I am using letter_regcog example from OpenCV, it used dataset from UCI which have structure like this:
Attribute Information:
1. lettr capital letter (26 values from A to Z)
2. x-box horizontal position of box (integer)
3. y-box vertical position of box (integer)
4. width width of box (integer)
5. high height of box (integer)
6. onpix total # on pixels (integer)
7. x-bar mean x of on pixels in box (integer)
8. y-bar mean y of on pixels in box (integer)
9. x2bar mean x variance (integer)
10. y2bar mean y variance (integer)
11. xybar mean x y correlation (integer)
12. x2ybr mean of x * x * y (integer)
13. xy2br mean of x * y * y (integer)
14. x-ege mean edge count left to right (integer)
15. xegvy correlation of x-ege with y (integer)
16. y-ege mean edge count bottom to top (integer)
17. yegvx correlation of y-ege with x (integer)
example:
T,2,8,3,5,1,8,13,0,6,6,10,8,0,8,0,8
I,5,12,3,7,2,10,5,5,4,13,3,9,2,8,4,10
now I have segmented image of letter and want to transform it into data like this to put recognize it but I don't understand the mean of all value like "6. onpix total # on pixels" what is it mean ? Can you please explain the mean of these value. thanks.
I am not familiar with OpenCV's letter_recog example, but this appears to be a feature vector, or set of statistics about the image of a letter that is used to classify the future occurrences of the letter. The results of your segmentation should leave you with a binary mask with 1's on the letter and 0's everywhere else. onpix is simply the total count of pixels that fall on the letter, or in other words, the sum of your binary mask.
Most of the rest values in the list need to be calculated based on the set of pixels with a value of 1 in your binary mask. x and y are just the position of the pixel. For instance, x-bar is just the sample mean of all of the x positions of all pixels that have a 1 in the mask. You should be able to easily find references on the web for mathematical definitions of mean, variance, covariance and correlation.
14-17 are a little different since they are based on edge pixels, but the calculations should be similar, just over a different set of pixels.
My name is Antonio Bernal.
In page 3 of this article you will find a good description for each value.
Letter Recognition Using Holland-Style Adaptive Classifiers.
If you have any doubt let me know.
I am trying to make this algorithm work, but my problem is that I do not know how to scale the values to fit them to the range 0-15.
Do you have any idea how to do this?
Another Link from Google scholar -> Letter Recognition Using Holland-Style Adaptive Classifiers
Related
I do not understand the creation of the chees board aruco this is the method:
static Ptr<CharucoBoard> cv::aruco::CharucoBoard::create ( int squaresX,
int squaresY,
float squareLength,
float markerLength,
const Ptr< Dictionary > & dictionary
)
Number of chessboard squares in X direction.
Number of chessboard squares in Y direction.
Length of square side.
Length of marker side.
The dictionary of the markers.
Ids of all the markers.
the first question I have to ask is if I have to choose the length of the squares and markers or do I have to choose according to something? this measure then do I have to check it when I print on sheet?
The dictionary does not understand there are several constants how do I choose the best?
Predefined markers dictionaries/sets Each dictionary indicates the number of bits and the number of markers contained.
and what does the description in the documentation mean?
for example between DICT_4X4_50 and DICT_6X6_1000 what do the bits used for the markers mean? do i need these bits to do some operations later to print the image or to show the image?
I am writing a disparity matching algorithm using block matching, but I am not sure how to find the corresponding pixel values in the secondary image.
Given a square window of some size, what techniques exist to find the corresponding pixels? Do I need to use feature matching algorithms or is there a simpler method, such as summing the pixel values and determining whether they are within some threshold, or perhaps converting the pixel values to binary strings where the values are either greater than or less than the center pixel?
I'm going to assume you're talking about Stereo Disparity, in which case you will likely want to use a simple Sum of Absolute Differences (read that wiki article before you continue here). You should also read this tutorial by Chris McCormick before you read more here.
side note: SAD is not the only method, but it's really common and should solve your problem.
