kelvins / lbph

Local Binary Patterns Histograms (LBPH) implementation in Go
https://goo.gl/HkovdB
MIT License
53 stars 13 forks source link

concatenate all histograms and create a new one which will be used to represent the image. #2

Closed tuji-sjp closed 6 years ago

tuji-sjp commented 6 years ago

Please tell me how to do this work?How to concatenate in detail?Thank you!

kelvins commented 6 years ago

@tuji-sjp I could not understand your question. This is a step of the algorithm, we don't need to do it manually. To use the algorithm, please follow this example.

To know more about the algorithm steps, please read this article.

Just to answer your question, the concatenation is quite simple:

Histogram 1: [1, 2, 3, 4] Histogram 2: [3, 6, 7, 8] Histogram 3: [4, 2, 7, 8]

Concatenated histogram: [1, 2, 3, 4, 3, 6, 7, 8, 4, 2, 7, 8]

tuji-sjp commented 6 years ago

@kelvins Thank you very much! This algorithm is almost clear to me. But I haven't learned the R. Could you tell me the version of python?

kelvins commented 6 years ago

@tuji-sjp this algorithm is implemented in Go, not Python. If you want to implement this algorithm in an easier language I would suggest you implement using Matlab.

tuji-sjp commented 6 years ago

@kelvins Please excuse my ignorance. Why should I use Matlab instead of python?

kelvins commented 6 years ago

@tuji-sjp It all depends on your purpose, but in general, Matlab provide some easier functions for example to load images, work with matrices, apply statistical functions, and so on. If you are implementing it for academic purposes I would suggest to use Matlab or R. Otherwise, if you are implementing it for commercial use I would suggest to use Python, G, C++, amongst other languages.

tuji-sjp commented 6 years ago

@kelvins Ok, thank you!

kelvins commented 6 years ago

You're welcome!