #
# Normalized cross-correlation. Similiar results upto 3 significant digits. #
# https://github.com/Sabrewarrior/normxcorr2-python/master/norxcorr2.py #
# http://lordsabre.blogspot.ca/2017/09/matlab-normxcorr2-implemented-in-python.html #
########################################################################################
import numpy as np
from scipy.signal import fftconvolve
def normxcorr2(template, image, mode="full"):
"""
Input arrays should be floating point numbers.
:param template: N-D array, of template or filter you are using for cross-correlation.
Must be less or equal dimensions to image.
Length of each dimension must be less than length of image.
:param image: N-D array
:param mode: Options, "full", "valid", "same"
full (Default): The output of fftconvolve is the full discrete linear convolution of the inputs.
Output size will be image size + 1/2 template size in each dimension.
valid: The output consists only of those elements that do not rely on the zero-padding.
same: The output is the same size as image, centered with respect to the ‘full’ output.
:return: N-D array of same dimensions as image. Size depends on mode parameter.
"""
# If this happens, it is probably a mistake
if np.ndim(template) > np.ndim(image) or \
len([i for i in range(np.ndim(template)) if template.shape[i] > image.shape[i]]) > 0:
print("normxcorr2: TEMPLATE larger than IMG. Arguments may be swapped.")
template = template - np.mean(template)
image = image - np.mean(image)
a1 = np.ones(template.shape)
# Faster to flip up down and left right then use fftconvolve instead of scipy's correlate
ar = np.flipud(np.fliplr(template))
out = fftconvolve(image, ar.conj(), mode=mode)
image = fftconvolve(np.square(image), a1, mode=mode) - \
np.square(fftconvolve(image, a1, mode=mode)) / (np.prod(template.shape))
# Remove small machine precision errors after subtraction
image[np.where(image < 0)] = 0
template = np.sum(np.square(template))
out = out / np.sqrt(image * template)
# Remove any divisions by 0 or very close to 0
out[np.where(np.logical_not(np.isfinite(out)))] = 0
return out
The returned object from normxcorr2() is the cross correlation matrix.
For this lab I need to sample 150 x-values from a Normal distribution using a mean of 0 and standard deviation of 10, then from the x-values construct a design matrix using the features {1,x,x^2}.
We have to sample parameters and then use the design matrix to create y values for regression data.
The problem is that my design matrix isn't square, and the Moore-Penrose Pseduoinverse needs square matrices, but I don't know how to get that to work given the earlier setup of the lab?
This is what I've done
#Linear Regression Lab
import numpy as np
import math
data = np.random.normal(0, 10, 150)
design_matrix = np.zeros((150,3))
for i in range(150):
design_matrix[i][0] = 1
design_matrix[i][1] = data[i]
design_matrix[i][2] = pow(data[i], 2)
print("-------------------Design Matrix---------------------")
print("|--------1--------|-------x-------|--------x^2--------|")
print(design_matrix[:20])
#sampling paramters
theta_0 = np.random.uniform(low = -30, high = 20)
theta_1 = np.random.uniform(low = -30, high = 20)
theta_2 = np.random.uniform(low = -30, high = 20)
print(theta_0, theta_1, theta_2)
theta = np.array([theta_0, theta_1, theta_2])
theta = np.transpose(theta)
#moore penrose psuedo inverse
MPpi = np.linalg.pinv(design_matrix) ##problem here
y_values = np.linalg.inv(MPpi)
Feel free to edit this incomplete answer
After running this code on Repl, I got the following error message
Traceback (most recent call last):
File "main.py", line 32, in <module>
y_values = np.linalg.inv(MPpi)
File "<__array_function__ internals>", line 5, in inv
File "/home/runner/.local/share/virtualenvs/python3/lib/python3.8/site-packages/numpy/linalg/linalg.py", line 542, in inv
_assert_stacked_square(a)
File "/home/runner/.local/share/virtualenvs/python3/lib/python3.8/site-packages/numpy/linalg/linalg.py", line 213, in _assert_stacked_square
raise LinAlgError('Last 2 dimensions of the array mustbe square')
numpy.linalg.LinAlgError: Last 2 dimensions of the array must be square
The first error propagates from taking the inverse of MPpi
By looking at the docs, it seems that pinv switches the last two dimensions [e.g., an m x n matrix becomes n x m], so we will need to format the matrix before calculating the psuedoinverse
As far as the Moore Penrose inverse AKA pinv is concerned, this article suggests that multiplying MPpi*data, which will yield x_0 {notation from Ross MacAusland}, which is the best fit for your least squares regression.
