Finding the optimal tiling strategy using squares of different sizes - algorithm

I have shapes constructed out of 8x8 squares. I need to tile them using the fewest number of squares of size 8x8, 16x16, 32x32 and 64x64. Four 8x8 squares arranged in a square can be replaced by a single 16x16 square, e.g.:
What algorithm can be used to achieve this?

This calls for a dynamic programming solution. I'll assume we have a square[r][c] array of booleans which is true if (r, c) has a 1x1 square (I've simplified the solution to work with 1x1, 2x2, 4x4 and 8x8 squares to make it easier to follow, but it's easy to adapt). Pad it with a wall of false sentinel values on the top row and left column.
Define a 2d count array, where count[r][c] refers to the number of 1x1 squares in the region above and to the left of (r, c). We can add them up using a dp algorithm:
count[0..n][0..m] = 0
for i in 1..n:
for j in 1..m:
count[i][j] = count[i-1][j] + count[i][j-1] -
count[i-1][j-1] + square[i][j]
The above works by adding up two regions we already know the sum of, subtracting the doubly-counted area and adding in the new cell. Using the count array, we can test if a square region is fully covered in 1x1 squares in constant time using a similar method:
// p1 is the top-left coordinate, p2 the bottom-right
function region_count(p1, p2):
return count[p1.r][p1.c] - count[p1.r][p2.c-1] -
count[p2.r-1][p1.c] + 2*count[p2.r-1][p2.c-1]
We then create a second 2d min_squares array, where min_squares[r][c] refers to the minimum number of squares required to cover the original 1x1 squares. These values can be calculates using another dp:
min_squares = count
for i in 1..n:
for j in 1..m:
for size in [2, 4, 8]:
if i >= size and j >= size and
region_count((i-size, j-size), (i, j)) == size*size:
min_squares[i][j] = min(min_squares[i][j],
min_squares[i-size-1][j] +
min_squares[i][j-size-1] -
min_squares[i-size-1][j-size-1] +
1)
In order to reconstruct the tiling needed to get the calculated minimum, we use an auxiliary size_used[r][c] array which we use to keep track of the size of square placed at (r, c). From this we can recursively reconstruct the tiling:
function reconstruct(i, j):
if i < 0 or j < 0:
return
place square of size size_used[i][j] at (i-size_used[i][j]+1, j-size_used[i][j]+1)
reconstruct(i-size_used[i][j], j)
reconstruct(i, j-size_used[i][j])

You might want to look at Optimal way for partitioning a cell based shape into a minimal amount of rectangles - if I understood correctly, this is the same problem but for rectangles instead of squares.

Related

Fast points in circle test with numpy

I have a large number of (x,y) grid points with integer coordinates which i want to test if they are in small number of circles given by radius and center. The points are some marked parts of an image, which means there are a small number of irregular shaped blocks, which contain the points. There i want to check for collisions and count the number of points inside a circle. My current approaches are rather slow (with python and numpy).
Now i have two tasks:
Test, if any point of set A are in any circle
Count the number of points of set B, which are in a circle
My current implementation looks like this (setA and setB are Nx2 numpy arrays and center is a 1x2 array.):
1) For each circle create an array of point - center, square it elementwise and take the sum, then check if it's smaller than radius**2
for circle in circles:
if (((setA - circle.center)**2).sum(axis=1) < circle.radius**2).any():
return "collision"
return "no collision"
This could be optimized by using a python loop and breaking on the first collision, but usually numpy loops are a lot faster than python loops and actually both versions were slower than expected.
2) For each circle create an array of distances and do an elementwise less than radius test. Add up all arrays and count the non-zero elements of the result.
pixels = sp.zeros(len(setB))
for circle in circles:
pixels += (((setB - circle.center)**2).sum(axis=1) < circle.radius**2)
return np.count_nonzero(pixels)
Is there an easy option to speedup this?
I do not want to over optimize (and make the program a lot more complicated), but just to use numpy in the most efficient way, using the numpy vectorization as much as possible.
So building the most perfect spatial tree or similiar isn't my goal, but i think a O(n^2) algorithm for a few thousand points and 10-20 circles should be possible in as fast way on an average desktop computer today.
Taking advantage of coordinates being integers:
create a lookup image
radius = max([circle.radius for circle in circles])
mask = np.zeros((image.shape[0] + 2*radius, image.shape[1] + 2*radius), dtype=int)
for circle in circles:
center = circle.center + radius
mask[center[0]-circle.radius:center[0]+circle.radius + 1,
center[1]-circle.radius:center[1]+circle.radius + 1] += circle.mask
circle.mask is a small square patch containing a mask of the disc of interior points
counting collisions is now as easy as
mask[radius:-radius, radius:-radius][setB[:,0], setB[:,1]].sum()
fast creation of discs (no multiplications, no square roots):
r = circle.radius
h2 = np.r_[0, np.add.accumulate(np.arange(1, 2*r+1, 2))]
w = np.searchsorted(h2[-1] - h2[::-1], h2)
q = np.zeros(((r+1), (r+1)), dtype=int)
q[np.arange(r+1), w[::-1]] = 1
q[1:, 0] -= 1
q = np.add.accumulate(q.ravel()).reshape(r+1, r+1)
h = np.c_[q, q[:, -2::-1]]
circle.mask = np.r_[h, h[-2::-1]]

