Introduction
Falcon AI-powered Image Processing services provide 24/7 protection against the risks of user-generated content on your brand channels, detecting undesirable images in near real-time.
The Falcon API is a collection of data interpretation and analysis tools available as a web service that allows you to systematize the process of exploring and categorizing significant numbers of unstructured data.
Common use cases for using Falcon Image processing API include the following:
Version 1: In this version you will be able to categorize your Images or videos into 2 different categories as Safe and Not Safe for work.
Version 2: In V2 you can get more detailed tags/labels for your image/video, Explicit Nudity, Partial Nudity and Sexual Scene.
The Falcon API Flow
Using Falcon API is simple:
  1. 1. Click here if you haven't already and create an account.
  2. 2. Get API access credentials. To get your API credential, To get your API credential, log in to your account and go to the API key page.
  3. 3. Submit your file to the API along with the necessary authentication details.
  4. 4. Receive an immediate response from the API that includes a comprehensive description of what was detected with the scores.
The API response provides all the necessary information in JSON format.
Image Processing
Falcon Image processing makes it simple to enhance image and video analysis to your application. You just provide an image or video to the Falcon Image Processing API, and the service can detect any improper content. Soon we will add other detection services to the API as well including facial analysis. Falcon Image Processing is based on the highly scalable, deep learning technology created by experts to analyze images and videos. It requires no machine learning knowledge to operate.
Content Moderation
Different communities require different rules, and the API makes it simple to establish the most effective ones. The API result enables you to create moderation rules that are tailored to your community and brand. For instance, partial nudity images might be permissible, while explicit nudity might not. Then you can filter out undesirable images based on your business policies using the predicted labels and their confidence scores returned by the API.
The format of the moderation outcome will be influenced by the API selected. For further information on the available APIs and the JSON response format, refer to the APIs Reference.
Head to the Quick start guide Falcon Image Moderation API:
Image Content Moderation, Version 1
Image Content Moderation, Version 2
The APIs accommodate the most common image formats including JPG, PNG, WebP, as well as animated multi-frame images like GIFs. The image must be transmitted as raw bytes, with a maximum size of 10MB for individual images or the entire bulk image group.
Version 1
  • The Version 1 API is designed to simply categorize images in two categories as below:
  • Not Safe for Work (NSFW)
  • Expect porn images, nudity, and genitals to be put into this category.
  • Safe
  • Completely safe images without any sensitive nudity or pornographic scene.
Get Started
Create an account to obtain your own API keys if you haven't done so already.
Quick Setup
Getting started only requires a minimal amount of coding effort. Our goal has been to simplify and speed up the integration process for you.
Endpoint
https://api.nixofalcon.com/ai/image
Parameters
Parameter Type Description
Token String Client API Key
Files Object Object Image to be sent for detection
Code Example

headers = {'Accept': 'application/json', 'Authorization': 'API Key' }

files = {'image': (open('/path/to/image', 'rb'))}

params = {
    'model': 'content-moderation',
    'version': '1'
}

response = requests.post('https://api.nixofalcon.com/ai/image', headers=headers, files=files, data=params)

response.text
/**
* requestAPI
* Request service API
* @param string $token API token generated in the project services
* @param string $filePath full path to images
* @param version $version service version, default to 1
* @param string $type Type of services default to 'im' service
* @return bool
*/

function requestApi(
    string $token,
    string $filePath,
    int $version = 1,
    string $type = "im"
): bool {
    $url = "https://api.nixofalcon.com/ai/" . $type . "/v" . $version;
    $file = curl_file_create($filePath);
    $headers = ["Authorization: Bearer " . $token, "Accept: application/json"];
    $curl = curl_init();
    curl_setopt($curl, CURLOPT_URL, $url);
    curl_setopt($curl, CURLOPT_HTTPHEADER, $headers);
    $data = [
        "image" => $file,
    ];
    curl_setopt($curl, CURLOPT_POSTFIELDS, $data);
    $response = curl_exec($curl);
    if (curl_errno($curl)) {
        echo "CURL ERROR - " . curl_error($curl);
    }
    curl_close($curl);
    return $response;
}
$apiKeys = "YOUR_API_KEY";
$filePath = "image/sample.jpg";
$version = 1;
$serviceType = "im";
$result = requestApi($apiKeys, $filePath, $version, $serviceType);

curl -k -X POST "https://api.nixofalcon.com/ai/image" -H "Authorization:

Bearer API Key" -H "Content-Type:

multipart/form-data" -H "Accept: application/json" -F "image=@/path/to/image.jpg"
Response
The response JSON will contain a summary object, a confidence score for each label, and the moderation decision based on your configuration.
  
