C# Camera API for Getting Video Frame

If you have a USB camera, how can you build a simple C# camera application on Windows 10? There are three options:  WIA (Windows Imaging Acquisition), DirectShow and MediaCapture. After trying some sample code that downloaded from CodeProject and GitHub, I got the conclusion: 1. WIA is not good because it does not support my webcam.  2. DirectShow can work well, but there is no C# API provided by Microsoft. You need to create a wrapper for C++ API. 3. MediaCapture class that designed for UWP apps provides C# APIs which provide low-level control over the capture pipeline and enable advanced capture scenarios. In this article, I want to share how to create a simple C# webcam app in which I can handle every preview frame myself.

Read more

How to Bridge C Code to Create Swift Barcode Reader on Mac

Have you implemented any barcode reader software in Swift on Mac OS X? Since Dynamsoft released the 1D/2D barcode SDK for Mac, I was wondering how I can bridge the C dylib with Swift. Why Swift, not Objective-C? Because Apple is encouraging developers to migrate to the new programming language, it is important to master the new tool if you want to create more excellent apps for Mac and iOS in the future. Swift is totally new to me, and thus I spent two days googling relevant resources and finally figured out the solution.

Read more

How to Make Java Barcode Reader with Dynamsoft Barcode SDK

Last week, Dynamsoft released Barcode Reader (DBR) SDK v2.0, which is available for Windows and Mac. The Windows installer contains Barcode libraries for ActiveX, C/C++, and .NET. If you are a Java developer, you have to use JNI to link native C/C++ libraries. In this tutorial, I’ll demonstrate how to invoke the native methods of Dynamsoft Barcode SDK via JNI to create a Java Barcode Reader.

Read more

Wrapping C++ OCR Library in Java

Dynamsoft OCR SDK is totally implemented in C++, which means it is easy to be wrapped in high-level programming languages, such as C#, Java, Python and so on. As a proprietary development SDK, so far, only .NET OCR library is available for commercial use. Because some of developers and users are hoping that Dynamsoft could provide a Java OCR library, I wrapped the C++ OCR library for test. Anyone can feel free to use the sample, and I’d like to receive feedbacks from you.

Read more

Barcode Generator with HTML5 WebSocket

Last week, I shared how to transmit an image between a WebSocket server and a Web client. Based on that, we can make a little bit of improvement. Today, I would like to share how to generate barcode on .Net server side and display it on web browser.

websocket barcode


Remote Barcode Generator

  • In your WinForms project, add DynamicBarcode.dll and set its property Copy to Output Directory as Copy always. As so, the Dynamic .NET TWAIN component is able to automatically load the barcode library.

barcode library

  • Encode the received message, and draw generated barcode onto a background image:
            float scale = 3;
            short sImageIndex = 0;
            dynamicDotNetTwain.MaxImagesInBuffer = 1;
            bool isLoaded = dynamicDotNetTwain.LoadImage("test.png");
            dynamicDotNetTwain.AddBarcode(sImageIndex, Dynamsoft.DotNet.TWAIN.Enums.Barcode.BarcodeFormat.QR_CODE, message, "", 0, 0, scale);
  • Send image width, height and data to web client:
            Image img = dynamicDotNetTwain.GetImage(sImageIndex);

            ImageData imageData = load_image(img);
            session.Send(imageData.Width + "," + imageData.Height);
            session.Send(imageData.Data, 0, imageData.Data.Length);
            imageData = null;
  • Display the barcode image on JavaScript web client:
               ws.onmessage = function (evt) {
                    var data = evt.data;
                    if (!data)

                    if (data instanceof ArrayBuffer) {
                    else {
                        var len = data.length;
                        var index = data.indexOf(",");
                        var w = data.substring(0, index);
                        var h = data.substring(index + 1, len);
                        imageWidth = parseInt(w);
                        imageHeight = parseInt(h);

Now, you can have fun with the remote barcode generator via all browsers (Chrome, Firefox, IE etc.) which support HTML5 WebSocket.

