Pi Camera Connect

Library to capture and stream Raspberry Pi camera data directly to NodeJS
Alternatives To Pi Camera Connect
Project NameStarsDownloadsRepos Using ThisPackages Using ThisMost Recent CommitTotal ReleasesLatest ReleaseOpen IssuesLicenseLanguage
Quaggajs4,793126443 months ago41June 07, 2017227mitJavaScript
An advanced barcode-scanner written in JavaScript
Restreamer2,792
5 months ago47apache-2.0HTML
The Restreamer is a complete streaming server solution for self-hosting. It has a visually appealing user interface and no ongoing license costs. Upload your live stream to YouTube, Twitch, Facebook, Vimeo, or other streaming solutions like Wowza. Receive video data from OBS and publish it with the RTMP and SRT server.
React Webcam1,4722512322 months ago65June 07, 202346mitTypeScript
Webcam component
Viseron1,142
7 days ago75mitPython
Self-hosted, local only NVR and AI Computer Vision software. With features such as object detection, motion detection, face recognition and more, it gives you the power to keep an eye on your home, office or any other place you want to monitor.
Ascii Camera1,074
4 years ago1mitJavaScript
Real-time ASCII representation of your webcam video stream, using the getUserMedia API.
Homebridge Camera Ffmpeg1,0237257 months ago202December 28, 202136apache-2.0TypeScript
Homebridge Plugin Providing FFmpeg-based Camera Support
Moonfire Nvr937
2 months ago64otherRust
Moonfire NVR, a security camera network video recorder
Hkcam817
5 months ago4February 16, 202227apache-2.0Go
Open-Source HomeKit Surveillance Camera
Neolink747
2 months ago83agpl-3.0Rust
An RTSP bridge to Reolink IP cameras
Hap Python5487352 months ago54June 28, 202278otherPython
A python implementation of the HomeKit Accessory Protocol (HAP)
Alternatives To Pi Camera Connect
Select To Compare


Alternative Project Comparisons
Readme

GitHub license npm

Pi Camera Connect - for NodeJS

pi-camera-connect is a library to capture and stream Raspberry Pi camera data directly to NodeJS.

Why use this?

There are many NPM modules for connecting to the Raspberry Pi camera, why use this?

  • Speed: JPEG images can be captured in ~33ms using a built in MJPEG parser
  • Efficient: Pictures and video streams are piped directly into Node as a Buffer, keeping all data in memory and eliminating disk I/O
  • Usable: Video streams are available as stream.Readable objects that can be piped or listened to
  • Tested: Contains automated tests using Jest
  • Modern: Uses the latest ESNext features and up to date development practices
  • Structure: Ships with TypeScript definition files

Install

NPM

$ npm install --save pi-camera-connect

Yarn

$ yarn add pi-camera-connect

Basic Usage

ESNext Syntax

Image capture:

import { StillCamera } from 'pi-camera-connect';
import * as fs from 'fs';

// Take still image and save to disk
const runApp = async () => {
  const stillCamera = new StillCamera();

  const image = await stillCamera.takeImage();

  fs.writeFileSync('still-image.jpg', image);
};

runApp();

Video capture:

import { StreamCamera, Codec } from 'pi-camera-connect';
import * as fs from 'fs';

// Capture 5 seconds of H264 video and save to disk
const runApp = async () => {
  const streamCamera = new StreamCamera({
    codec: Codec.H264,
  });

  const videoStream = streamCamera.createStream();

  const writeStream = fs.createWriteStream('video-stream.h264');

  videoStream.pipe(writeStream);

  await streamCamera.startCapture();

  await new Promise(resolve => setTimeout(() => resolve(), 5000));

  await streamCamera.stopCapture();
};

runApp();

Compatible Syntax

Image capture:

const { StillCamera } = require('pi-camera-connect');

const stillCamera = new StillCamera();

stillCamera.takeImage().then(image => {
  fs.writeFileSync('still-image.jpg', image);
});

Video capture:

const { StreamCamera, Codec } = require('pi-camera-connect');

const streamCamera = new StreamCamera({
  codec: Codec.H264,
});

const writeStream = fs.createWriteStream('video-stream.h264');

const videoStream = streamCamera.createStream();

videoStream.pipe(writeStream);

streamCamera.startCapture().then(() => {
  setTimeout(() => streamCamera.stopCapture(), 5000);
});

Capturing an image

There are 2 ways to capture an image with pi-camera-connect:

  1. StillCamera.takeImage() - Slow, but higher quality

    This is equivalent to running the raspistill command. Under the hood, the GPU will run a strong noise reduction algorithm to make the image appear higher quality.

    import { StillCamera } from 'pi-camera-connect';
    
    const runApp = async () => {
      const stillCamera = new StillCamera();
    
      const image = await stillCamera.takeImage();
    
      // Process image...
    };
    
    runApp();
    
  2. StreamCamera.takeImage() - Fast, but lower quality

    This works by grabbing a single JPEG frame from a Motion JPEG (MJPEG) video stream . Images captured from the video port tend to have a grainier appearance due to the lack of a strong noise reduction algorithm.

