How to train more than one positive image

Hello everyone, I’m new here.
I just wanna know how to train more than one positive image, so any idea? I scaned all the internet but I didn’t get a clear idea

can you elaborate on that

train what on more than one positive image ? please explain

opencv_traincascade :slight_smile:
i wanna train more than positive image

explain why you think that

what information do you work with, what guides?

ok look, i might didnt explain it right.
i have a watch image to train, So I wanna train more watches images in the same cascade file, without keep doing opencv_traincascade for each image in my folder.

is it clear yet?

opencv_createsamples -img {here i wanna use 5 imgs for ex} -bg bg.txt -info info/info.lst -pngoutput info -maxxangle 0.5 -maxyangle -0.5 -maxzangle 0.5 -num 2000

please read the tutorial

for many train images, put your image file names and rectangles into the info file, and do NOT use -img or -num 2000

then, get more positive images (like 500).
you can neither successfully train it on 5 images, nor does it make sense to augment your dataset from 5 to 2000