Which machine learning algorithm should I use to identify these sections?

1 次查看(过去 30 天)
I'm looking to identify the part I marked in red (this part is written with many variations but the structure remains the same, i.e size, orientation may change, shape may change very slightly) in the image below from a set of images:
I can train it using some images and then I want to test it on some unknown cases. For example here are some examples of the same word written in different styles, that is with slight variations.
I am only interested in identifying the part marked in red.
So for example if I cut out my part I am interested in from all of the images in the above link and train the classifier using those. Then when a similar new image is given to the system it should be able to identify id and locate it in the original image? Would that be possible?
The machine learning algorithms that I have so far come across are SVM, SIFT and neural network. I did not want to go into too much detail of each method unless I was sure it was the solution I was looking for.
Which machine learning algorithm should I look into?
  2 个评论
Faraz
Faraz 2014-5-2
I will be using the IFN Databse.
I want to cut out the red part and train the classifier with at least 10 different variations of it. And then see if it can match the test data with the trained data. Is that possible?

请先登录,再进行评论。

采纳的回答

Image Analyst
Image Analyst 2014-5-2
Would this help:
See how other people have done it, and use those successful, published algorithms.
  2 个评论
Faraz
Faraz 2014-5-2
Thank you. Ill look into those, I believe I will have to get an idea from the abstracts as the full paper is available after purchase.
Image Analyst
Image Analyst 2014-5-2
If you're at a university I believe they can get you the paper "for free" as part of (included in) your tuition or employment.

请先登录,再进行评论。

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Image Data Workflows 的更多信息

Community Treasure Hunt

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

Start Hunting!

Translated by