Best way to find all points of lattice in sphere

Given a bunch of arbitrary vectors (stored in a matrix A) and a radius r, I'd like to find all integer-valued linear combinations of those vectors which land inside a sphere of radius r. The necessary coordinates I would then store in a Matrix V. So, for instance, if the linear combination
K=[0; 1; 0]
lands inside my sphere, i.e. something like
if norm(A*K) <= r then
V(:,1)=K
end
etc.
The vectors in A are sure to be the simplest possible basis for the given lattice and the largest vector will have length 1. Not sure if that restricts the vectors in any useful way but I suspect it might. - They won't have as similar directions as a less ideal basis would have.
I tried a few approaches already but none of them seem particularly satisfying. I can't seem to find a nice pattern to traverse the lattice.
My current approach involves starting in the middle (i.e. with the linear combination of all 0s) and go through the necessary coordinates one by one. It involves storing a bunch of extra vectors to keep track of, so I can go through all the octants (in the 3D case) of the coordinates and find them one by one. This implementation seems awfully complex and not very flexible (in particular it doesn't seem to be easily generalizable to arbitrary numbers of dimension - although that isn't strictly necessary for the current purpose, it'd be a nice-to-have)
Is there a nice* way to find all the required points?
(*Ideally both efficient and elegant**. If REALLY necessary, it wouldn't matter THAT much to have a few extra points outside the sphere but preferably not that many more. I definitely do need all the vectors inside the sphere. - if it makes a large difference, I'm most interested in the 3D case.
**I'm pretty sure my current implementation is neither.)
Similar questions I found:
Find all points in sphere of radius r around arbitrary coordinate - this is actually a much more general case than what I'm looking for. I am only dealing with periodic lattices and my sphere is always centered at 0, coinciding with one point on the lattice.
But I don't have a list of points but rather a matrix of vectors with which I can generate all the points.
How to efficiently enumerate all points of sphere in n-dimensional grid - the case for a completely regular hypercubic lattice and the Manhattan-distance. I'm looking for completely arbitary lattices and euclidean distance (or, for efficiency purposes, obviously the square of that).
Offhand, without proving any assertions, I think that 1) if the set of vectors is not of maximal rank then the number of solutions is infinite; 2) if the set is of maximal rank, then the image of the linear transformation generated by the vectors is a subspace (e.g., plane) of the target space, which intersects the sphere in a lower-dimensional sphere; 3) it follows that you can reduce the problem to a 1-1 linear transformation (kxk matrix on a k-dimensional space); 4) since the matrix is invertible, you can "pull back" the sphere to an ellipsoid in the space containing the lattice points, and as a bonus you get a nice geometric description of the ellipsoid (principal axis theorem); 5) your problem now becomes exactly one of determining the lattice points inside the ellipsoid.
The latter problem is related to an old problem (counting the lattice points inside an ellipse) which was considered by Gauss, who derived a good approximation. Determining the lattice points inside an ellipse(oid) is probably not such a tidy problem, but it probably can be reduced one dimension at a time (the cross-section of an ellipsoid and a plane is another ellipsoid).
