@innovatrics/jslib-html5-camera-photo

jslib-html5-camera-photo

The first objective of this package comes from the need to have a js library that can help me to capture picture from mobile or desktop camera through the browser with the HTML5 video and canvas elements. So instead of using the native navigator.mediaDevices.getUserMedia() and manage the stream and the constraints, i need an abstraction it into a small lib that can switch between camera and get the desired resolution.

Another js camera ? Yes! I found webcamjs and jpeg_camera but i need to switch easily from camera environment and user. You need to build the constraint for getUserMedia()... Another need is to have a sizeFactor instead of a fixing 'width' and 'height' that can not fit with the ratio of the resolution that camera can pick.

Features of the library.

1. Choose between environment or user camera, fall back to the default camera.
2. Set ideal resolution, fall back to the default resolution.
3. Get the maximum resolution of the camera, fall back to the default resolution.
4. Choose dataURI image format type between jpg or png.
5. If image format is jpg, choose the compression value is the between [0, 1].
6. Choose dataURI image mirror if you want to get an mirror dataURI of the camera.

Simple Live Demo

https://mabelanger.github.io/jslib-html5-camera-photo/

supported browsers (getUserMedia)

https://caniuse.com/#search=getUserMedia ...(as April 2018)

Available camera facingModes

FACING_MODES []	Description
USER	The source is facing toward the user (a self-view camera).
ENVIRONMENT	The source is facing away from the user (viewing the environment).

Below is an illustration of the video facing modes in relation to the user.

src : https://www.w3.org/TR/mediacapture-streams/#dom-videofacingmodeenum

Getting started

You can use the library with vanilla JavaScript, React, Jquery, Angular...

Installation

npm install --save jslib-html5-camera-photo

yarn add jslib-html5-camera-photo

Both Yarn and npm download packages from the npm registry.

Usage

Constructor

import CameraPhoto, { FACING_MODES, IMAGE_TYPES } from 'jslib-html5-camera-photo';

// get your video element with his corresponding id from the html
let videoElement = document.getElementById('videoId');

// pass the video element to the constructor.
let cameraPhoto = new CameraPhoto(videoElement);

Start the default mode (facing Mode & resolution)

If you do not specify any prefer resolution and facing mode, the default is used. The function return a promise. If the promises success it will give you the stream if you want to use it. If it fail, it will give you the error.

// default camera and resolution
cameraPhoto.startCamera()
  .then((stream)=>{/* ... */})
  .catch((error)=>{/* ... */});

Start with ideal facing Mode & default resolution

// environment (camera point to environment)
cameraPhoto.startCamera(FACING_MODES.ENVIRONMENT, {})
  .then((stream)=>{/* ... */})
  .catch((error)=>{/* ... */});

// OR user (camera point to the user)
cameraPhoto.startCamera(FACING_MODES.USER, {})
  .then((stream)=>{/* ... */})
  .catch((error)=>{/* ... */});

Start with ideal (facing Mode & resolution)

// example of ideal resolution 640 x 480
cameraPhoto.startCamera(facingMode, {width: 640, height: 480})
  .then((stream)=>{/* ... */})
  .catch((error)=>{/* ... */});

Start with the maximum resolution

it will try the range of width [3840, 2560, 1920, 1280, 1080, 1024, 900, 800, 640, default] px to take the maximum width of 3840px if it can't, 2560px and so on ... until the fall back of the default value of the camera. The facingMode is optional.

// It will try the best to get the maximum resolution with the specified facingMode
cameraPhoto.startCameraMaxResolution(facingMode)
  .then((stream)=>{/* ... */})
  .catch((error)=>{/* ... */});

getDataUri()

Function that return the dataUri of the current frame of the camera. To use that function build the configuration object with the corresponding properties. To use the default value, just ommit the properties:

Parameters

• sizeFactor (Number): Used to get a desired resolution. Example, a sizeFactor of 1 get the same resolution of the camera while sizeFactor of 0.5 get the half resolution of the camera. The sizeFactor can be between range of ]0, 1] and the default value is 1.

