**Question/Issue: Couldn’t fix the/edit the code to communicate!
**Project ID: 336309
**Context/Use case: So, I’ve created a model for object detection for my esp32 cam module from edge impulse, Now I want to modify this code for serial communication with my Arduino whether the module detect object or not. But I couldn’t fix the code according to my needs. If anyone kind enough to help!
Hi @Mamun2468,
After calling run_classifier()
your inference results should be stored in an ei_impulse_result_bounding_box_t
struct (documentation for that can be found here). You can see an example of this here.
You need to compare your bb.label
to some threshold (e.g. bb.label > 0.6
) to make sure that your model is confident in the bounding box class. Your bb.label
will correspond to your object’s label. You can then use bb.x
, bb.y
, bb.width
, bb.height
to figure out where the object is in your image.