I found a method that makes me a lot happier for now. There may still be possible improvements, so if you have a better method, or find an error in this code, definitely please share. Though here is what I have for now: (all written in SciLab)
Step 1: Figure out the maximal ranges as defined by a bounding n-parallelotope aligned with the axes of the lattice vectors. Thanks for ElKamina's vague suggestion as well as this reply to another of my questions over on math.se by chappers: https://math.stackexchange.com/a/1230160/49989
function I=findMaxComponents(A,r) //given a matrix A of lattice basis vectors
//and a sphere radius r,
//find the corners of the bounding parallelotope
//built from the lattice, and store it in I.
[dims,vecs]=size(A); //figure out how many vectors there are in A (and, unnecessarily, how long they are)
U=eye(vecs,vecs); //builds matching unit matrix
iATA=pinv(A'*A); //finds the (pseudo-)inverse of A^T A
iAT=pinv(A'); //finds the (pseudo-)inverse of A^T
I=[]; //initializes I as an empty vector
for i=1:vecs //for each lattice vector,
t=r*(iATA*U(:,i))/norm(iAT*U(:,i)) //find the maximum component such that
//it fits in the bounding n-parallelotope
//of a (n-1)-sphere of radius r
I=[I,t(i)]; //and append it to I
end
I=[-I;I]; //also append the minima (by symmetry, the negative maxima)
endfunction
In my question I only asked for a general basis, i.e, for n dimensions, a set of n arbitrary but linearly independent vectors. The above code, by virtue of using the pseudo-inverse, works for matrices of arbitrary shapes and, similarly, Scilab's "A'" returns the conjugate transpose rather than just the transpose of A so it equally should work for complex matrices.
In the last step I put the corresponding minimal components.
For one such A as an example, this gives me the following in Scilab's console:
A =
0.9701425 - 0.2425356 0.
0.2425356 0.4850713 0.7276069
0.2425356 0.7276069 - 0.2425356
r=3;
I=findMaxComponents(A,r)
I =
- 2.9494438 - 3.4186986 - 4.0826424
2.9494438 3.4186986 4.0826424
I=int(I)
I =
- 2. - 3. - 4.
2. 3. 4.
The values found by findMaxComponents are the largest possible coefficients of each lattice vector such that a linear combination with that coefficient exists which still land on the sphere. Since I'm looking for the largest such combinations with integer coefficients, I can safely drop the part after the decimal point to get the maximal plausible integer ranges. So for the given matrix A, I'll have to go from -2 to 2 in the first component, from -3 to 3 in the second and from -4 to 4 in the third and I'm sure to land on all the points inside the sphere (plus superfluous extra points, but importantly definitely every valid point inside) Next up:
Step 2: using the above information, generate all the candidate combinations.
function K=findAllCombinations(I) //takes a matrix of the form produced by
//findMaxComponents() and returns a matrix
//which lists all the integer linear combinations
//in the respective ranges.
v=I(1,:); //starting from the minimal vector
K=[];
next=1; //keeps track of what component to advance next
changed=%F; //keeps track of whether to add the vector to the output
while or(v~=I(2,:)) //as long as not all components of v match all components of the maximum vector
if v <= I(2,:) then //if each current component is smaller than each largest possible component
if ~changed then
K=[K;v]; //store the vector and
end
v(next)=v(next)+1; //advance the component by 1
next=1; //also reset next to 1
changed=%F;
else
v(1:next)=I(1,1:next); //reset all components smaller than or equal to the current one and
next=next+1; //advance the next larger component next time
changed=%T;
end
end
K=[K;I(2,:)]'; //while loop ends a single iteration early so add the maximal vector too
//also transpose K to fit better with the other functions
endfunction
