0
$\begingroup$

Are there Machine learning algorithms which will take input dataset of all geometrical objects shapes as images in gif,jpg,tiff formats & output the geometrical shapes names?

i.e. Geometrical Objects pattern identification.

Examples : Square, Rectangle, Kite, Circle, Sphere, Cone, Rhombus, Ellipse, Cylinder, Pentagon etc

These are examples of potential inputs with corresponding labels

$\endgroup$
1
  • 1
    $\begingroup$ CNN to classify labelled Geometrical Objects images, you can train it if you have the data to do so $\endgroup$
    – Multivac
    Sep 22, 2020 at 20:59

1 Answer 1

1
$\begingroup$

This is quite doable. Just make sure of the following:

  1. Any test data should be one of the shapes you have trained on.
  2. Make sure you have balanced data. (You can introduce bias if required)
  3. To make it more challenging you can create or find data like google street view images for numbers.
  4. Segmentation and Classification, both are possible.

VGG should do just fine, you could also use resnet. The overall model can be fit into an inventory pipeline where you sort the inventory through shapes!

$\endgroup$
3
  • $\begingroup$ Thanks. Will it be more complex to interpret & identify colour filled geometrical shapes objects viz Red Rectangle, Blue Circle, Green Triangle, Yellow Pentagon, Brown Trapezoid etc images? $\endgroup$ Sep 24, 2020 at 2:40
  • 1
    $\begingroup$ It definitely will be more complex but with the right dataset you can easily get a good performing model! $\endgroup$
    – Academic
    Sep 24, 2020 at 6:00
  • 1
    $\begingroup$ You could also make a pipeline where one detects color and other detects shape and you combine the results! $\endgroup$
    – Academic
    Sep 24, 2020 at 6:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.