I am trying to perform a 2d convolution in python using numpy
I have a 2d array as follows with kernel H_r for the rows and H_c for the columns
data = np.zeros((nr, nc), dtype=np.float32)
#fill array with some data here then convolve
for r in range(nr):
data[r,:] = np.convolve(data[r,:], H_r, 'same')
for c in range(nc):
data[:,c] = np.convolve(data[:,c], H_c, 'same')
data = data.astype(np.uint8);
It does not produce the output that I was expecting, does this code look OK, I think the problem is with the casting from float32 to 8bit. Whats the best way to do this
Thanks
Maybe it is not the most optimized solution, but this is an implementation I used before with numpy library for Python:
def convolution2d(image, kernel, bias):
m, n = kernel.shape
if (m == n):
y, x = image.shape
y = y - m + 1
x = x - m + 1
new_image = np.zeros((y,x))
for i in range(y):
for j in range(x):
new_image[i][j] = np.sum(image[i:i+m, j:j+m]*kernel) + bias
return new_image
I hope this code helps other guys with the same doubt.
Regards.
Edit [Jan 2019]
#Tashus comment bellow is correct, and #dudemeister's answer is thus probably more on the mark. The function he suggested is also more efficient, by avoiding a direct 2D convolution and the number of operations that would entail.
Possible Problem
I believe you are doing two 1d convolutions, the first per columns and the second per rows, and replacing the results from the first with the results of the second.
Notice that numpy.convolve with the 'same' argument returns an array of equal shape to the largest one provided, so when you make the first convolution you already populated the entire data array.
One good way to visualize your arrays during these steps is to use Hinton diagrams, so you can check which elements already have a value.
Possible Solution
You can try to add the results of the two convolutions (use data[:,c] += .. instead of data[:,c] = on the second for loop), if your convolution matrix is the result of using the one dimensional H_r and H_c matrices like so:
Another way to do that would be to use scipy.signal.convolve2d with a 2d convolution array, which is probably what you wanted to do in the first place.
Since you already have your kernel separated you should simply use the sepfir2d function from scipy:
from scipy.signal import sepfir2d
convolved = sepfir2d(data, H_r, H_c)
On the other hand, the code you have there looks all right ...
I checked out many implementations and found none for my purpose, which should be really simple. So here is a dead-simple implementation with for loop
def convolution2d(image, kernel, stride, padding):
image = np.pad(image, [(padding, padding), (padding, padding)], mode='constant', constant_values=0)
kernel_height, kernel_width = kernel.shape
padded_height, padded_width = image.shape
output_height = (padded_height - kernel_height) // stride + 1
output_width = (padded_width - kernel_width) // stride + 1
new_image = np.zeros((output_height, output_width)).astype(np.float32)
for y in range(0, output_height):
for x in range(0, output_width):
new_image[y][x] = np.sum(image[y * stride:y * stride + kernel_height, x * stride:x * stride + kernel_width] * kernel).astype(np.float32)
return new_image
It might not be the most optimized solution either, but it is approximately ten times faster than the one proposed by #omotto and it only uses basic numpy function (as reshape, expand_dims, tile...) and no 'for' loops:
def gen_idx_conv1d(in_size, ker_size):
"""
Generates a list of indices. This indices correspond to the indices
of a 1D input tensor on which we would like to apply a 1D convolution.