So now that I have that, all that remains is to check whether a given combination actually does lie inside or outside the sphere. All I gotta do for that is:
Step 3: Filter the combinations to find the actually valid lattice points
function points=generatePoints(A,K,r)
possiblePoints=A*K; //explicitly generates all the possible points
points=[];
for i=possiblePoints
if i'*i<=r*r then //filter those that are too far from the origin
points=[points i];
end
end
endfunction
And I get all the combinations that actually do fit inside the sphere of radius r.
For the above example, the output is rather long: Of originally 315 possible points for a sphere of radius 3 I get 163 remaining points.
The first 4 are: (each column is one)
- 0.2425356 0.2425356 1.2126781 - 0.9701425
- 2.4253563 - 2.6678919 - 2.4253563 - 2.4253563
1.6977494 0. 0.2425356 0.4850713
so the remainder of the work is optimization. Presumably some of those loops could be made faster and especially as the number of dimensions goes up, I have to generate an awful lot of points which I have to discard, so maybe there is a better way than taking the bounding n-parallelotope of the n-1-sphere as a starting point.
Let us just represent K as X.
The problem can be represented as:
(a11x1 + a12x2..)^2 + (a21x1 + a22x2..)^2 ... < r^2
(x1,x2,...) will not form a sphere.
This can be done with recursion on dimension--pick a lattice hyperplane direction and index all such hyperplanes that intersect the r-radius ball. The ball intersection of each such hyperplane itself is a ball, in one lower dimension. Repeat. Here's the calling function code in Octave:
function lat_points(lat_bas_mx,rr)
% **globals for hyperplane lattice point recursive function**
clear global; % this seems necessary/important between runs of this function
global MLB;
global NN_hat;
global NN_len;
global INP; % matrix of interior points, each point(vector) a column vector
global ctr; % integer counter, for keeping track of lattice point vectors added
% in the pre-allocated INP matrix; will finish iteration with actual # of points found
ctr = 0; % counts number of ball-interior lattice points found
MLB = lat_bas_mx;
ndim = size(MLB)(1);
% **create hyperplane normal vectors for recursion step**
% given full-rank lattice basis matrix MLB (each vector in lattice basis a column),
% form set of normal vectors between successive, nested lattice hyperplanes;
% store them as columnar unit normal vectors in NN_hat matrix and their lengths in NN_len vector
NN_hat = [];
for jj=1:ndim-1
tmp_mx = MLB(:,jj+1:ndim);
tmp_mx = [NN_hat(:,1:jj-1),tmp_mx];
NN_hat(:,jj) = null(tmp_mx'); % null space of transpose = orthogonal to columns
tmp_len = norm(NN_hat(:,jj));
NN_hat(:,jj) = NN_hat(:,jj)/tmp_len;
NN_len(jj) = dot(MLB(:,jj),NN_hat(:,jj));
if (NN_len(jj)<0) % NN_hat(:,jj) and MLB(:,jj) must have positive dot product
% for cutting hyperplane indexing to work correctly
NN_hat(:,jj) = -NN_hat(:,jj);
NN_len(jj) = -NN_len(jj);
endif
endfor
NN_len(ndim) = norm(MLB(:,ndim));
NN_hat(:,ndim) = MLB(:,ndim)/NN_len(ndim); % the lowest recursion level normal
% is just the last lattice basis vector
% **estimate number of interior lattice points, and pre-allocate memory for INP**
vol_ppl = prod(NN_len); % the volume of the ndim dimensional lattice paralellepiped
% is just the product of the NN_len's (they amount to the nested altitudes
% of hyperplane "paralellepipeds")
vol_bll = exp( (ndim/2)*log(pi) + ndim*log(rr) - gammaln(ndim/2+1) ); % volume of ndim ball, radius rr
est_num_pts = ceil(vol_bll/vol_ppl); % estimated number of lattice points in the ball
err_fac = 1.1; % error factor for memory pre-allocation--assume max of err_fac*est_num_pts columns required in INP
INP = zeros(ndim,ceil(err_fac*est_num_pts));
% **call the (recursive) function**
