Is there a way that I could prove that the pattern of distribution of white pixels in the two attached images are different?

1 次查看（过去 30 天）

Penny13 2019-4-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/458801-is-there-a-way-that-i-could-prove-that-the-pattern-of-distribution-of-white-pixels-in-the-two-attach

评论： Rodrick 2019-4-27

Hello,

In the two attached binary images, I want to find a way to prove that the pattern of white pixels in the two images are different, so is there any way in MATLAB statistics or digital image processing to approach such a problem?

Thanks for your guidance,

6 个评论
显示 4更早的评论隐藏 4更早的评论

Penny13 2019-4-27

Thank you @Image Analyst. I will look into particle size analysis. I also attached gray level images. The other thing is I want to possibly use SVM for training and testing because I want to find the difference between those two group of the images first.

Penny13 2019-4-27

Thank you @ Walter Roberson. I will look fractal dimention too.

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Image Analyst 2019-4-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/458801-is-there-a-way-that-i-could-prove-that-the-pattern-of-distribution-of-white-pixels-in-the-two-attach#answer_372515

Well there are tons of things you could measure. Area fraction, nearest neighbor distances, area distribution, blob intensities, etc. Literally dozens upon dozens of things.

One approach is to just measure everything you can think of, then use principal components analysis, pca(), to make a model. You could threshold on the first principal component if there are two classes.

Or you could try kmeans() if you know there are two classes.

Or you should try the Classification Learner app to find out which classification/prediction scheme works best given the measurements you have made. Often random forest, though very ad hoc, turns out to be best. Or try several of them and have them "vote" like "adaboost" methods.

But don't discount deep learning - it may end up being the best of them all.

4 个评论
显示 2更早的评论隐藏 2更早的评论

Penny13 2019-4-27

Thanks a lot. I appreciate your guidance.

Rodrick 2019-4-27

Hello Penny13. I am writing a Matlab code for a project (Intelligent Tobacco Curing Control Based on Color Recognition). I am using a Raspberry Pi and Pi Camera to capture images of leaves and then the code has to determine the curing stage. The curing stages are identified by the color of the leaf and there are basically 3 main colors(identifiers). There are two parameters(temperature and humidity) which must then be adjusted to required levels unique to each of the 3 stages.

The colors are: 1. Green (Initial Stage) 2. Yellow (Yellowing Stage) 3. Lemonish Colour (Colour fixing stage). A captured process image will be compared with a dataset in Matlab to recognise the curing stage. After recognising the color and stage MATLAB has to then command the Pi to Switch on a control mechanism or end effector, for the prototype it (the end effector) may just be a small Fan to show control of humidity and temperature adjustments will be shown by switiching on of a LED for each temperature level.

So my question is, what are the resources you recommend for me to read to accomplish my goals and have you done something almost similar before ?

请先登录，再进行评论。