For instance, with a 1D input array of size 5 and a kernel of size 3, the
1D convolution product will successively looks at elements of indices [0,1,2],
[1,2,3] and [2,3,4] in the input array. In this case, the function idx_conv1d(5,3)
outputs the following array: array([0,1,2,1,2,3,2,3,4]).
args:
in_size: (type: int) size of the input 1d array.
ker_size: (type: int) kernel size.
return:
idx_list: (type: np.array) list of the successive indices of the 1D input array
access to the 1D convolution algorithm.
example:
>>> gen_idx_conv1d(in_size=5, ker_size=3)
array([0, 1, 2, 1, 2, 3, 2, 3, 4])
"""
f = lambda dim1, dim2, axis: np.reshape(np.tile(np.expand_dims(np.arange(dim1),axis),dim2),-1)
out_size = in_size-ker_size+1
return f(ker_size, out_size, 0)+f(out_size, ker_size, 1)
def repeat_idx_2d(idx_list, nbof_rep, axis):
"""
Repeats an array of indices (idx_list) a number of time (nbof_rep) "along" an axis
(axis). This function helps to browse through a 2d array of size
(len(idx_list),nbof_rep).
args:
idx_list: (type: np.array or list) a 1D array of indices.
nbof_rep: (type: int) number of repetition.
axis: (type: int) axis "along" which the repetition will be applied.
return
idx_list: (type: np.array) a 1D array of indices of size len(idx_list)*nbof_rep.
example:
>>> a = np.array([0, 1, 2])
>>> repeat_idx_2d(a, 3, 0) # repeats array 'a' 3 times along 'axis' 0
array([0, 0, 0, 1, 1, 1, 2, 2, 2])
>>> repeat_idx_2d(a, 3, 1) # repeats array 'a' 3 times along 'axis' 1
array([0, 1, 2, 0, 1, 2, 0, 1, 2])
>>> b = np.reshape(np.arange(3*4), (3,4))
>>> b[repeat_idx_2d(np.arange(3), 4, 0), repeat_idx_2d(np.arange(4), 3, 1)]
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11])
"""
assert axis in [0,1], "Axis should be equal to 0 or 1."
tile_axis = (nbof_rep,1) if axis else (1,nbof_rep)
return np.reshape(np.tile(np.expand_dims(idx_list, 1),tile_axis),-1)
def conv2d(im, ker):
"""
Performs a 'valid' 2D convolution on an image. The input image may be
a 2D or a 3D array.
The output image first two dimensions will be reduced depending on the
convolution size.
The kernel may be a 2D or 3D array. If 2D, it will be applied on every
channel of the input image. If 3D, its last dimension must match the
image one.
args:
im: (type: np.array) image (2D or 3D).
ker: (type: np.array) convolution kernel (2D or 3D).
returns:
im: (type: np.array) convolved image.
example:
>>> im = np.reshape(np.arange(10*10*3),(10,10,3))/(10*10*3) # 3D image
>>> ker = np.array([[0,1,0],[-1,0,1],[0,-1,0]]) # 2D kernel
>>> conv2d(im, ker) # 3D array of shape (8,8,3)
"""
if len(im.shape)==2: # if the image is a 2D array, it is reshaped by expanding the last dimension
im = np.expand_dims(im,-1)
im_x, im_y, im_w = im.shape
if len(ker.shape)==2: # if the kernel is a 2D array, it is reshaped so it will be applied to all of the image channels
ker = np.tile(np.expand_dims(ker,-1),[1,1,im_w]) # the same kernel will be applied to all of the channels
assert ker.shape[-1]==im.shape[-1], "Kernel and image last dimension must match."