% for output, global variable INP (matrix of interior points)
% stores each valid lattice point (as a column vector)
clp = zeros(ndim,1); % confirmed lattice point (start at origin)
bpt = zeros(ndim,1); % point at center of ball (initially, at origin)
rd = 1; % initial recursion depth must always be 1
hyp_fun(clp,bpt,rr,ndim,rd);
printf("%i lattice points found\n",ctr);
INP = INP(:,1:ctr); % trim excess zeros from pre-allocation (if any)
endfunction
Regarding the NN_len(jj)*NN_hat(:,jj) vectors--they can be viewed as successive (nested) altitudes in the ndim-dimensional "parallelepiped" formed by the vectors in the lattice basis, MLB. The volume of the lattice basis parallelepiped is just prod(NN_len)--for a quick estimate of the number of interior lattice points, divide the volume of the ndim-ball of radius rr by prod(NN_len). Here's the recursive function code:
function hyp_fun(clp,bpt,rr,ndim,rd)
%{
clp = the lattice point we're entering this lattice hyperplane with
bpt = location of center of ball in this hyperplane
rr = radius of ball
rd = recrusion depth--from 1 to ndim
%}
global MLB;
global NN_hat;
global NN_len;
global INP;
global ctr;
% hyperplane intersection detection step
nml_hat = NN_hat(:,rd);
nh_comp = dot(clp-bpt,nml_hat);
ix_hi = floor((rr-nh_comp)/NN_len(rd));
ix_lo = ceil((-rr-nh_comp)/NN_len(rd));
if (ix_hi<ix_lo)
return % no hyperplane intersections detected w/ ball;
% get out of this recursion level
endif
hp_ix = [ix_lo:ix_hi]; % indices are created wrt the received reference point
hp_ln = length(hp_ix);
% loop through detected hyperplanes (updated)
if (rd<ndim)
bpt_new_mx = bpt*ones(1,hp_ln) + NN_len(rd)*nml_hat*hp_ix; % an ndim by length(hp_ix) matrix
clp_new_mx = clp*ones(1,hp_ln) + MLB(:,rd)*hp_ix; % an ndim by length(hp_ix) matrix
dd_vec = nh_comp + NN_len(rd)*hp_ix; % a length(hp_ix) row vector
rr_new_vec = sqrt(rr^2-dd_vec.^2);
for jj=1:hp_ln
hyp_fun(clp_new_mx(:,jj),bpt_new_mx(:,jj),rr_new_vec(jj),ndim,rd+1);
endfor
else % rd=ndim--so at deepest level of recursion; record the points on the given 1-dim
% "lattice line" that are inside the ball
INP(:,ctr+1:ctr+hp_ln) = clp + MLB(:,rd)*hp_ix;
ctr += hp_ln;
return
endif
endfunction
This has some Octave-y/Matlab-y things in it, but most should be easily understandable; M(:,jj) references column jj of matrix M; the tic ' means take transpose; [A B] concatenates matrices A and B; A=[] declares an empty matrix.
Updated / better optimized from original answer:
"vectorized" the code in the recursive function, to avoid most "for" loops (those slowed it down a factor of ~10; the code now is a bit more difficult to understand though)
pre-allocated memory for the INP matrix-of-interior points (this speeded it up by another order of magnitude; before that, Octave was having to resize the INP matrix for every call to the innermost recursion level--for large matrices/arrays that can really slow things down)
Because this routine was part of a project, I also coded it in Python. From informal testing, the Python version is another 2-3 times faster than this (Octave) version.
For reference, here is the old, much slower code in the original posting of this answer:
% (OLD slower code, using for loops, and constantly resizing
% the INP matrix) loop through detected hyperplanes
if (rd<ndim)
for jj=1:length(hp_ix)
bpt_new = bpt + hp_ix(jj)*NN_len(rd)*nml_hat;
clp_new = clp + hp_ix(jj)*MLB(:,rd);
dd = nh_comp + hp_ix(jj)*NN_len(rd);
rr_new = sqrt(rr^2-dd^2);
hyp_fun(clp_new,bpt_new,rr_new,ndim,rd+1);
endfor
else % rd=ndim--so at deepest level of recursion; record the points on the given 1-dim
% "lattice line" that are inside the ball
for jj=1:length(hp_ix)
clp_new = clp + hp_ix(jj)*MLB(:,rd);
INP = [INP clp_new];
endfor
return
endif

