Python image recognition libraries process the image sent with the api and look at the object, classification, intense colors of the image and whether the image is safe or not.
Files you need to upload to the home path after downloading the home folder
https://github.com/OlafenwaMoses/ImageAI/releases/download/essentials-v5/resnet50_imagenet_tf.2.0.h5 https://github.com/OlafenwaMoses/ImageAI/releases/download/1.0/yolo.h5
python api.py
{
"image": "https://github.com/OlafenwaMoses/ImageAI/blob/master/test-images/11.jpg?raw=true"
}
{
"width": 800,
"height": 534,
"security": {
"nsfw": "0.00035329186",
"sfw": "0.99964666"
},
"color_palette": [
{
"color": "#ECECEB",
"score": 0.56
},
{
"color": "#7F7A7D",
"score": 0.12
},
{
"color": "#C3917D",
"score": 0.11
},
{
"color": "#312B29",
"score": 0.09
},
{
"color": "#CCB698",
"score": 0.05
},
{
"color": "#8E6345",
"score": 0.03
},
{
"color": "#704433",
"score": 0.02
},
{
"color": "#494D5D",
"score": 0.01
}
],
"classification": [
{
"label": "jean",
"score": 14.029912650585175
},
{
"label": "notebook",
"score": 11.251913011074066
},
{
"label": "stretcher",
"score": 8.27689915895462
},
{
"label": "monitor",
"score": 8.269056677818298
},
{
"label": "lab_coat",
"score": 6.923883408308029
}
],
"objectDetection": [
{
"label": "person",
"score": 99.90594983100891
},
{
"label": "person",
"score": 99.98345375061035
},
{
"label": "person",
"score": 99.86221194267273
},
{
"label": "person",
"score": 99.63626861572266
},
{
"label": "laptop",
"score": 99.60939884185791
},
{
"label": "dining table",
"score": 41.51776731014252
},
{
"label": "person",
"score": 98.27415943145752
},
{
"label": "cup",
"score": 99.72758293151855
},
{
"label": "cup",
"score": 98.5138475894928
},
{
"label": "cup",
"score": 88.39868307113647
},
{
"label": "cell phone",
"score": 53.04287672042847
},
{
"label": "keyboard",
"score": 31.74397349357605
},
{
"label": "cell phone",
"score": 79.6405017375946
},
{
"label": "cup",
"score": 98.25559854507446
}
]
}
The code is available under the MIT License.