ker_x = ker.shape[0]
ker_y = ker.shape[1]
# shape of the output image
out_x = im_x - ker_x + 1
out_y = im_y - ker_y + 1
# reshapes the image to (out_x, ker_x, out_y, ker_y, im_w)
idx_list_x = gen_idx_conv1d(im_x, ker_x) # computes the indices of a 1D conv (cf. idx_conv1d doc)
idx_list_y = gen_idx_conv1d(im_y, ker_y)
idx_reshaped_x = repeat_idx_2d(idx_list_x, len(idx_list_y), 0) # repeats the previous indices to be used in 2D (cf. repeat_idx_2d doc)
idx_reshaped_y = repeat_idx_2d(idx_list_y, len(idx_list_x), 1)
im_reshaped = np.reshape(im[idx_reshaped_x, idx_reshaped_y, :], [out_x, ker_x, out_y, ker_y, im_w]) # reshapes
# reshapes the 2D kernel
ker = np.reshape(ker,[1, ker_x, 1, ker_y, im_w])
# applies the kernel to the image and reduces the dimension back to the one of original input image
return np.squeeze(np.sum(im_reshaped*ker, axis=(1,3)))
I tried to add a lot of comments to explain the method but the global idea is to reshape the 3D input image to a 5D one of shape (output_image_height, kernel_height, output_image_width, kernel_width, output_image_channel) and then to apply the kernel directly using the basic array multiplication. Of course, this methods is then using more memory (during the execution the size of the image is thus multiply by kernel_height*kernel_width) but it is faster.
To do this reshape step, I 'over-used' the indexing methods of numpy arrays, especially, the possibility of giving a numpy array as indices into a numpy array.
This methods could also be used to re-code the 2D convolution product in Pytorch or Tensorflow using the base math functions but I have no doubt in saying that it will be slower than the existing nn.conv2d operator...
I really enjoyed coding this method by only using the numpy basic tools.
One of the most obvious is to hard code the kernel.
img = img.convert('L')
a = np.array(img)
out = np.zeros([a.shape[0]-2, a.shape[1]-2], dtype='float')
out += a[:-2, :-2]
out += a[1:-1, :-2]
out += a[2:, :-2]
out += a[:-2, 1:-1]
out += a[1:-1,1:-1]
out += a[2:, 1:-1]
out += a[:-2, 2:]
out += a[1:-1, 2:]
out += a[2:, 2:]
out /= 9.0
out = out.astype('uint8')
img = Image.fromarray(out)
This example does a box blur 3x3 completely unrolled. You can multiply the values where you have a different value and divide them by a different amount. But, if you honestly want the quickest and dirtiest method this is it. I think it beats Guillaume Mougeot's method by a factor of like 5. His method beating the others by a factor of 10.
It may lose a few steps if you're doing something like a gaussian blur. and need to multiply some stuff.
Try to first round and then cast to uint8:
data = data.round().astype(np.uint8);
I wrote this convolve_stride which uses numpy.lib.stride_tricks.as_strided. Moreover it supports both strides and dilation. It is also compatible to tensor with order > 2.
import numpy as np
from numpy.lib.stride_tricks import as_strided
from im2col import im2col
def conv_view(X, F_s, dr, std):
X_s = np.array(X.shape)
F_s = np.array(F_s)
dr = np.array(dr)
Fd_s = (F_s - 1) * dr + 1
if np.any(Fd_s > X_s):
raise ValueError('(Dilated) filter size must be smaller than X')
std = np.array(std)
X_ss = np.array(X.strides)
Xn_s = (X_s - Fd_s) // std + 1
Xv_s = np.append(Xn_s, F_s)
Xv_ss = np.tile(X_ss, 2) * np.append(std, dr)
return as_strided(X, Xv_s, Xv_ss, writeable=False)
def convolve_stride(X, F, dr=None, std=None):
if dr is None:
dr = np.ones(X.ndim, dtype=int)
if std is None:
std = np.ones(X.ndim, dtype=int)
if not (X.ndim == F.ndim == len(dr) == len(std)):
raise ValueError('X.ndim, F.ndim, len(dr), len(std) must be the same')
Xv = conv_view(X, F.shape, dr, std)