You already have the right idea. Make windows, move windows, sum pixels, find minimums. So I'll give you what I think might help:
To start:
If you have color images, first you will want to convert them to black and white. In python you might use a simple function like this per pixel, where x is a pixel that contains RGB.
def rgb_to_bw(x):
return int(x[0]*0.299 + x[1]*0.587 + x[2]*0.114)
You will want this to be black and white to make the SAD easier to computer. If you're wondering why you don't loose significant information from this, you might be interested in learning what a Bayer Filter is. The Bayer Filter, which is typically RGGB, also explains the multiplication ratios of the Red, Green, and Blue portions of the pixel.
Calculating the SAD:
You already mentioned that you have a window of some size, which is exactly what you want to do. Let's say this window is n x n in size. You would also have some window in your left image WL and some window in your right image WR. The idea is to find the pair that has the smallest SAD.
So, for each left window pixel pl at some location in the window (x,y) you would the absolute value of difference of the right window pixel pr also located at (x,y). you would also want some running value, which is the sum of these absolute differences. In sudo code:
SAD = 0
from x = 0 to n:
from y = 0 to n:
SAD = SAD + absolute_value|pl - pr|
After you calculate the SAD for this pair of windows, WL and WR you will want to "slide" WR to a new location and calculate another SAD. You want to find the pair of WL and WR with the smallest SAD - which you can think of as being the most similar windows. In other words, the WL and WR with the smallest SAD are "matched". When you have the minimum SAD for the current WL you will "slide" WL and repeat.
Disparity is calculated by the distance between the matched WL and WR. For visualization, you can scale this distance to be between 0-255 and output that to another image. I posted 3 images below to show you this.
Typical Results:
Left Image:
Right Image:
Calculated Disparity (from the left image):
you can get test images here: http://vision.middlebury.edu/stereo/data/scenes2003/
According to the HOG process, as described in the paper Histogram of Oriented Gradients for Human Detection (see link below), the contrast normalization step is done after the binning and the weighted vote.
I don't understand something - If I already computed the cells' weighted gradients, how can the normalization of the image's contrast help me now?
As far as I understand, contrast normalization is done on the original image, whereas for computing the gradients, I already computed the X,Y derivatives of the ORIGINAL image. So, if I normalize the contrast and I want it to take effect, I should compute everything again.
Is there something I don't understand well?
Should I normalize the cells' values?
Is the normalization in HOG not about contrast anyway, but is about the histogram values (counts of cells in each bin)?
Link to the paper:
http://lear.inrialpes.fr/people/triggs/pubs/Dalal-cvpr05.pdf
The contrast normalization is achieved by normalization of each block's local histogram.
The whole HOG extraction process is well explained here: http://www.geocities.ws/talh_davidc/#cst_extract
When you normalize the block histogram, you actually normalize the contrast in this block, if your histogram really contains the sum of magnitudes for each direction.
The term "histogram" is confusing here, because you do not count how many pixels has direction k, but instead you sum the magnitudes of such pixels. Thus you can normalize the contrast after computing the block's vector, or even after you computed the whole vector, assuming that you know in which indices in the vector a block starts and a block ends.
The steps of the algorithm due to my understanding - worked for me with 95% success rate:
Define the following parameters (In this example, the parameters are like HOG for Human Detection paper):
A cell size in pixels (e.g. 6x6)
A block size in cells (e.g. 3x3 ==> Means that in pixels it is 18x18)
Block overlapping rate (e.g. 50% ==> Means that both block width and block height in pixels have to be even. It is satisfied in this example, because the cell width and cell height are even (6 pixels), making the block width and height also even)
Detection window size. The size must be dividable by a half of the block size without remainder (so it is possible to exactly place the blocks within with 50% overlapping). For example, the block width is 18 pixels, so the windows width must be a multiplication of 9 (e.g. 9, 18, 27, 36, ...). Same for the window height. In our example, the window width is 63 pixels, and the window height is 126 pixels.
Calculate gradient:
Compute the X difference using convolution with the vector [-1 0 1]
Compute the Y difference using convolution with the transpose of the above vector
Compute the gradient magnitude in each pixel using sqrt(diffX^2 + diffY^2)
Compute the gradient direction in each pixel using atan(diffY / diffX). Note that atan will return values between -90 and 90, while you will probably want the values between 0 and 180. So just flip all the negative values by adding to them +180 degrees. Note that in HOG for Human Detection, they use unsigned directions (between 0 and 180). If you want to use signed directions, you should make a little more effort: If diffX and diffY are positive, your atan value will be between 0 and 90 - leave it as is. If diffX and diffY are negative, again, you'll get the same range of possible values - here, add +180, so the direction is flipped to the other side. If diffX is positive and diffY is negative, you'll get values between -90 and 0 - leave them the same (You can add +360 if you want it positive). If diffY is positive and diffX is negative, you'll again get the same range, so add +180, to flip the direction to the other side.
