Print

Generally, the SEO experts are in perception that the text in images are not read by Google and it does not support as keyword promotions. Yes, this has been the scenario till date and now the things have changed.

Google now changes the trend, the text in the imagse can also be read by google. Interesting is it? It is also definitely aided for the SEO. Further, Google has been the first to make this possible.

Now I’m sure you would like to try it, like how I felt when I first got to know it. I have tried out this as below. It’s very interesting! On January 6th our team published this blog post. It contained one unique word in this image.

screenshots1

I did not add that word to tags, alt text, image names etc..

I then searched that word in google image search and the particular image displayed on my screen. It was quite interesting! I was thinking how it could be possible like how Google scans text in images.

Later I searched that keyword in google no results shown for that keyword. This makes sure that only the image search the word is been identified. 

Google search

In image search the following are the images shown in the search results.

 

Image search

It was astonishing to see this image on google!

I tested it multiple times with different logic and the right images were displayed in Google search perfectly. If you had any doubt, check our blog post test from your end and see this magic.

How it works:

What is Google OCR?

It is all about OCR, but what is Google OCR? The answer will begin with brief information about what is OCR.
OCR, the short term of Optical Character Recognition, it refers to the mechanical or electronic conversion of image-based files into machine-encoded text.

  • Login to your Google Account and go to Google Drive
  • Upload image or PDF into Google Drive by click New>File Upload
  • Check the uploaded scanned PDF or image on the right panel, right click on the scanned PDF or image, open with Google Docs
  • The image or pdf converts to documents.

OCR helps save time by not having to retype the words that someone already typed in the images. Although there are better OCR solutions, this one is Free and requires no software.

The same method is implementing in google image search. Google automatically perform OCR on image-based files to make it searchable

Google can read text using Optical Character Recognition this is one of the Google vision API features.

What is Google Cloud Vision API?

Google Cloud Vision API is an innovative tool that allows developers or enterprises to see a potential in service of image recognition, understand the contents of an image through encapsulation of powerful machine learning models in a format easy to use API REST.

pastedImage

The Prime Functions:

Label / Entity detection identifies the main object in the image by determining its category, for example, animals, transport, etc.

Optical Character Recognition (OCR) saves time, by automatically extracting data from scanned images and then making the data available for electronic processing. It is a subset of image recognition and is widely used as a form of data entry with the input being some sort of printed document or data records such as bank statements, sales invoices, passports, resumes and business cards. The document is either scanned or a picture that is taken and it is up to the program to recognize the characters and give an output in the form of a text document.

Detect and extract the text from an image. It includes support for a wide range of languages, along with support for automatic language identification

Detection of inappropriate content, it can also detect adult or violent content type in an image.

Face Detection, it does detect faces in photos, as well as areas associated with faces like eyes, mouth, nose and last but not least facial expressions and emotions such as sadness or happiness. It does not offer face recognition.

Landscape analyzes to detect natural structures or human-made (sculptures, buildings, paintings, etc.).

Identification of companies logos inside an image.

Tajmahal P7

Try the API upload an image to see the features: https://cloud.google.com/vision/

It should have definitely been interesting and have a great experience experimenting and exploring.