Object Detection issue [closed]
I am using Haar trainer and classifier to train my database at end it is not not detection these pest, help me how can i detect this insect or can suggest any other method for this ?
i am using Haar cascade in making.I am using 500 positive images and 1300 negatives image and perform 20 stages to create XML file . below are the images i collected
it is working on 2 or 3 images but with the error :(
i follow all the steps but still not detecting :( i can also send xml file if anyone can check it ?
Please provide at least an insight in what you did, how your model performed and some decent samples which you trained with!
i attach the samples above and other related info
Did you cut out the positives or annotated them? How about test samples?
yes by using objectmarkers.exe i crop out these images and save that data into info.txt.
can someone help me?
While using the CascadeClassifier::detectMultiScale, you can play around with the function arguments, particularly scale and size of the object detected. This should weed out small objects such as the leaf stem. Take a look here
PS: The third image from the bottom right in your training set suspiciously resembles the falsely detected leaf stem in the sample test image you've provided, hence the misclassification. Adjusting the minSize argument in detectMultiscale might help fix this issue.
thanks for the help bro but i had not develop the code i am using c++ exes files given here in the link below : https://www.cs.auckland.ac.nz/~m.reza...@v4ven27 i am very helpful if u help me on this :)