Analysing Images: Detecting Landmarks (etc.)
I have a tremendous love for all things related to engineering - whether it be mechanical, structural, digital, etc. and so when I first came across Google Cloud's Vision API, I was immediately interested.
According to Google's documentation, the Cloud Vision API allows developers to easily integrate vision detection features within applications, including image labeling, face and landmark detection, optical character recognition (OCR), and tagging of explicit content.
Therefore by registering with the Google Cloud Platform and obtaining the necessary credentials, I was able to integrate this API into my website for all of you to try out!
At the start I was only going to implement the "LANDMARK_DETECTION" feature that analyses an uploaded image and returns a result of what location the uploaded photo is from, with a percentage certainty. However, after reading about and testing out the other features - face and emotion, logo, label, optical character recognition and language, safe search, image property detection - I decided they were way too cool to not implement and include as well.
I am still mostly interested in the "LANDMARK_DETECTION" feature because I want to test whether the API is able to pick up as many South African (particularly Johannesburg) features and landmarks compared to other more 'famous' countries (and cities). I have a sneaky suspicion that there will be WAY more North American and European landmarks in the API's database compared to those in Africa. What do you think?
Check out my blogs about this API:
Please try out the API below for yourself!
Feel free to contact me and let
me know if you find anything interesting and what you think about it!
Google Cloud Vision API