• imageType (String): Used to get the desired image type between jpg or png. to specify the imageType use the constant IMAGE_TYPES, for example to specify jpg format use IMAGE_TYPES.JPG. The default imageType is png.

• imageCompression (Number): Used to get the desired compression when jpg is selected. choose a compression between [0, 1], 1 is maximum, 0 is minimum. The default value imageCompression is 0.92.

• isImageMirror (Boolean): Used to get an image mirror when is set to true, the result of the dataUri is the mirror of the actual camera data. Many software that use camera mirror like hangout etc... Please note if you want to enable this option, for consistency with the camera video, you need to use css transform: rotateY(180deg) to the <video> tag to mirror the stream, because the stream is not mirrored. It's only apply to the canvas dataUri. The default value is false (no mirror).

Available image types

IMAGE_TYPES []	Description
JPG	set image image/jpeg to the data URI
PNG	set image image/png to the data URI (the default value)

// Use all the default value
const config = {};
let dataUri = cameraPhoto.getDataUri(config);

// OR

// Specify sizeFactor, imageType, imageCompression, isImageMirror

const config = {
  sizeFactor : 1;
  imageType : IMAGE_TYPES.JPG
  imageCompression : .95;
  isImageMirror : false;
}

let dataUri = cameraPhoto.getDataUri(config);

Get the camera settings

the function return null if no stream exist (camera not started) or an object with the camera settings attributes of (aspectRatio, frameRate, height, width).

let cameraSettigs = cameraPhoto.getCameraSettings();
if (cameraSettigs) {
  let {aspectRatio, frameRate, height, width} = cameraSettigs;
  let settingsStr =
      `aspectRatio:${aspectRatio} ` +
      `frameRate: ${frameRate} ` +
      `height: ${height} ` +
      `width: ${width}`;
  console.log(settingsStr);
}

Get the input video device infos

the function return empty array [] if no input video device exist. In order to read the video devices, the browser need to start the camera with with startCamera() or startCameraMaxResolution() before calling the function getInputVideoDeviceInfos() it return a list of objects with the device info attributes of (kind, label, deviceId).

let inputVideoDeviceInfos = cameraPhoto.getInputVideoDeviceInfos();
inputVideoDeviceInfos.forEach((inputVideoDeviceInfo) => {
  let {kind, label, deviceId} = inputVideoDeviceInfo;
  let inputVideoDeviceInfoStr = `
        kind: ${kind}
        label: ${label}
        deviceId: ${deviceId}
    `;
  console.log(inputVideoDeviceInfoStr)
});

Stop the camera

Function that stop the camera. If it success, no value is returned. It can fail if they is no camera to stop because the camera has already been stopped or never started. It will give a parameter of Error('no stream to stop!'). Note that each time we start the camera, it internally using this stop function to be able to apply new constraints.

// It stop the camera
cameraPhoto.stopCamera()
  .then(()=>{/* ... */})
  .catch((error)=>{/* ... */});

Full example vanilla Js & HTML

HTML

<!-- needed to by the camera stream -->
<video id="videoId" autoplay="true"></video>

<!-- needed if you want to display the image when you take a photo -->
<img alt="imgId" id="imgId">

<!--buttons to trigger the actions -->
<button id="takePhotoButtonId">takePhoto</button>
<button id="stopCameraButtonId">stopCamera</button>

JavaScript

import CameraPhoto, { FACING_MODES } from 'jslib-html5-camera-photo';

// get video and image elements from the html
let videoElement = document.getElementById('videoId');
let imgElement = document.getElementById('imgId');

// get buttons elements from the html
let takePhotoButtonElement = document.getElementById('takePhotoButtonId');
let stopCameraButtonElement = document.getElementById('stopCameraButtonId');

// instantiate CameraPhoto with the videoElement
let cameraPhoto = new CameraPhoto(videoElement);

/*
 * Start the camera with ideal environment facingMode
 * if the environment facingMode is not available, it will fallback
 * to the default camera available.
 */
cameraPhoto.startCamera(FACING_MODES.ENVIRONMENT)
  .then(() => {
    console.log('Camera started !');
  })
  .catch((error) => {
    console.error('Camera not started!', error);
  });

