The goal is to identify activities performed by non-employees in real timeWhich solution should the agency consider?
Use a proxy server at each local office and for each camera, and stream the RTSP feed to a unique Amazon Kinesis Video Streams video stream. On each stream, use Amazon Rekognition Video and create a stream processor to detect faces from a collection of known employees, and alert when non-employees are detected.
Use a proxy server at each local office and for each camera, and stream the RTSP feed to a unique Amazon Kinesis Video Streams video stream. On each stream, use Amazon Rekognition Image to detect faces from a collection of known employees and alert when non-employees are detected.
Install AWS DeepLens cameras and use the DeepLens_Kinesis_Video module to stream video to Amazon Kinesis Video Streams for each camera. On each stream, use Amazon Rekognition Video and create a stream processor to detect faces from a collection on each stream, and alert when non-employees are detected.
Install AWS DeepLens cameras and use the DeepLens_Kinesis_Video module to stream video to Amazon Kinesis Video Streams for each camera. On each stream, run an AWS Lambda function to capture image fragments and then call Amazon Rekognition Image to detect faces from a collection of known employees, and alert when non-employees are detected.
Explanations:
This option effectively utilizes Amazon Kinesis Video Streams for real-time streaming of RTSP feeds. By employing Amazon Rekognition Video, it can analyze video streams continuously and detect faces in real time. The stream processor alerts when non-employees are identified, which aligns with the agency’s goal of monitoring non-employee activities in real time.
This option suggests using Amazon Rekognition Image, which is designed for analyzing individual images rather than streaming video. As the requirement is to identify non-employees in real-time, this approach lacks the capability for continuous monitoring and timely alerts during streaming, making it unsuitable for the agency’s needs.
Although this option involves AWS DeepLens and Kinesis Video Streams, it is not the best choice because it lacks direct streaming from video feeds in real time for continuous monitoring. Instead, it processes video streams on a per-frame basis, which may not be optimal for real-time detection of non-employees as required by the agency.
While this option uses AWS DeepLens and Kinesis Video Streams, relying on AWS Lambda to capture image fragments and then analyze them with Amazon Rekognition Image is inefficient for real-time detection. It introduces latency, as image fragments need to be processed sequentially, which is not suitable for immediate alerts when non-employees are detected.
I calculate that the answer is:
Use a proxy server at each local office and for each camera, and stream the RTSP feed to a unique Amazon Kinesis Video Streams video stream. On each stream, use Amazon Rekognition Video and create a stream processor to detect faces from a collection of known employees, and alert when non-employees are detected.