Skip to content

Store user input #6

Description

@svjharris

Make sure that user input on the location of eye structures can be set and interpreted properly. Since the code isn't very accurate right now, most pictures will probably be processed by the code but structures will be identified by normal human vision, not computer vision (at least in the first release).

Activity

  1. svjharris commented on Nov 24, 2013

    @svjharris
    ContributorAuthor

    This is being implemented in the resetEyes and resetPupils methods in controller. These methods have yet to be working properly/fully tested. Also resetWhiteDot and resetCrescents need to be made.

    Steven made a good point:

    If the first detection is wrong all the subsequent detections should be redone.
    For example, if the user resets Eye a new Eye object should be made so that findPupil and the methods it calls can be run again
    If they reset pupil, findWhiteDot and findCrescent should be run again.

    We'll need to change the reset code to be in line with Steven's point because now it just blindly resets the data it is given without making any new objects.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions