401 Error When Using FatSecret Image Recognition API – Scope Issue?

[Omni Knight]

Hello, FatSecret team

I’m currently developing a side project focused on nutrition, and I’d like to test it using your company’s API.  

I'm currently unable to use the Image Recognition API and consistently receive a **401 Unauthorized** response.

It appears that the issue may be related to API scopes. On my side, I am requesting the following scopes when generating the access token:

`basic premier image-recognition`

However, requests to the Image Recognition v2 endpoint (`/rest/image-recognition/v2`) still fail with a 401 error, even though the token is successfully issued.

Could you please confirm:

* Whether the **image-recognition** scope needs to be explicitly enabled on your side for our client ID, and
* If there are any additional scopes, permissions, or account-level settings required to access the Image Recognition API?

For reference, the token is being passed correctly as a Bearer token in the Authorization header.

Thank you for your support, and please let us know if you need any additional details from our end.

FYI: Here's the part of the python code.

def get_token():

    if _token_cache["token"] and time.time() < _token_cache["expires_at"]:

        return _token_cache["token"]

   

    try:

        # ADDED 'image-recognition' to the scope list

        payload = {

            "grant_type": "client_credentials",

            "scope": "basic premier image-recognition"

        }

        response = requests.post(TOKEN_URL, auth=(CLIENT_ID, CLIENT_SECRET), data=payload)

       

        if response.status_code != 200:

            print("Token Error:", response.text)

            return None

           

        data = response.json()

        _token_cache["token"] = data["access_token"]

        _token_cache["expires_at"] = time.time() + data["expires_in"] - 60

        return _token_cache["token"]

    except Exception as e:

        print(f"Auth Exception: {e}")

        return None

import base64

RECOGNITION_V2_URL = "https://platform.fatsecret.com/rest/image-recognition/v2"

@app.route('/api/recognize', methods=['POST'])

def recognize_image():

    if 'image' not in request.files:

        return jsonify({"error": "No image uploaded"}), 400

   

    file = request.files['image']

   

    token = get_token()

    if not token:

        # If token failed (likely due to invalid_scope), return that error

        return jsonify({"error": "Authentication failed. Check API Scopes."}), 401

    try:

        # 1. PROCESS IMAGE (Resize to 512x512 to fit 1MB limit and improve AI accuracy)

        img = Image.open(file)

       

        # Convert to RGB (in case of PNG with transparency)

        if img.mode in ("RGBA", "P"):

            img = img.convert("RGB")

           

        # Resize maintaining aspect ratio or force 512x512

        img.thumbnail((512, 512))

       

        # Save to buffer

        buffer = io.BytesIO()

        img.save(buffer, format="JPEG", quality=85)

        buffer.seek(0)

        image_data = buffer.read()

        # 2. ENCODE

        base64_encoded = base64.b64encode(image_data).decode('utf-8')

       

        # 3. SAFETY CHECK

        if len(base64_encoded) > 999982:

            return jsonify({"error": "Image too large even after resize."}), 413

        payload = {

            "image_b64": base64_encoded,

            "include_food_data": True,

            "region": "US",

            "language": "en"

        }

        headers = {

            "Authorization": f"Bearer {token}",

            "Content-Type": "application/json"

        }

        # 4. SEND

        res = requests.post(RECOGNITION_V2_URL, headers=headers, json=payload)

       

        if res.status_code != 200:

            return jsonify(res.json()), res.status_code

           

        return jsonify(res.json())

    except Exception as e:

        print(f"Exception: {str(e)}")

        return jsonify({"error": str(e)}), 500

Best regards,

[seba...@fatsecret.com]

Thanks for reaching out.

- Image Recognition is only available once you have activated the subscription via your API Account Dashboard while a 14 day free trial is available for all developers

- You can leave scopes blank which will then simply revert back to using all available scopes

For any detailed analysis on your API key please contact our support team: https://platform.fatsecret.com/contact

[Omni Knight]

Thank you for your prompt response.

I am considering using the service for business purposes and am willing to accept all terms of service. Could you please confirm if this is possible? Additionally, I would like to inquire about the pricing for unlimited API calls in this case.

Thank you for your assistance.

Best regards,

[seba...@fatsecret.com]

Thanks for following up. Please get in touch with our support team and we will seek to answer all the above and any additional questions:
https://platform.fatsecret.com/contact