return np.tensordot(Xv, F, axes=X.ndim)
%timeit -n 100 -r 10 convolve_stride(A, F)
#31.2 ms ± 1.31 ms per loop (mean ± std. dev. of 10 runs, 100 loops each)
Super simple and fast convolution using only basic numpy:
import numpy as np
def conv2d(image, kernel):
# apply kernel to image, return image of the same shape
# assume both image and kernel are 2D arrays
# kernel = np.flipud(np.fliplr(kernel)) # optionally flip the kernel
k = kernel.shape[0]
width = k//2
# place the image inside a frame to compensate for the kernel overlap
a = framed(image, width)
b = np.zeros(image.shape) # fill the output array with zeros; do not use np.empty()
# shift the image around each pixel, multiply by the corresponding kernel value and accumulate the results
for p, dp, r, dr in [(i, i + image.shape[0], j, j + image.shape[1]) for i in range(k) for j in range(k)]:
b += a[p:dp, r:dr] * kernel[p, r]
# or just write two nested for loops if you prefer
# np.clip(b, 0, 255, out=b) # optionally clip values exceeding the limits
return b
def framed(image, width):
a = np.zeros((image.shape[0]+2*width, image.shape[1]+2*width))
a[width:-width, width:-width] = image
# alternatively fill the frame with ones or copy border pixels
return a
Run it:
Image.fromarray(conv2d(image, kernel).astype('uint8'))
Instead of sliding the kernel along the image and computing the transformation pixel by pixel, create a series of shifted versions of the image corresponding to each element in the kernel and apply the corresponding kernel value to each of the shifted image versions.
This is probably the fastest you can get using just basic numpy; the speed is already comparable to C implementation of scipy convolve2d and better than fftconvolve. The idea is similar to #Tatarize. This example works only for one color component; for RGB just repeat for each (or modify the algorithm accordingly).
Typically, Convolution 2D is a misnomer. Ideally, under the hood,
whats being done is a correlation of 2 matrices.
pad == same
returns the output as the same as input dimension
It can also take asymmetric images. In order to perform correlation(convolution in deep learning lingo) on a batch of 2d matrices, one can iterate over all the channels, calculate the correlation for each of the channel slices with the respective filter slice.
For example: If image is (28,28,3) and filter size is (5,5,3) then take each of the 3 slices from the image channel and perform the cross correlation using the custom function above and stack the resulting matrix in the respective dimension of the output.
def get_cross_corr_2d(W, X, pad = 'valid'):
if(pad == 'same'):
pr = int((W.shape[0] - 1)/2)
pc = int((W.shape[1] - 1)/2)
conv_2d = np.zeros((X.shape[0], X.shape[1]))
X_pad = np.zeros((X.shape[0] + 2*pr, X.shape[1] + 2*pc))
X_pad[pr:pr+X.shape[0], pc:pc+X.shape[1]] = X
for r in range(conv_2d.shape[0]):
for c in range(conv_2d.shape[1]):
conv_2d[r,c] = np.sum(np.inner(W, X_pad[r:r+W.shape[0], c:c+W.shape[1]]))
return conv_2d
else:
pr = W.shape[0] - 1
pc = W.shape[1] - 1
conv_2d = np.zeros((X.shape[0] - W.shape[0] + 2*pr + 1,
X.shape[1] - W.shape[1] + 2*pc + 1))
X_pad = np.zeros((X.shape[0] + 2*pr, X.shape[1] + 2*pc))
X_pad[pr:pr+X.shape[0], pc:pc+X.shape[1]] = X
for r in range(conv_2d.shape[0]):
for c in range(conv_2d.shape[1]):
conv_2d[r,c] = np.sum(np.multiply(W, X_pad[r:r+W.shape[0], c:c+W.shape[1]]))
return conv_2d
This code incorrect:
for r in range(nr):
data[r,:] = np.convolve(data[r,:], H_r, 'same')
for c in range(nc):
data[:,c] = np.convolve(data[:,c], H_c, 'same')
See Nussbaumer transformation from multidimentional convolution to one dimentional.