Python Tutorial

In this tutorial, we are learning how to get images' emotions using Microsoft Emotion API in Python. In English, emotion is one of the few words that does not have a specific meaning. It is abstract. But almost every decision we make in our lives is driven by emotions. Business research has proven that accurately predicting the emotions which can be a huge factor in business growth, and that is what we will look at today: Reading the emotions.

The idea is to reproduce people's thought processes based on the training data in the form of images of people and videos. Then, try to segment existing emotions into objects. For the review in this section, we will focus on pre-recorded and emotional videos, but the same can be applied to live video streams for real-time analysis.

The computer vision is a part of artificial intelligence used to process visual data. Using machine learning and deep learning models, modern computer systems can process digital images and videos to understand and emotionally identify the video content's characteristics.

All human beings have emotions like happiness, sadness, surprise, neutrality, sorrow, etc. These images' emotions can be extracted using Microsoft Emotion API in Python. It is very easy to use and can be called from the API in the terminal or any language, such as Python or PHP. Microsoft offers a 30-day free trial for making a total request of 30,000.

Face Detection:

In any case, face detection is the first step. Check out the API to see the faces in the image and return a rectangle of their positions. The face detection also returns a unique ID that represents the facet of the data. This is used to recognize or identify faces in subsequent processing.

However, facial recognition can remove facial features such as head shape, age, emotion, beard, glasses, etc. These attributes are general predictions, not actual classifications. When users add themselves to the facial service, certain features help ensure that your app receives accurate facial data. For example, your app might tell users how to get the emotion of the images.

Program Code:

Now, we give the program code for getting images' emotions using Microsoft emotion API in Python. The code is now given below -

# To analyse the emotion of image, we use Python script
# Here, we imported all the necessary libraries
import http.client, urllib.request
import urllib.parse, urllib.error
import base64, sys
import simplejson as json
  
# Here we replace with sub_key which you obtained after registration
sub_key = '34dtyb32fcaf55ertcf78181b6a30rtsv56kl'
  
headers = {
  
    # Request for the headers. And replace the placeholder key with your sub_key  or subscription key.
    'Content-Type': 'application/json',
    'Ocp-Apim-Subscription-Key': subscription_key,
}
  
params = urllib.parse.urlencode({
})
  
# Replace the URL and below with the URL of the image you want to analyze.
url1 = 'HERE ADDED THE IMAGE URL'
body = { 'url': url1 }
newbody = str(body)
  
try:
    # You must use the same region in your REST for the call as you used to obtain your subscription keys.
    # For instance, if you obtained your subscription keys from westcentralus, replace "westus" in the URL below with "westcentralus".
    conn = http.client.HTTPSConnection('westus.api.cognitive.microsoft.com')
    conn.request("POST", "/emotion/v1.0/recognize?%s" % params, newbody, headers)
    response = conn.getresponse()
    data = response.read()
  
    my_parsed = json.loads(data)
    print ("The Response of it is:")
    print (json.dumps(parsed, sort_keys=True, indent=2))
  
    # Here, the emotion of the image will be the maximum value of any emotion obtained from the different scores of each emotion
# Here, we calculate the value
    value = parsed[0]["scores"]

# Here, we calculate the result
    result = max(value, key = value.get)
# Here, we print the output of the result
    print ("\nEmotion output is: ", result)
  
# Close all the connections using this below command
    conn.close()
except Exception as e:
    print(e.args)

Output:

The Response is:
{
  "faceRectangle": {
    "top": 68,
    "left": 191,
    "width": 98,
    "height": 98
  },
  "scores": {
    "anger": 0.001,
    "contempt": 0.001,
    "disgust": 0.002,
    "fear": 0.004,
    "happiness": 0.978,
    "neutral": 0.008,
    "sadness": 0.002,
    "surprise": 0.004
  }
}

Dominant Emotion:  happiness

Conclusion:

This tutorial teaches us how to get the emotions of images using Microsoft Emotion API in Python. All human beings have emotions like happiness, sadness, surprise, neutrality, sorrow, etc. These images' emotions can be extracted using Microsoft Emotion API in Python. Using API, we can easily get the emotion of the images. We also learned a program code for getting the emotion of the images.

Next Topicget_log Driver Method - Selenium Python

← prev next →