"Bin" the directions. For example, 9 unsigned bins: 0-20, 20-40, ..., 160-180. You can easily achieve that by dividing each value by 20 and flooring the result. Your new binned directions will be between 0 and 8.
Do for each block separately, using copies of the original matrix (because some blocks are overlapping and we do not want to destroy their data):
Split to cells
For each cell, create a vector with 9 members (one for each bin). For each index in the bin, set the sum of all the magnitudes of all the pixels with that direction. We have totally 6x6 pixels in a cell. So for example, if 2 pixels have direction 0 while the magnitude of the first one is 0.231 and the magnitude of the second one is 0.13, you should write in index 0 in your vector the value 0.361 (= 0.231 + 0.13).
Concatenate all the vectors of all the cells in the block into a large vector. This vector size should of course be NUMBER_OF_BINS * NUMBER_OF_CELLS_IN_BLOCK. In our example, it is 9 * (3 * 3) = 81.
Now, normalize this vector. Use k = sqrt(v[0]^2 + v[1]^2 + ... + v[n]^2 + eps^2) (I used eps = 1). After you computed k, divide each value in the vector by k - thus your vector will be normalized.
Create final vector:
Concatenate all the vectors of all the blocks into 1 large vector. In my example, the size of this vector was 6318
I'm trying to write a GIS, and are using shapefiles from kortforsyningen.dk
I have the problem, that i cant find out what the m (mesure) value of a vertex is.
I know x value is east/west
y is north/south
z is the height, elevation
but m, whats that? In physics, it would be time or 4.th dimention, but none of those fit with the word "mesure"
The Documentation doesn't tell, first time the word is used, it just says "plus a m (mesure) value. (page 10)
EDIT:
when i wrote "The Documentation" i meant the shapefile documentation, this one
http://www.esri.com/library/whitepapers/pdfs/shapefile.pdf
m seems to be any value that you can assign to a point. E.g You measure the temperature at spefic measure points. then x,y contains the geo coordinates, an m the temperature. Then there is the PointZ type whoch contains x,y,z,m: which i undrstand as a 3d point with an assigned measure, e.g temperature or airpressure, etc.
I'm trying to figure out the best way to approach the following:
Say I have a flat representation of the earth. I would like to create a grid that overlays this with each square on the grid corresponding to about 3 square kilometers. Each square would have a unique region id. This grid would just be stored in a database table that would have a region id and then probably the long/lat coordinates of the four corners of the region, right? Any suggestions on how to generate this table easily? I know I would first need to find out the width and height of this "flattened earth" in kms, calculate the number of regions, and then somehow assign the long/lats to each intersection of vertical/horizontal line; however, this sounds like a lot of manual work.
Secondly, once I have that grid table created, I need to design a fxn that takes a long/lat pair and then determines which logical "region" it is in. I'm not sure how to go about this.
Any help would be appreciated.
Thanks.
Assume the Earth is a sphere with radius R = 6371 km.
Start at (lat, long) = (0, 0) deg. Around the equator, 3km corresponds to a change in longitude of
dlong = 3 / (2 * pi * R) * 360
= 0.0269796482 degrees
If we walk around the equator and put a marker every 3km, there will be about (2 * pi * R) / 3 = 13343.3912 of them. "About" because it's your decision how to handle the extra 0.3912.
From (0, 0), we walk North 3 km to (lat, long) (0.0269796482, 0). We will walk around the Earth again on a path that is locally parallel to the first path we walked. Because it is a little closer to the N Pole, the radius of this circle is a bit smaller than that of the first circle we walked. Let's use lower case r for this radius
r = R * cos(lat)
= 6371 * cos(0.0269796482)
= 6 368.68141 km
We calculate dlong again using the smaller radius,
dlong = 3 / (2 * pi * r) * 360
= 0.0269894704 deg
We put down the second set of flags. This time there are about (2 * pi * r) / 3 = 13 338.5352 of them. There were 13,343 before, but now there are 13,338. What's that? five less.
