2017-06-18 06:25:06 -0600 | received badge | ● Famous Question (source) |
2016-04-04 12:09:54 -0600 | received badge | ● Student (source) |
2015-11-11 06:36:53 -0600 | received badge | ● Notable Question (source) |
2015-05-21 05:34:13 -0600 | received badge | ● Popular Question (source) |
2014-04-05 13:12:27 -0600 | asked a question | Haar cascade vs Lbp cascade vs Hog Cascade in object detection I am doing an project about recognizing one kind of leaf. Well I am using opencv c++. I have read about using 3 different features extraction but I do not know when I can use haar, lbp or hog. I want to use my program in a embedded system so the thing is what type of feature exctraction is better in emebedded system? and why? |
2014-02-17 15:11:37 -0600 | asked a question | Haar cascade using grayscale images without background Helo, I have a question about haar cascade using emgucv and visual studio 2010. I want a lot of positive images and negative images and It does not matter the color of my object that I want to recgonize so for that reason I want to convert positive images and negative images to grayscale to training my haar cascade and I want to eliminate background of positives images and leave only object that I want to recognize to trainning my haar cascade too. Is that a good idea?? |
2014-02-14 19:25:31 -0600 | asked a question | Detecting leaves of trees using haar classifier Well hello everybody, my project is to do a sotfware that detect one kind of leaf call saman using kinect. Leaves have a size of 5 cm or less, my question is ¿Can haar classifier detect that leaves with those size? and if the answer is yes. if I decide to use haar clasifiers and I must take a lot of photo of object that I want to recognize but If I show one object that is very similiar 70% or 80% of the photos that I gave to training the machine, the machine recognize it is a leaf or is mandatory that leaf that I show to machine must be exactly of photos that I gave to training the machine |