What does it mean to get the (MSE) mean error squared for 2 images?

The MSE is the average of the channel error squared.
What does that mean in comparing two same size images?
For two pictures A, B you take the square of the difference between every pixel in A and the corresponding pixel in B, sum that up and divide it by the number of pixels.
Pseudo code:
sum = 0.0
for(x = 0; x < width;++x){
for(y = 0; y < height; ++y){
difference = (A[x,y] - B[x,y])
sum = sum + difference*difference
}
}
mse = sum /(width*height)
printf("The mean square error is %f\n",mse)
Conceptually, it would be:
1) Start with red channel
2) Compute the difference between each pixel's gray level value in the two image's red channels pixel-by-pixel (redA(0,0)-redB(0,0) etc for all pixel locations.
3) Square the differences of every one of those pixels (redA(0,0)-redB(0,0)^2
4) Compute the sum of the squared difference for all pixels in the red channel
5) Repeat above for the green and blue channels
6) Add the 3 sums together and divide by 3, i.e, (redsum+greensum+bluesum)/3
7) Divide by the area of the image (Width*Height) to form the mean or average, i.e., (redsum+greensum+bluesum)/(3*Width*Height) = MSE
Note that the E in error is synonymous with difference. So it could be called the Mean Squared Difference. Also mean is the same as average. So it could also be called the Average Squared Difference.
You can have a look at following article: http://en.wikipedia.org/wiki/Mean_squared_error#Definition_and_basic_properties. There "Yi" represents the true values and "hat_Yi" represents the values with which we want to compare the true values.
So, in your case you can consider one image as the reference image and the second image as the image whose pixel values you would like to compare with the first one....and you do so by calculating the MSE which tells you "how different/similar is the second image to the first one"
Check out wikipedia for MSE, it's a measure of the difference between each pixel value. Here's a sample implementation
def MSE(img1, img2):
squared_diff = (img1 -img2) ** 2
summed = np.sum(squared_diff)
num_pix = img1.shape[0] * img1.shape[1] #img1 and 2 should have same shape
err = summed / num_pix
return err
Let's us assume you have two points in a 2-dimensional space A(x1,y1) and B(x2,y2), the distance between the two points is calculated as sqrt((x1-x2)^2+(y1-y2)^2). If the the two points are in 3-dimensional space, it can be calculated as sqrt((x1-x2)^2+(y1-y2)^2+(z1-z2)^2). For two points in n-dimensional space, the distance formulae can be extended as sqrt(sumacrossdimensions(valueofAindim-valueofBindim)^2) (since latex is not allowed).
Now, the image with n pixels can be viewed as a point in n-dimensional space. The distance between two images with n pixels can be thoughts as the distance between 2 points in n-dimensional space. This distance is called MSE.

How to implement the arnold transform of image

I need to implement the arnold transformation on colour image as it is part of my project please suggest how to implement it for MxN image
In the classical sense, the continuous Arnold's map is defined on the unit square, and therefore the discrete version is defined on square images:
Consider your image as three NxN matrices; one NxN matrix for each colour channel (assuming you're working with RGB images). Do the following transformation with each of the matrices:
Map the (i,j) element of the input matrix to the ((i + j) mod N, (i + 2j) mod N) element of the output matrix.
It's a concatenation of a vertical and a horizontal shearing transformation, "wrapped around" to the original image rectangle:
(the image is from the corresponding Wikipedia article)
In pseudocode for a single colour channel:
Image arnold(inputImage){
outputImage = Image(inputImage.width, inputImage.height);
for(x = 0; x < inputImage.width; x++){
for(y = 0; y < inputImage.height; y++){
pixel = inputImage[x][y];
outputImage[(2*x + y) mod inputImage.width][(x + y) mod inputImage.height] = pixel;
}
}
return outputImage;
}
(note that conventionally we index matrices by (row,column) and images by (column,row))
So that's for square (NxN) images. What you want (Arnold's map for MxN images, where possibly M != N) is somewhat ill-posed in the sense that it's not clear whether it preserves some interesting properties of Arnold's map. However, if that doesn't bother you, you can generalize the map for MxN images the following way:
Perform a vertical shear with wrap-around in such a way, that the j-th column is circulalry shifted upward by j*M/N (note that this leaves the first column and the last column in-place) *
Perform a horizontal shear with wrap-around in such a way, that the i-th row is circulalry right-shifted by i*N/M (this leaves the first and the last row in-place) *
* : shearing is simply shifting columns/rows
Edit: updated my answer for the generalized MxN case