    Using this method, you can capture a JPEG image at more or less the frame rate of the stream, eg. 30 fps = ~33ms capture times.

    import { StreamCamera, Codec } from 'pi-camera-connect';
    
    const runApp = async () => {
      const streamCamera = new StreamCamera({
        codec: Codec.MJPEG,
      });
    
      await streamCamera.startCapture();
    
      const image = await streamCamera.takeImage();
    
      // Process image...
    
      await streamCamera.stopCapture();
    };
    
    runApp();
    

Capturing a video stream

Capturing a video stream is easy. There are currently 2 codecs supported: H264 and MJPEG.

Why H264 and MJPEG?

The GPU on the Raspberry Pi comes with a hardware-accelerated H264 encoder and JPEG encoder. To capture videos in real time, using these hardware encoders are required.

Stream

A standard NodeJS readable stream is available after calling createStream(). As with any readable stream, it can be piped or listened to.

import { StreamCamera, Codec } from 'pi-camera-connect';
import * as fs from 'fs';

const runApp = async () => {
  const streamCamera = new StreamCamera({
    codec: Codec.H264,
  });

  const videoStream = streamCamera.createStream();

  const writeStream = fs.createWriteStream('video-stream.h264');

  // Pipe the video stream to our video file
  videoStream.pipe(writeStream);

  await streamCamera.startCapture();

  // We can also listen to data events as they arrive
  videoStream.on('data', data => console.log('New data', data));
  videoStream.on('end', data => console.log('Video stream has ended'));

  // Wait for 5 seconds
  await new Promise(resolve => setTimeout(() => resolve(), 5000));

  await streamCamera.stopCapture();
};

runApp();

You can test the video by viewing it in omxplayer (ships with Raspbian):

$ omxplayer video-stream.h264

Note that this example produces a raw H264 video. Wrapping it in a video container (eg. MP4, MKV, etc) is out of the scope of this module.

API

StillCamera

A class for taking still images. Equivalent to running the raspistill command.

constructor (options?: StillOptions): StillCamera

Instantiates a new StillCamera class.

const stillCamera = new StillCamera({
    ...
});

StillOptions are:

  • width: number - Default: Max sensor width
  • height: number - Default: Max sensor height
  • rotation: Rotation - Default: Rotation.Rotate0
  • flip: Flip - Default: Flip.None
  • delay: number - Default: 1 ms
  • shutter: number - Default: Auto calculated based on framerate (1000000µs/fps). Number is in microseconds
  • sharpness: number - Range: -100-100; Default: 0
  • contrast: number - Range: -100-100; Default: 0
  • brightness: number - Range: 0-100; Default: 50
  • saturation: number - Range: -100-100; Default: 0
  • iso: number - Range: 100-800; Default: Auto
  • exposureCompensation: number - Range: -10-10; Default: 0
  • exposureMode: ExposureMode - Default: ExposureMode.Auto
  • awbMode: AwbMode - Default: AwbMode.Auto
  • analogGain: number - Default: 0
  • digitalGain: number - Default: 0

StillCamera.takeImage(): Promise<Buffer>

Takes a JPEG image from the camera. Returns a Promise with a Buffer containing the image bytes.

const stillCamera = new StillCamera();

const image = await stillCamera.takeImage();

StreamCamera

A class for capturing a stream of camera data, either as H264 or MJPEG.

constructor(options?: StreamOptions): StreamCamera

Instantiates a new StreamCamera class.

const streamCamera = new StreamCamera({
    ...
});

StreamOptions are:

  • width: number - Default: Max sensor width
  • height: number - Default: Max sensor height
  • rotation: Rotation - Default: Rotation.Rotate0
  • flip: Flip - Default: Flip.None
  • bitRate: number - Default: 17000000 (17 Mbps)
  • fps: number - Default: 30 fps
  • codec: Codec - Default: Codec.H264
  • sensorMode: SensorMode - Default: SensorMode.AutoSelect
  • shutter: number - Default: Auto calculated based on framerate (1000000µs/fps). Number is in microseconds
  • sharpness: number - Range: -100-100; Default: 0
  • contrast: number - Range: -100-100; Default: 0
  • brightness: number - Range: 0-100; Default: 50
  • saturation: number - Range: -100-100; Default: 0
  • iso: number - Range: 100-800; Default: Auto
  • exposureCompensation: number - Range: -10-10; Default: 0
  • exposureMode: ExposureMode - Default: ExposureMode.Auto
  • awbMode: AwbMode - Default: AwbMode.Auto
  • analogGain: number - Default: 0
  • digitalGain: number - Default: 0

startCapture(): Promise<void>

Begins the camera stream. Returns a Promise that is resolved when the capture has started.

stopCapture(): Promise<void>

Ends the camera stream. Returns a Promise that is resolved when the capture has stopped.

createStream(): stream.Readable

Creates a readable stream of video data. There is no limit to the number of streams you can create.

