Ask Your Question
1

Approach for Red Light Recognition on Android

asked 2013-02-08 17:10:01 -0600

this post is marked as community wiki

This post is a wiki. Anyone with karma >50 is welcome to improve it.

Hi,

I am new to OpenCV and have been developing Android for quite a bit. I am not a good C/C++ programmer as compared to Java. I am not sure which approach should I take to implement this. I have not used NDK before.

I need to help blind people walk straight on a lane crossing, may be using the Red light on the other end as a reference and calculating the deviation from it.

After going through the different approaches i can take, given here http://opencv.org/platforms/android/opencv4android-usage-models.html [BASIC and ADVANCED].

Any Pointers to help me decide will be appreciated.

I am not sure which approach will be more suitable for this project. I am more inclined

edit retag flag offensive close merge delete

Comments

A highly recommended approach is to work on a desktop application while you develop your algorithm. Writing algorithms directly on the phone is like building planes and trying to fly them without engine. As soon as you have a stable, fast, accurate algorithm, tested with pre-recorded videos, you can start porting.

sammy gravatar imagesammy ( 2013-02-09 00:44:54 -0600 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2013-02-08 22:07:03 -0600

this post is marked as community wiki

This post is a wiki. Anyone with karma >50 is welcome to improve it.

The most of OpenCV C++ API is available in Java. So, in most most cases you can use Java only. C++ is used usually for performance optimizations and in case, when you cannot access some algorithms ans 3rd party libraries from Java.

edit flag offensive delete link more

Question Tools

Stats

Asked: 2013-02-08 17:10:01 -0600

Seen: 320 times

Last updated: Feb 08 '13