Version 2
This version of Falcon Image Moderation enables you to detect inappropriate content in milliseconds and allows you to filter the images based on your application requirements. The API provides a list of labels with confidence scores to enable fine-grained control over images you want to allow.
This API uses a two-level hierarchical taxonomy to label categories of inappropriate content. Each top-level category has several second-level categories that allow you more flexibility.
Labels (V2 API):
  • Sexual Scene
  • Pornographic Scene
  • Explicit Nudity
  • Bare Chested Male
  • Graphic Male Nudity
  • Breast
  • Graphic Female Nudity
  • Partial Nudity
  • Bikini or Lingerie
  • Revealing Clothes
  • Mini Skirt or Short
  • Male Swimwear or Underwear
Get Started
Create an account to obtain your own API keys if you haven't done so already.
Quick Setup
Getting started only requires a minimal amount of coding effort. Our goal has been to simplify and speed up the integration process for you.
Endpoint
https://api.nixofalcon.com/ai/image
Parameters
Parameter Type Description
Token String Client API Key
Files Object Object Image to be sent for detection
Configuration
Image moderation Configuration simplifies the process of customizing rules and actions, accessible from your online dashboard. You have control over the analysis of each image and the criteria that determine acceptance or rejection.
To personalize the process flow, go to the Configuration section in your dashboard. There, you can choose the regulations that will be in place for each label and determine what steps should be taken as a result of those regulations - ACCEPT, REVIEW, or REJECT.

configuration.png
Note that once your configuration’s change has been submitted, it will apply immediately and reflect in the API response (Remark element).
Code Example

headers = {'Accept': 'application/json', 'Authorization': 'API Key' }

files = {'image': (open('/path/to/image', 'rb'))}

params = {
    'model': 'content-moderation',
    'version': '2'
}

response = requests.post('https://api.nixofalcon.com/ai/image', headers=headers, files=files, data=params)

response.text
/**
* requestAPI
* Request service API
* @param string $token API token generated in the project services
* @param string $filePath full path to images
* @param version $version service version, default to 2
* @param string $type Type of services default to 'im' service
* @return bool
*/

function requestApi(
    string $token,
    string $filePath,
    int $version = 2,
    string $type = "im"
): bool {
    $url = "https://api.nixofalcon.com/ai/" . $type . "/v" . $version;
    $file = curl_file_create($filePath);
    $headers = ["Authorization: Bearer " . $token, "Accept: application/json"];
    $curl = curl_init();
    curl_setopt($curl, CURLOPT_URL, $url);
    curl_setopt($curl, CURLOPT_HTTPHEADER, $headers);
    $data = [
        "image" => $file,
    ];
    curl_setopt($curl, CURLOPT_POSTFIELDS, $data);
    $response = curl_exec($curl);
    if (curl_errno($curl)) {
        echo "CURL ERROR - " . curl_error($curl);
    }
    curl_close($curl);
    return $response;
}
$apiKeys = "YOUR_API_KEY";
$filePath = "image/sample.jpg";
$version = 2;
$serviceType = "im";
$result = requestApi($apiKeys, $filePath, $version, $serviceType);

curl -k -X POST "https://api.nixofalcon.com/ai/image" -H "Authorization:

Bearer API Key" -H "Content-Type:

multipart/form-data" -H "Accept: application/json" -F "image=@/path/to/image.jpg"
Response
The response JSON will contain a summary object, a confidence score for each label, and the moderation decision based on your configuration.
  
Bulk Image Moderation
However, the easiest way to use the Image Moderation API is by submitting images individually, in some cases, it might be more efficient to submit multiple images at the same time.
Our API accepts that you submit multiple images at the same time within a single request.
Limitation
There is no limit on the number of images, however, the combined size of all images must not exceed 10 MB.
Code Example
Moderate three images through direct upload in one request.
headers = {'Accept': 'application/json', 'Authorization': 'API Key'}

files = {
  ('image[]': open('/path/to/image_01.jpg', 'rb')),
  ('image[]': open('/path/to/image_02.jpg', 'rb')),
  ('image[]': open('/path/to/image_03.jpg', 'rb'))
}

params = {
    'model': 'content-moderation',
    'version': '1'
}

response = requests.post('https://api.nixofalcon.com/ai/image', headers=headers, files=files, data=params)

response.text

/**
* requestAPI
* Request service API
* @param string $token API token generated in the project services
* @param string $filePath full path to images
* @param version $version service version, default to 1
* @param string $type Type of services default to 'im' service
* @return bool
*/

function requestApi(
    string $token,
    string $filePaths,
    int $version = 1,
    string $type = "im"
): bool {
    $url = "https://api.nixofalcon.com/ai/" . $type . "/v" . $version;
    $data = [];
    foreach ($filePaths as $key => $filePath) {
	    $data["image[$key]"] = curl_file_create($filePath);
    }
    $headers = ["Authorization: Bearer " . $token, "Accept: application/json"];
    $curl = curl_init();
    curl_setopt($curl, CURLOPT_URL, $url);
    curl_setopt($curl, CURLOPT_HTTPHEADER, $headers);
    curl_setopt($curl, CURLOPT_POSTFIELDS, $data);
    $response = curl_exec($curl);
    if (curl_errno($curl)) {
        echo "CURL ERROR - " . curl_error($curl);
    }
    curl_close($curl);
    return $response;
}
$apiKeys = "YOUR_API_KEY";
$filePaths = ["path/to/image_01.jpg", "path/to/image_02.jpg", "path/to/image_03.jpg"];
$version = 1;
$serviceType = "im";
$result = requestApi($apiKeys, $filePaths, $version, $serviceType);

curl -k -X POST "https://api.nixofalcon.com/ai/image" -H "Authorization:

Bearer API Key" -H "Content-Type:

multipart/form-data" -H "Accept: application/json"
    -F "image[]=@/path/to/image_01.jpg"
    -F "image[]=@/path/to/image_02.jpg"
    -F "image[]=@/path/to/image_03.jpg"
Response
The API will provide a JSON response where the moderation outcomes for each image will be enclosed in separate fields. Using the "image" field within each object is crucial to link each image to its respective moderation result as the order cannot be guaranteed.