How do we draw a ribbon of squares when there are five less corners in the top line? In fact, as we walked around the Earth, we'd find that we started off with pretty good squares, but that the shape of the regions sheared out into pretty extreme parallelograms.
We need a different strategy that gives us the same number of corners above and below. If the lower boundary (SW-SE) is 3 km long, then the top should be a little shorter, to make a ribbon of trapeziums.
There are many ways to craft a compromise that approximates your ideal square grid. This wikipedia article on map projections that preserve a metric property, links to several dozen such strategies.
The specifics of your app may allow you to simplify things considerably, especially if you don't really need to map the entire globe.
Microsoft has been investing in spatial data types in their SQL Server 2008 offering. It could help you out here. Because it has data types to represent your flattened earth regions, operators to determine when a set of coordinates is inside a geometry, etc. Even if you choose not to use this, consider checking out the following links. The second one in particular has a lot of good background information on the problem and a discussion on some of the industry standard data formats for spatial data.
http://www.microsoft.com/sqlserver/2008/en/us/spatial-data.aspx
http://jasonfollas.com/blog/archive/2008/03/14/sql-server-2008-spatial-data-part-1.aspx
First, Paul is right. Unfortunately the earth is round which really complicates the heck out of this stuff.
I created a grid similar to this for a topographical mapping server many years ago. I just recoreded the coordinates of the upper left coder of each region. I also used UTM coordinates instead of lat/long. If you know that each region covers 3 square kilometers and since UTM is based on meters, it is straight forward to do a range query to discover the right region.
You do realize that because the earth is a sphere that "3 square km" is going to be a different number of degrees near the poles than near the equator, right? And that at the top and bottom of the map your grid squares will actually represent pie-shaped parts of the world, right?
I've done something similar with my database - I've broken it up into quad cells. So what I did was divide the earth into four quarters (-180,-90)-(0,0), (-180,0)-(0,90) and so on. As I added point entities to my database, if the "cell" got more than X entries, I split the cell into 4. That means that in areas of the world with lots of point entities, I have a lot of quad cells, but in other parts of the world I have very few.
My database for the quad tree looks like:
\d areaids;
Table "public.areaids"
Column | Type | Modifiers
--------------+-----------------------------+-----------
areaid | integer | not null
supercededon | timestamp without time zone |
supercedes | integer |
numpoints | integer | not null
rectangle | geometry |
Indexes:
"areaids_pk" PRIMARY KEY, btree (areaid)
"areaids_rect_idx" gist (rectangle)
Check constraints:
"enforce_dims_rectangle" CHECK (ndims(rectangle) = 2)
"enforce_geotype_rectangle" CHECK (geometrytype(rectangle) = 'POLYGON'::text OR rectangle IS NULL)
"enforce_srid_rectangle" CHECK (srid(rectangle) = 4326)
I'm using PostGIS to help find points in a cell. If I look at a cell, I can tell if it's been split because supercededon is not null. I can find its children by looking for ones that have supercedes equal to its id. And I can dig down from top to bottom until I find the ones that cover the area I'm concerned about by looking for ones with supercedeson null and whose rectangle overlaps my area of interest (using the PostGIS '&' operator).
There's no way you'll be able to do this with rectangular cells, but I've just finished an R package dggridR which would make this easy to do using a grid of hexagonal cells. However, the 3km cell requirement might yield so many cells as to overload your machine.
You can use R to generate the grid:
install.packages('devtools')
install.packages('rgdal')
library(devtools)
devools.install_github('r-barnes/dggridR')
library(dggridR)
library(rgdal)
#Construct a discrete global grid (geodesic) with cells of ~3 km^2
dggs <- dgconstruct(area=100000, metric=FALSE, resround='nearest')
#Get a hexagonal grid for the whole earth based on this dggs
grid <- dgearthgrid(dggs,frame=FALSE)
#Save the grid
writeOGR(grid, "grid_3km_cells.kml", "cells", "KML")
The KML file then contains the ids and edge vertex coordinates of every cell.
The grid looks a little like this:
My package is based on Kevin Sahr's DGGRID which can generate this same grid to KML directly, though you'll need to figure out how to compile it yourself.