Source Code



Image Transmission between a HTML5 WebSocket Server and a Web Client

HTML5 WebSocket facilitates the communication between web browsers and local/remote servers. If you want to learn a simple websocket example, creating a WebSocket Server in C# and a Web client in JavaScript, you can refer to SuperWebSocket, which is a .NET implementation of Web Socket Server.

In this article, I would like to share how I implemented a simple WebSocket solution for image transmission based on the basic sample code of SuperWebSocket and Dynamic .NET TWAIN, especially what problem I have solved.

Here is a quick look of the WebSocket server and the HTML5 & JavaScript client.

WebSocket ServerWebSocket Client


Create a .NET WebSocket Server

Create a new WinForms project, and add the following references which are located at the folder of SuperWebSocket.

WebSocket References

Add the required namespaces:

using Dynamsoft.DotNet.TWAIN;

using SuperSocket.SocketBase;
using SuperWebSocket;

In the sample code, the server is launched with port 2012, no IP specified by default. You can specify the IP address for remote access. For example:

if (!appServer.Setup("", 2012)) //Setup with listening port
     MessageBox.Show("Failed to setup!");

You can use Dynamic Web TWAIN component to do some image operation. With two lines of code, I can load an image file:

bool isLoad = dynamicDotNetTwain.LoadImage("dynamsoft_logo_black.png"); // load an image
Image img = dynamicDotNetTwain.GetImage(0);

Be careful of the image format. It is PNG. If you want to display it in a Web browser, you need to convert the format from PNG to BMP:

            byte[] result;
            using (System.IO.MemoryStream stream = new System.IO.MemoryStream())
                img.Save(stream, System.Drawing.Imaging.ImageFormat.Bmp);   // convert png to bmp
                result = stream.GetBuffer();

It’s not done yet. The byte array contains some extra image information, which is 54 bytes in length. So the actual data length is:

int iRealLen = result.Length - 54;
byte[] image = new byte[iRealLen];

Here are some tricky things that if you just send the subset of the byte array to your Web browser, you will find the displayed image is upside-down, and the color is also incorrect.

To fix the position issue, you need to sort the bytes of original data array from bottom to top. As to the color, exchange the position of blue and red. See the code:

            int iIndex = 0;
            int iRowIndex = 0;
            int iWidth = width * 4;
            for (int i = height - 1; i >= 0; --i)
                iRowIndex = i * iWidth;
                for (int j = 0; j < iWidth; j += 4)
                    // RGB to BGR
                    image[iIndex++] = result[iRowIndex + j + 2 + 54]; // B
                    image[iIndex++] = result[iRowIndex + j + 1 + 54]; // G
                    image[iIndex++] = result[iRowIndex + j + 54];     // R
                    image[iIndex++] = result[iRowIndex + j + 3 + 54]; // A

Now, you can send the data via:

session.Send(imageData.Data, 0, imageData.Data.Length);

Create a JavaScript Client

To receive the image as ArrayBuffer on the client side, you have to specify the binaryType after creating a WebSocket:

ws.binaryType = "arraybuffer";

Once the image data is received, draw all bytes onto a new canvas, and finally create an image element to display the canvas data:

                var imageWidth = 73, imageHeight = 73; // hardcoded width & height. 
                var byteArray = new Uint8Array(data);

                var canvas = document.createElement('canvas');
                canvas.height = imageWidth;
                canvas.width = imageHeight;
                var ctx = canvas.getContext('2d');

                var imageData = ctx.getImageData(0, 0, imageWidth, imageHeight); // total size: imageWidth * imageHeight * 4; color format BGRA
                var dataLen = imageData.data.length;
                for (var i = 0; i < dataLen; i++)
                    imageData.data[i] = byteArray[i];
                ctx.putImageData(imageData, 0, 0);

                // create a new element and add it to div
                var image = document.createElement('img');
                image.width = imageWidth;
                image.height = imageHeight;
                image.src = canvas.toDataURL();

                var div = document.getElementById('img');

Download Sample