Loading ...
Sorry, an error occurred while loading the content.
 

What features can i use for shape recognition ?

Expand Messages
  • compueng_asabrin
    Hello! I have doing project, and I have a problem. I am working in pattern classification project to classify circle , rectangle and triangle, and I use some
    Message 1 of 3 , Nov 9, 2007
      Hello!

      I have doing project, and I have a problem.
      I am working in pattern classification project to classify circle ,
      rectangle and triangle, and I use some feature such as compactness,
      roundness, eccentricity and no. of point of convex hull , but it not
      give me needed accuracy, so I looking for another features.
      Can somebody point me to some feature I can extract it?
    • Bart van Delft
      ... Hough Transformation maybe? http://en.wikipedia.org/wiki/Hough_transform Or are you more looking for things to parse to a neural network?
      Message 2 of 3 , Nov 9, 2007
        --- In artificialintelligencegroup@yahoogroups.com, "compueng_asabrin"
        <compueng_asabrin@...> wrote:
        >
        > Hello!
        >
        > I have doing project, and I have a problem.
        > I am working in pattern classification project to classify circle ,
        > rectangle and triangle, and I use some feature such as compactness,
        > roundness, eccentricity and no. of point of convex hull , but it not
        > give me needed accuracy, so I looking for another features.
        > Can somebody point me to some feature I can extract it?
        >

        Hough Transformation maybe? http://en.wikipedia.org/wiki/Hough_transform
        Or are you more looking for things to parse to a neural network?
      • Mohammad Tazaree
        Hello 1. each transform domain(such as fourier and wavelet) has features that don t exist in spacial domain. you can use the transfrom domain features. 2.
        Message 3 of 3 , Nov 9, 2007
          Hello
          1. each transform domain(such as fourier and wavelet) has features that don't exist in spacial domain. you can use the transfrom domain features.

          2. SHAPE MOMENT can help you in shape recognition.

          3. however, you may change your ALGORITHM. a bad algorithm, with good features can't help you.
          I think you can use a good classifier say SVM to recognize the shape.

          4. don't remember! shape recognition is a research topic and researches work on it! you don't expect complete this topic in one night

          Bests

          compueng_asabrin <compueng_asabrin@...> wrote:
          Hello!

          I have doing project, and I have a problem.
          I am working in pattern classification project to classify circle ,
          rectangle and triangle, and I use some feature such as compactness,
          roundness, eccentricity and no. of point of convex hull , but it not
          give me needed accuracy, so I looking for another features.
          Can somebody point me to some feature I can extract it?





          Send instant messages to your online friends http://uk.messenger.yahoo.com

          [Non-text portions of this message have been removed]
        Your message has been successfully submitted and would be delivered to recipients shortly.