Largest rectangular sub matrix with the same number

I am trying to come up with a dynamic programming algorithm that finds the largest sub matrix within a matrix that consists of the same number:
example:
{5 5 8}
{5 5 7}
{3 4 1}
Answer : 4 elements due to the matrix
5 5
5 5
This is a question I already answered here (and here, modified version). In both cases the algorithm was applied to binary case (zeros and ones), but the modification for arbitrary numbers is quite easy (but sorry, I keep the images for the binary version of the problem). You can do this very efficiently by two pass linear O(n) time algorithm - n being number of elements. However, this is not a dynamic programming - I think using dynamic programming here would be clumsy and inefficient in the end, because of the difficulties with problem decomposition, as the OP mentioned - unless its a homework - but in that case you can try to impress by this algorithm :-) as there's obviously no faster solution than O(n).
Algorithm (pictures depict binary case):
Say you want to find largest rectangle of free (white) elements.
Here follows the two pass linear O(n) time algorithm (n being number of elemets):
1) in a first pass, go by columns, from bottom to top, and for each element, denote the number of consecutive elements available up to this one:
repeat, until:
Pictures depict the binary case. In case of arbitrary numbers you hold 2 matrices - first with the original numbers and second with the auxiliary numbers that are filled in the image above. You have to check the original matrix and if you find a number different from the previous one, you just start the numbering (in the auxiliary matrix) again from 1.
2) in a second pass you go by rows, holding data structure of potential rectangles, i.e. the rectangles containing current position somewhere at the top edge. See the following picture (current position is red, 3 potential rectangles - purple - height 1, green - height 2 and yellow - height 3):
For each rectangle we keep its height k and its left edge. In other words we keep track of the sums of consecutive numbers that were >= k (i.e. potential rectangles of height k). This data structure can be represented by an array with double linked list linking occupied items, and the array size would be limited by the matrix height.
Pseudocode of 2nd pass (non-binary version with arbitrary numbers):
var m[] // original matrix
var aux[] // auxiliary matrix filled in the 1st pass
var rect[] // array of potential rectangles, indexed by their height
// the occupied items are also linked in double linked list,
// ordered by height
foreach row = 1..N // go by rows
foreach col = 1..M
if (col > 1 AND m[row, col] != m[row, col - 1]) // new number
close_potential_rectangles_higher_than(0); // close all rectangles
height = aux[row, col] // maximal height possible at current position
if (!rect[height]) { // rectangle with height does not exist
create rect[height] // open new rectangle
if (rect[height].next) // rectangle with nearest higher height
// if it exists, start from its left edge
rect[height].left_col = rect[height].next.left_col
else
rect[height].left_col = col;
}
close_potential_rectangles_higher_than(height)
end for // end row
close_potential_rectangles_higher_than(0);
// end of row -> close all rect., supposing col is M+1 now!
end for // end matrix
The function for closing rectangles:
function close_potential_rectangles_higher_than(height)
close_r = rectangle with highest height (last item in dll)
while (close_r.height > height) { // higher? close it
area = close_r.height * (col - close_r.left_col)
if (area > max_area) { // we have maximal rectangle!
max_area = area
max_topleft = [row, close_r.left_col]
max_bottomright = [row + height - 1, col - 1]
}
close_r = close_r.prev
// remove the rectangle close_r from the double linked list
}
end function
This way you can also get all maximum rectangles. So in the end you get:
