How about labeling manually a dataset for training, then let machine learning do the rest (of course local binary pattern could be used as feature)? By labeling, I mean for each image draw the region where there are grass, and the region where there are cover. On Mon, Nov 6, 2017 at 4:15 AM, Zhenjiang Zhou <zhenjiang.zhou@slu.se> wrote:

Hi there,

I am new python user, trying to use python to solve a question in my research, which is simple but not easy question:

The attached image is mixture of two plants: grass and clover, grass is the one with long leaf, clover is the one with round leaf.

So they have different shape, but very similar color

So the question is how to use python to estimate the percentage of each plant in the whole image (%)

From the literature, local binary pattern might be used, but I am not so sure it can do good job here.

Any help is highly appreciated, I can say that this question is a big issue in my research area.

Looking forwards to your reply.

Zhenjiang Zhou

Postdoc

*Swedish University of Agricultural Sciences*

Department of Agricultural Research for Northern Sweden 901 83 UMEÅ Sweden Visiting address: Skogsmarksgränd Phone:+46 90-786 8731 <+46%2090%20786%2087%2031>

*zhenjiang.zhou@slu.se <zhenjiang.zhou@slu.se> *[image: slu-logo-rgb-web-medium]

_______________________________________________ scikit-image mailing list scikit-image@python.org https://mail.python.org/mailman/listinfo/scikit-image

-- *Dzung Nguyen* PhD Student Electrical Engineering and Computer Science, Northwestern University, IL, USA http://users.eecs.northwestern.edu/~dtn419/