The traffic recognition API classifies an image into one of the following
Sparse traffic Dense traffic Accident Fire
With this, from images of live traffic, you can tell if an accident has occured, if their is traffic gridlock or if a vehicle is on fire.
To use this API, you need to set VISION-TRAFFIC=True when starting DeepStack
sudo docker run -e VISION-TRAFFIC=True -v localstorage:/datastore \ -p 80:5000 deepquestai/deepstack
If using the GPU Version, run
sudo docker run --rm --runtime=nvidia -e VISION-TRAFFIC=True -v localstorage:/datastore \ -p 80:5000 deepquestai/deepstack:gpu
Note also that you can have multiple endpoints activated, for example, both traffic and object detection are activated below
sudo docker run -e VISION-TRAFFIC=True -e VISION-DETECTION=True -v localstorage:/datastore \ -p 80:5000 deepquestai/deepstack
Example
const request = require("request")
const fs = require("fs")
image_stream = fs.createReadStream("test-image4.jpg")
var form = {"image":image_stream}
request.post({url:"http://localhost:80/v1/vision/traffic", formData:form},function(err,res,body){
response = JSON.parse(body)
console.log(response)
})
Result
{ success: true, label: 'accident', confidence: 0.9488776 }