Deep learning result in json format

[{"id":"22700825.3f9d38","type":"tab","label":"Flow 5","disabled":false,"info":""},{"id":"98922031.33529","type":"cocossd","z":"22700825.3f9d38","name":"","x":390,"y":200,"wires":[["f13b7b29.1af548","54a7b73a.b97128"]]},{"id":"d442ac3a.97611","type":"debug","z":"22700825.3f9d38","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"false","statusVal":"","statusType":"auto","x":870,"y":180,"wires":[]},{"id":"705149ce.c18248","type":"fileinject","z":"22700825.3f9d38","name":"","x":130,"y":200,"wires":[["98922031.33529","d26ab02d.79838"]]},{"id":"d26ab02d.79838","type":"image","z":"22700825.3f9d38","name":"","width":160,"data":"annotatedinput","dataType":"msg","thumbnail":false,"active":true,"pass":false,"outputs":0,"x":340,"y":300,"wires":[]},{"id":"f13b7b29.1af548","type":"image","z":"22700825.3f9d38","name":"","width":160,"data":"annotatedinput","dataType":"msg","thumbnail":false,"active":true,"pass":false,"outputs":0,"x":600,"y":300,"wires":[]},{"id":"54a7b73a.b97128","type":"json","z":"22700825.3f9d38","name":"","property":"payload","action":"","pretty":false,"x":650,"y":180,"wires":[["d442ac3a.97611"]]}]

So as you can see from there, I want to get object detection output result in json format, but I don't know how to do that as there is sth missing I guess.

Can you show the debug output of the "cocossd" node ? It is not a string that can be converted to json.

@MK1992 Perhaps it is the image (or lack thereof based on your screenshot) that is throwing the error.

The JSON node works as expected for me, but I am using a different contrib of tf coco.

[
	{
		"bbox": [
			189.49939727783204,
			170.23276805877686,
			1080.4227828979493,
			540.3283023834229
		],
		"class": "car",
		"score": 0.5844805836677551
	}
]

Hi Gunner, yeah I try to install tf.model but I have some installation errors. Could you help me check? I just want to load my model.json into node that supports my issue.(laoding json model) .

I just started here... But as I understand it, you shouldn't have more than one TF contrib installed due conflicts?

To try to keep all of the threads from @MK1992 together, here is the one regarding the install problem: Tf.model node error - #3 by knolleary

The issue being he has installed things in the wrong location.

2 Likes

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.