CODEWITHSUNDEEP
pythoncurlflask
SimonFreeman
SimonFreeman

Reputation: 205

Python - Flask - pass image URL instead of file

I got this code using Python 3.6.5 - Flask 0.12.2 exposing a service and receiving images files:

@app.route('/image', methods=['POST'])
def image():
    try:
        image_file = request.files['image']  # get the image

        # Set an image confidence threshold value to limit returned data
        threshold = request.form.get('threshold')
        if threshold is None:
            threshold = 0.5
        else:
            threshold = float(threshold)

        # finally run the image through tensor flow object detection`
        image_object = Image.open(image_file)
        objects = od_ws_api.get_objects(image_object, threshold)
        return objects

    except Exception as e:
        print(e)

When I run the program using this CURL command, everything works fine as expected:

curl -F "[email protected]" http://localhost:5000/image

My goal is to pass using same POST method an image URL instead local file, something like:

curl -F "image=https://i.ytimg.com/vi/aeLgjgoy_kE/maxresdefault.jpg" http://localhost:5000/image

If I do so I receive following error message:

400 Bad Request: The browser (or proxy) sent a request that this server could not understand.
127.0.0.1 - - [10/Apr/2018 08:20:52] "POST /image HTTP/1.1" 200

Shall I use other method or library? Thank you Regs S

I did modify as:

@app.route('/image_url', methods=['POST'])
def image_url():
    try:
        image_url = request.value['image_url']  # get the image URL
        local_filename='c:/tensorflow/temp.jpg'
        local_filename, headers = urllib.request.urlretrieve(image_url)
        # Set an image confidence threshold value to limit returned data
        threshold = request.form.get('threshold')
        if threshold is None:
            threshold = 0.5
        else:
            threshold = float(threshold)

        # finally run the image through tensor flow object detection`
        image_object = Image.open(image_file)
        objects = od_ws_api.get_objects(image_object, threshold)
        return objects

    except Exception as e:
        print(e)
        return 'error'

After tring CURL:

curl -F "image_url=@https://i.ytimg.com/vi/aeLgjgoy_kE/maxresdefault.jpg" http://localhost:5000/image_url

I receive following error:

Warning: setting file https://i.ytimg.com/vi/aeLgjgoy_kE/maxresdefault.jpg
Warning: failed!
curl: (26) read function returned funny value

Upvotes: 1

Views: 6150

Answers (2)

SimonFreeman
SimonFreeman

Reputation: 205

I did modify the method to make it works:

import requests
@app.route('/image_url', methods=['GET'])
def image_url():
    try:
        f = open('c:/tensorflow1/temp.jpg','wb')
        image_url = request.args['image_url']  # get the image URL
        f.write(requests.get(image_url).content)
        f.close()
        # Set an image confidence threshold value to limit returned data
        threshold = request.form.get('threshold')
        if threshold is None:
            threshold = 0.5
        else:
            threshold = float(threshold)

        # finally run the image through tensor flow object detection`
        image_object = Image.open('c:/tensorflow1/temp.jpg')
        objects = od_ws_api.get_objects(image_object, threshold)
        return objects

    except Exception as e:
        print(e)
        return 'error'

So now I've use simple CURL command:

curl http://localhost:5000/image_url?image_url=https://i.ytimg.com/vi/aeLgjgoy_kE/maxresdefault.jpg

and it works! Thx for your help

Upvotes: 0

MrLeeh
MrLeeh

Reputation: 5589

If you use curl with a filename it will not be stored in request.files but in request.values. So to get the image url you need to call

image_url = request.value['image']

Now you will need to download the image, e.g. by using urlretrieve:

import urllib.request
local_filename, headers = urllib.request.urlretrieve(image_url)

The image is now stored in a temporary file that you can access over the local_filename.

Upvotes: 1

Related Questions