I am using strongest SURFfeatures to identify geese (several hundred per image) in aerial photos. Is there a way to acquire the centroid for each of the clustered SURFfeatures around geese? I want to crop images of geese for BagOfFeatures analysis.

1 visualización (últimos 30 días)
Tony
Tony el 10 de Nov. de 2016
Comentada: Tony el 10 de Nov. de 2016

Iniciar sesión para responder a esta pregunta.

Respuesta aceptada

Image Analyst
Image Analyst el 10 de Nov. de 2016
You need an unsupervised cluster analysis to tell you how many clusters (geese) there are. See http://www.mathworks.com/help/stats/cluster-analysis.html Once you know which cluster each circle belongs to then you can take the average of all the green circle centroids (little green +) to get the overall centroid of all the circles in each goose.

Más respuestas (0)

Categorías

Más información sobre Image Category Classification en Help Center y File Exchange.

Etiquetas

Productos

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by