https://github.com/timatooth/catscanface

Detect motion and cat faces with OpenCV 😺 🔎
https://github.com/timatooth/catscanface

aws-lambda cats mice opencv raspberry-pi

Last synced: 9 months ago
JSON representation

Detect motion and cat faces with OpenCV 😺 🔎

• Host: GitHub
• URL: https://github.com/timatooth/catscanface
• Owner: timatooth
• License: mit
• Created: 2016-11-05T06:07:01.000Z (about 9 years ago)
• Default Branch: master
• Last Pushed: 2017-05-19T12:05:33.000Z (over 8 years ago)
• Last Synced: 2025-04-01T07:26:05.120Z (10 months ago)
• Topics: aws-lambda, cats, mice, opencv, raspberry-pi
• Language: Python
• Homepage:
• Size: 60.5 KB
• Stars: 18
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

          
# catscanface 😺

> Detect motion and cat faces with Raspberry Pi, OpenCV & AWS Lambda

![Thug cat detected with OpenCV](https://i.imgur.com/RDmYoce.png)

Contains Raspberry Pi agent for uploading motion detected frames to S3 which triggers a Python lambda function

for doing further cat Haar Cascade facial processing.

Makes optional use of Datadog for statsd metrics & Raspberry Pi Sense Hat.

#### Blog post

http://timatooth.com/development/2017/01/01/raspberry-pi-motion-detection-aws-lambda-opencv.html

## Raspberry Pi Agent usage

Configuration can be set on the command line or via environment variables. Usage of S3 requires AWS client ID & secrets

to be configured.

    usage: agent.py [-h] [--resolution RESOLUTION] [--fps FPS]

                    [--delta-threshold DELTA_THRESHOLD] [--min-area MIN_AREA]

                    [--enable-sensehat] [--enable-statsd] [--enable-annotate]

                    [--image-path IMAGE_PATH] [--enable-s3]

                    [--s3-bucket S3_BUCKET] [--s3-prefix S3_PREFIX]

    Motion detect and upload frames to S3

    optional arguments:

      -h, --help            show this help message and exit

      --resolution RESOLUTION

                            e.g 640x480

      --fps FPS             Framerate e.g: 18

      --delta-threshold DELTA_THRESHOLD

      --min-area MIN_AREA

      --enable-sensehat     Use Sense Hat display

      --enable-statsd       Send metrics

      --enable-annotate     Draw detected regions to image

      --image-path IMAGE_PATH

                            Where to save images locally eg /tmp

      --enable-s3           Enable saving frames to AWS S3

      --s3-bucket S3_BUCKET

                            AWS S3 bucket to save frames

      --s3-prefix S3_PREFIX

                            AWS S3 bucket prefix path e.g cats/

#### Requirements

OpenCV on Raspberry Pi

```apt-get install libopencv-dev python-opencv python-dev```

pip dependencies for AWS & Datadog

```pip install boto3 dogstatsd```

## AWS Lambda Function for Cat Haar-Cascade facial detection:

#### CloudFormation Stack

    aws cloudformation create-stack --stack-name catscanface --template-body file://cloudformation.yaml --capabilities CAPABILITY_IAM --parameters ParameterKey=S3BucketName,ParameterValue=timatooth

#### Lambda Deployment

This will download [OpenCV compiled on Amazon Linux](https://github.com/aeddi/aws-lambda-python-opencv) and update the AWS Lambda function code.

    ./deploy-lambda.sh