And what the complexity will be? You see that the function close_potential_rectangles_higher_than is O(1) per closed rectangle. Because for each field we create 1 potential rectangle at the maximum, the total number of potential rectangles ever present in particular row is never higher than the length of the row. Therefore, complexity of this function is O(1) amortized!
So the whole complexity is O(n) where n is number of matrix elements.
A dynamic solution:
Define a new matrix A wich will store in A[i,j] two values: the width and the height of the largest submatrix with the left upper corner at i,j, fill this matrix starting from the bottom right corner, by rows bottom to top. You'll find four cases:
case 1: none of the right or bottom neighbour elements in the original matrix are equal to the current one, i.e: M[i,j] != M[i+1,j] and M[i,j] != M[i,j+1] being M the original matrix, in this case, the value of A[i,j] is 1x1
case 2: the neighbour element to the right is equal to the current one but the bottom one is different, the value of A[i,j].width is A[i+1,j].width+1 and A[i,j].height=1
case 3: the neighbour element to the bottom is equal but the right one is different, A[i,j].width=1, A[i,j].height=A[i,j+1].height+1
case 4: both neighbours are equal: A[i,j].width = min(A[i+1,j].width+1,A[i,j+1].width) and A[i,j].height = min(A[i,j+1]+1,A[i+1,j])
the size of the largest matrix that has the upper left corner at i,j is A[i,j].width*A[i,j].height so you can update the max value found while calculating the A[i,j]
the bottom row and the rightmost column elements are treated as if their neighbours to the bottom and to the right respectively are different
in your example, the resulting matrix A would be:
{2:2 1:2 1:1}
{2:1 1:1 1:1}
{1:1 1:1 1:1}
being w:h width:height
Modification to the above answer:
Define a new matrix A wich will store in A[i,j] two values: the width and the height of the largest submatrix with the left upper corner at i,j, fill this matrix starting from the bottom right corner, by rows bottom to top. You'll find four cases:
case 1: none of the right or bottom neighbour elements in the original matrix are equal to the current one, i.e: M[i,j] != M[i+1,j] and M[i,j] != M[i,j+1] being M the original matrix, in this case, the value of A[i,j] is 1x1
case 2: the neighbour element to the right is equal to the current one but the bottom one is different, the value of A[i,j].width is A[i+1,j].width+1 and A[i,j].height=1
case 3: the neighbour element to the bottom is equal but the right one is different, A[i,j].width=1, A[i,j].height=A[i,j+1].height+1
case 4: both neighbours are equal:
Three rectangles are considered:
1. A[i,j].width=A[i,j+1].width+1; A[i,j].height=1;
A[i,j].height=A[i+1,j].height+1; a[i,j].width=1;
A[i,j].width = min(A[i+1,j].width+1,A[i,j+1].width) and A[i,j].height = min(A[i,j+1]+1,A[i+1,j])
The one with the max area in the above three cases will be considered to represent the rectangle at this position.
The size of the largest matrix that has the upper left corner at i,j is A[i,j].width*A[i,j].height so you can update the max value found while calculating the A[i,j]
the bottom row and the rightmost column elements are treated as if their neighbours to the bottom and to the right respectively are different.
This question is a duplicate. I have tried to flag it as a duplicate. Here is a Python solution, which also returns the position and shape of the largest rectangular submatrix:
#!/usr/bin/env python3
import numpy
s = '''5 5 8
5 5 7
3 4 1'''
nrows = 3
ncols = 3
skip_not = 5
area_max = (0, [])
a = numpy.fromstring(s, dtype=int, sep=' ').reshape(nrows, ncols)
w = numpy.zeros(dtype=int, shape=a.shape)
h = numpy.zeros(dtype=int, shape=a.shape)
for r in range(nrows):
for c in range(ncols):
if not a[r][c] == skip_not:
continue
if r == 0:
h[r][c] = 1
else:
h[r][c] = h[r-1][c]+1
if c == 0:
w[r][c] = 1
else:
w[r][c] = w[r][c-1]+1
minw = w[r][c]
for dh in range(h[r][c]):
minw = min(minw, w[r-dh][c])
area = (dh+1)*minw
if area > area_max[0]:
area_max = (area, [(r, c, dh+1, minw)])
print('area', area_max[0])
for t in area_max[1]:
print('coord and shape', t)
Output:
area 4
coord and shape (1, 1, 2, 2)

Resources