US8-ComputerVision Model to detect yellow lane violations

[HusainAdel]

Research open source datasets, find and integrate a model to detect the violations, connect it to the frontend.

[yomnak]

-I 've created an ubuntu EC2 instance.

• git clone https://github.com/aws-samples/amazon-kinesis-video-streams-consumer-library-for-python.git cd amazon-kinesis-video-streams-consumer-library-for-python
• updated the KVS stream parameters in kvs_consumer_library_example_visualization.py

[yomnak]

update :

[yomnak]

update: when the yellow lane is detected, the left of the yellow lane will be cropped. when there is no yellow lane detected, the image will be deleted.

[yomnak]

-update:

• I modified the "kvs_consumer_library_example_object_detection_and_tracking.py " file to detect the license plate of the violated cars . *NOTE: I used dynamodb, S3 and SQS that mohammed and Husain created to test the code
• the result:

[cerashdan]

This only works for static images, next step make it work for a video

[Dalmajed]

Last update:

1- We updated the lambda function to make it take the data processed from s3 and textract and save them in the RDS in the "violations" table, the picture shows an example of violated cars tested and saved their data into RDS to finally process them to the dashboard