p and q - order of the moments, it is the analogue of the mechanical moments (i.e. of force or inertia). Image moments are very basic properties of image, invariant to rotation, could be used as simple descriptor.
Moment of inertia Ix= I2 = mass *x_coordinate^2.
Image moment Ix2 = intensity * *x_coordinate^2.
Because wo want to get the centroid of the image(a block/patch) by the intensity.
m00:p = q = 0,sum the intensity matrix.
m10:p =1,q = 0,sum of the x-direction.
m01:p = 0,q = 1,sum of the y-direction.
(m10/m00,m01/m00) is the centroid.
For class matching, Hietanen in  compares several
binary descriptors and SIFT with different detectors, including
a dense grid. SIFT preformed better than other descriptors and
dense grid responds very well.
This result leads us to test dense detector for different
It seems that a dense detector is one that uses a dense grid on which to ...
I never personally tried it but using image registration algorithm on Voxel Data you may have some results by trying registration on sub-cluster of voxel.
I would be interested in any feedback about this idea。
I tried PHOT (suggested by @applesoup) using the unofficial implementation given under https://github.com/thinkng but it did not work for these image sets. Maybe one needs to further investigate or tweak the algorithm a bit.
First, if you have sufficient amount of data, I do not believe that one could easily outperform a good deep architecture in the task ...
A possible way to approach the problem would be to extract the maximum values of the discrete circular convolution of each pre-normalized sequence with an NXN matrix formed by the elements of an orthonormal base of the Nx1 sequence space.
The output of this step would be a Nx1 vector that would contain a measure of the similitude of each sequence with the ...