// function called by the buttons.
function takePhoto () {
  const config = {};
  let dataUri = cameraPhoto.getDataUri(config);
  imgElement.src = dataUri;
}

function stopCamera () {
  cameraPhoto.stopCamera()
    .then(() => {
      console.log('Camera stoped!');
    })
    .catch((error) => {
      console.log('No camera to stop!:', error);
    });
}

// bind the buttons to the right functions.
takePhotoButtonElement.onclick = takePhoto;
stopCameraButtonElement.onclick = stopCamera;

Full example with React

A project with react is build with this library react-html5-camera-photo

import React from 'react';
import CameraPhoto, { FACING_MODES } from 'jslib-html5-camera-photo';

class App extends React.Component {
  constructor (props, context) {
    super(props, context);
    this.cameraPhoto = null;
    this.videoRef = React.createRef();
    this.state = {
      dataUri: ''
    }
  }

  componentDidMount () {
    // We need to instantiate CameraPhoto inside componentDidMount because we
    // need the refs.video to get the videoElement so the component has to be
    // mounted.
    this.cameraPhoto = new CameraPhoto(this.videoRef.current);
  }

  startCamera (idealFacingMode, idealResolution) {
    this.cameraPhoto.startCamera(idealFacingMode, idealResolution)
      .then(() => {
        console.log('camera is started !');
      })
      .catch((error) => {
        console.error('Camera not started!', error);
      });
  }

  startCameraMaxResolution (idealFacingMode) {
    this.cameraPhoto.startCameraMaxResolution(idealFacingMode)
      .then(() => {
        console.log('camera is started !');
      })
      .catch((error) => {
        console.error('Camera not started!', error);
      });
  }

  takePhoto () {
    const config = {
      sizeFactor: 1
    };

    let dataUri = this.cameraPhoto.getDataUri(config);
    this.setState({ dataUri });
  }

  stopCamera () {
    this.cameraPhoto.stopCamera()
      .then(() => {
        console.log('Camera stoped!');
      })
      .catch((error) => {
        console.log('No camera to stop!:', error);
      });
  }

  render () {
    return (
      <div>
        <button onClick={ () => {
          let facingMode = FACING_MODES.ENVIRONMENT;
          let idealResolution = { width: 640, height: 480 };
          this.startCamera(facingMode, idealResolution);
        }}> Start environment facingMode resolution ideal 640 by 480 </button>

        <button onClick={ () => {
          let facingMode = FACING_MODES.USER;
          this.startCamera(facingMode, {});
        }}> Start user facingMode resolution default </button>

        <button onClick={ () => {
          let facingMode = FACING_MODES.USER;
          this.startCameraMaxResolution(facingMode);
        }}> Start user facingMode resolution maximum </button>

        <button onClick={ () => {
          this.takePhoto();
        }}> Take photo </button>

        <button onClick={ () => {
          this.stopCamera();
        }}> Stop </button>

        <video
          ref={this.videoRef}
          autoPlay="true"
        />
        <img
          alt="imgCamera"
          src={this.state.dataUri}
        />
      </div>
    );
  }
}

export default App;

import the UMD module with HTML script tag

You can build the dist and then serve it with :

$ npm run buildBrowser
$ npm run serve:dist

Or you can copy the dist folder of the repo.

Example :

<script src="/jslib-html5-camera-photo.min.js"></script>
<script>
  ...
  var FACING_MODES = JslibHtml5CameraPhoto.FACING_MODES;
  var cameraPhoto = new JslibHtml5CameraPhoto.default(videoElement);
  ...
</script

Development

I choose the env dev of create-react-app even if it is vanilla js library because it's simple to use and really efficient to develop but you don't necessarily need react to use it. If you want to fix bug or add functionalities please contribute :)

Documentations

• Auto scan resolution camera
• Great info of new API
• W3 specs

Other interesting projects

• Webrtc/adapter
• GetUserMedia.js