EX3 3.) c.)

Moderator: Computer Vision

schwanal
Erstie
Erstie
Beiträge: 20
Registriert: 5. Mai 2010 12:56

EX3 3.) c.)

Beitrag von schwanal »

Dear all,

in Exercise 3 Question 3 Part c.) we should compute a combined gradmag and laplace histogram. How should the histogram looks like? Should we multiply / add the values or just merge them one after another?

Thank u in advance,

Alexander Schwan

DanielR
Mausschubser
Mausschubser
Beiträge: 83
Registriert: 19. Feb 2008 13:15

Re: EX3 3.) c.)

Beitrag von DanielR »

Hi Alex.

When I understand our proposals right, they are unfortunately all not sufficient.

When you calculate the histograms seperately and combine them afterwards, you will only gather the information about occurences of certain Lap and Mag values. What you need (and what you loose through this seperation) is the information, which Lap values occured _together_ with certain Mag values. You will therefore have to compute a joint histogram (which is simply a 2d histogram similar to the rg-histgram that we already implemented in Ex2). Therefore you increment our histogram when a certain pair of Mag and Lap occured in your picture - because of the discretisation you will have to look for a pair of bin indices, of course.

diane
Windoof-User
Windoof-User
Beiträge: 25
Registriert: 21. Jun 2009 17:33

Re: EX3 3.) c.)

Beitrag von diane »

The co-occurrences of Mag values and Lap values (i.e. considering them as couples) are more informative than a simple concatenation of these values.
I indeed expect that you use a 2D histogram (nb_bins X nb_bins) for computing the mag/lap histogram.
As always, at the end of the function, the returned representation is a 1D histogram (which contains here nb_bins^2 bins).

Antworten

Zurück zu „Computer Vision“