Name: Google Cloud Vision API
Uploaded: 2025-01-22T12:30:52.357Z
Duration: 2 min 33 s
Description: A cloud-based service offering powerful image analysis capabilities, including labeling, face detection, and OCR.

google-cloud-vision-api

A cloud-based service offering powerful image analysis capabilities, including labeling, face detection, and OCR.

Google Cloud Vision API

Hello everyone. Welcome to coding money. In today's tutorial, we're diving into Google's Vision API to see how AI can help us analyze images. Today we're going to be developing a practical, a simple node.js application that will analyze product images and determine if they're attractive to children. 

I recently posted a video where I share 10 ideas of how you can make money with AI Vision. Check out the video if you haven't already. Whether you're a developer, a suggestion to share AI Agent Store with colleagues, or just curious about AI tools, this video has got you covered. 

Make sure to like, subscribe, and hit the notification bell to stay updated with the latest tech tutorials. Now let's get [Music] started. The very first step is to set up the Google Vision API. Let's do that first. Head over to Google Cloud console console.

Cloud.com create a new project or select an existing one and click on API in Services. Click on library and search for vision API. Click Cloud Vision API click enable. Next go to IM and administration. I am am and admin click service accounts create a new service account name your service account like children product packaging analysis. Click create and continue assign the uh editor role. Click continue this is optional. Click done. 

Click the three dots button and click manage keys and click add key. Click create a new key select Json and click create. It's going to create the service key file and it will automatically download to your computer so remember that our um service account key file has been downloaded into or downloads folder. Next we'll create a new folder we'll call it.

Google Cloud Vision API and we'll go ahead and copy the key file we downloaded from the downloads file. We'll paste it inside this folder. [Music] Now we'll open this folder with Visual Studio code. Npm install at Google Cloud SLV Vision. It's going to download and install the library for you. 

Create a new file index.js. First, we'll import the Google Cloud Vision client library. Then we'll create a client. Here we need to replace this with the service account key file. As you can see, we have already copied that into our project folder and we just need to provide this path. 

The second step is to write the nodejs code to analyze our product images. We'll write a function to analyze an image and determine if it's attractive to Children. Here's the function. As you can see, it's an async.

Function it's got the tri catch block. We perform the label detection over here. Then we perform image properties detection and we're particularly looking for dominant colors here. We'll check for child attractive labels like toy, cartoon, game, animal, fun, colorful. Here we're checking for bright and vibrant colors. 

Here we're deciding if the image is attracted to children. If it has child-friendly labels or is colorful, then we console log the image is likely attractive to children and we return true. Else we'll say the image is not particularly attractive to children and return false. 

In case there's an error, we'll display the error uh with a message saying error analyzing image. The third step is to run and test the program. Now let's test this function with an image we will need to.

Replace this variable with the path to the image file. It could be anything on your local machine. Like locally, it could be somewhere in the project folder. Or you could also replace uh pad to an image on the internet. 

Let's choose this. Right click, open link in a new tab, and we'll get the link to this image. We'll go ahead and replace this with pad to the image. We'll save this file and now we can run this node index.js. 

We're getting an error message. Okay, so the error message so I have two Json json.

Json file which is not correct right. It's it's saying no such file or directory. Is looking for json.

Json which does not exist it has only one extension. So I'll save this file and then I'll try again node index.js. Okay so now it's saying the image is likely attractive to children. 

Uh now let's try with a product image that isn't supposed to be attractive to children. So I'm trying with this. I'll copy the path to the image. Let's replace the path over here. Save this file and run the code again. 

All right so it's saying that image is likely attractive to children and it's not giving me the desired result. It's uh this image I know isn't supposed to be attractive to children. It's not having the brightest of the colors but as you can see the colors are a bit bright here. 

So I want to look for even more brighter colors and I'm also going to adjust some of these um parameters. So here I

Will change the color score from 0.2 to 0.3 and change the pixel fraction from 10% to 20%. So now it was just looking for uh 10% uh of the product image to be brighter colors. Now it's going to look for 20%. 

And so since this picture won't have those really bright color and it's not filling up 20% of the product image then it should give me like the the response that I'm looking for. So let me run the code again. 

All right so this time it's giving me the desired result. It's showing uh attract attracted to children as false which is correct. I this is not the kind of product that packaging or product that is supposed to be attractive to Children. 

I've gone ahead created a GitHub repo at this URL uh just make sure to replace the service account key with your key in index.js. I have

Instructions on how to run this code. And there you have it a step-by-step guide to using Google Vision API for analyzing images. Check out the links in the description for source code and more resources. I have a video where I share 10 ideas of making money with AI Vision API. Check out the video if you haven't already. The link should pop up on the top left corner of the screen. This tool is powerful for marketers, developers, and anyone looking to integrate AI into their workflow.

Note, that there may be slight differences in text and video. Watch video for the original content.

Google Cloud Vision API Video