Open jmorris644 opened 2 years ago
Hi @jmorris644,
I see now that I have not described this on my readme page. The input message is forwarded to the output.
Suppose your image buffer arrives in the input image as msg.payload
and your recognition result is stored (via output field also in msg.payload
. Then your image in the message will be overwritten by your recognition results. But if you store e.g. your recognition results in msg.myresults
, then your msg.payload
should still contain your original input image...
Bart
That makes perfect sense.
Optionally it would be nice to output the cropped vehicle image.
Yes indeed that can be useful. But as a separate node, so it can be reused to crop the image output of other nodes also. Will put it on my todo list...
I am currently using node-red-contrib-image-tools to do the cropping.
Is it possible to also output the incoming buffer as part of the JSON output string? I need to do some follow-on image manipulation once I have the JSON info. For example, I want to crop the original image to just save the bounding box of the vehicle.
Optionally it would be nice to output the cropped vehicle image. This would be similar to what the current Stream SDK does.