“ I have checked the documents of AWS to resolve the issue of deep analysis of image and video using amazon rekognition. So I checked for various solutions and found out that it can be possible using amazon rekognition to get the parameters checked for analysis of image and video. Pricing of amazon rekognition depends on images processed per month and videos processed per month.”
Amazon Rekognition is an image and video analysis service provided by AWS. It is a cloud based software as a service computer vision platform that was launched in 2016. Using Amazon Rekognition, you can easily add video and image analysis to your application. You just have to provide a video or an image to the rekognition API and it will automatically identify people, objects, texts, activities and scenes.
In this post, you will get to know deep analysis of image and video using amazon rekognition. Here I have used a demos section and video demos section under amazon rekognition to have different available options to analyze the image and video as per require output and also can see the metrics available in amazon rekognition with video content in s3 bucket.
Architecture Overview
The architecture diagram shows the overall deployment architecture with data flow, amazon rekognition, cloudwatch and s3 bucket.
Solution overview
The blog post consists of the following phases:
- Try Uploading Image in Multiple Options Under Demos Section in Amazon Rekognition
- Try Uploading Sample Video in Stored Video Analysis Option Under Video Demos Section in Amazon Rekognition
- Output as Metrics to Identify the Successful Calls and Error Counts
Phase 1: Try Uploading Image in Multiple Options Under Demos Section in Amazon Rekognition
- Open the Amazon Rekognition, Under demos section goto label detection then upload any image or image url once the image uploads it will showcase the results, request and response. Similarly various options are available under the demos section like facial analysis which analyzes the face from image, face comparison which compares the face whether its similar as equal or unequal, text in image which identify the text written on the image.
Phase 2: Try Uploading Sample Video in Stored Video Analysis Option Under Video Demos Section in Amazon Rekognition
- Open the Amazon Rekognition, Under video demos section goto stored video analysis then upload a sample video to find out the results of analysis. Also check the streaming video events option to find analysis of live streaming video.
Phase 3: Output as Metrics to Identify the Successful Calls and Error Counts
Clean-up
Delete S3 bucket.
Pricing
I review the pricing and estimated cost of this example.
Cost of Rekognition =
Amazon Rekognition USE1-Group1-ImagesProcessed(Images Processed: Non Free Tier : 0 - 1000000 images processed per month) = 4 Images Processed = $0.00
Amazon Rekognition USE1-Group2-ImagesProcessed(Images Processed: Non Free Tier : 0 - 1000000 images processed per month) = 12 Images Processed = $0.01
Amazon Rekognition USE1-MinsOfArchVideoProcessed(MinsOfArchVideoProcessed: 0.10 USD per minute of Archived Video processed) = 0.264 Minutes Processed = $0.03
Cost of CloudWatch = $0.00
Cost of Simple Storage Class = $0.01
Total Cost = ($0.00 + $0.01 + $0.03 + $0.00 + $0.01) = $0.05
Summary
In this post, I showed “deep analysis of image and video using amazon rekognition”.
For more details on Amazon Rekognition, Checkout Get started Amazon Rekognition, open the Amazon Rekognition console. To learn more, read the Amazon Rekognition documentation.
Thanks for reading!
Connect with me: Linkedin
Top comments (0)