Be aware that, as with any readable stream, data will buffer in memory until it is read. If you create a video stream but do not read its data, your program will quickly run out of memory.

Ways to read data so that it does not remain buffered in memory include:

  • Switching the stream to 'flowing' mode by calling either resume(), pipe(), or attaching a listener to the 'data' event
  • Calling read() when the stream is in 'paused' mode

See the readable stream documentation for more information on flowing/paused modes.

const streamCamera = new StreamCamera({
  codec: Codec.H264,
});

const videoStream = streamCamera.createStream();

await streamCamera.startCapture();

videoStream.on('data', data => console.log('New video data', data));

// Wait 5 seconds
await new Promise(resolve => setTimeout(() => resolve(), 5000));

await streamCamera.stopCapture();

takeImage(): Promise<Buffer>

Takes a JPEG image frame from an MJPEG camera stream, resulting in very fast image captures. Returns a Promise with a Buffer containing the image bytes.

Note: StreamOptions.codec must be set to Codec.MJPEG, otherwise takeImage() with throw an error.

const streamCamera = new StreamCamera({
  codec: Codec.MJPEG,
});

await streamCamera.startCapture();

const image = await streamCamera.takeImage();

await streamCamera.stopCapture();

Rotation

Image rotation options.

  • Rotation.Rotate0
  • Rotation.Rotate90
  • Rotation.Rotate180
  • Rotation.Rotate270
import { Rotation } from 'pi-camera-connect';

Flip

Image flip options.

  • Flip.None
  • Flip.Horizontal
  • Flip.Vertical
  • Flip.Both
import { Flip } from 'pi-camera-connect';

Codec

Stream codec options.

  • Codec.H264
  • Codec.MJPEG
import { Codec } from 'pi-camera-connect';

SensorMode

Stream sensor mode options.

  • SensorMode.AutoSelect
  • SensorMode.Mode1
  • SensorMode.Mode2
  • SensorMode.Mode3
  • SensorMode.Mode4
  • SensorMode.Mode5
  • SensorMode.Mode6
  • SensorMode.Mode7
import { SensorMode } from 'pi-camera-connect';

These are slightly different depending on the version of Raspberry Pi camera you are using.

Camera version 1.x (OV5647):

Mode Size Aspect Ratio Frame rates FOV Binning
0 automatic selection
1 1920x1080 16:9 1-30fps Partial None
2 2592x1944 4:3 1-15fps Full None
3 2592x1944 4:3 0.1666-1fps Full None
4 1296x972 4:3 1-42fps Full 2x2
5 1296x730 16:9 1-49fps Full 2x2
6 640x480 4:3 42.1-60fps Full 2x2 plus skip
7 640x480 4:3 60.1-90fps Full 2x2 plus skip

Camera version 2.x (IMX219):

Mode Size Aspect Ratio Frame rates FOV Binning
0 automatic selection
1 1920x1080 16:9 0.1-30fps Partial None
2 3280x2464 4:3 0.1-15fps Full None
3 3280x2464 4:3 0.1-15fps Full None
4 1640x1232 4:3 0.1-40fps Full 2x2
5 1640x922 16:9 0.1-40fps Full 2x2
6 1280x720 16:9 40-90fps Partial 2x2
7 640x480 4:3 40-90fps Partial 2x2

ExposureMode

Exposure mode options.

  • ExposureMode.Off
  • ExposureMode.Auto
  • ExposureMode.Night
  • ExposureMode.NightPreview
  • ExposureMode.Backlight
  • ExposureMode.Spotlight
  • ExposureMode.Sports
  • ExposureMode.Snow
  • ExposureMode.Beach
  • ExposureMode.VeryLong
  • ExposureMode.FixedFps
  • ExposureMode.AntiShake
  • ExposureMode.Fireworks
import { ExposureMode } from 'pi-camera-connect';

AwbMode

White balance mode options.

  • AwbMode.Off
  • AwbMode.Auto
  • AwbMode.Sun
  • AwbMode.Cloud
  • AwbMode.Shade
  • AwbMode.Tungsten
  • AwbMode.Fluorescent
  • AwbMode.Incandescent
  • AwbMode.Flash
  • AwbMode.Horizon
  • AwbMode.GreyWorld
import { AwbMode } from 'pi-camera-connect';
Popular Camera Projects
Popular Stream Projects
Popular Hardware Categories
Related Searches

Get A Weekly Email With Trending Projects For These Categories
No Spam. Unsubscribe easily at any time.
Typescript
Stream
Camera
Mjpeg
Raspberry Pi Camera