C# Barcode Reader How-Tos
PM > Install-Package OnBarcode.Barcode.Reader
C#: How to Start
Read multi-barcodes
Scan image specified regions
Apply image processing
Async & multi-thread scanning
Process Barcode Text
Unicode text
Structured Append data
GS1 data elements
Setup .NET Projects
ASP.NET Core
ASP.NET Core MVC
ASP.NET Framework
WinForms (.NET Core)
WPF (.NET Core)
2D Barcode in C#
Data Matrix
PDF417
QR Code
Linear Barcode in C#
Code 39
Code 128
EAN-13
GS1-128
UPC-A

HHow to process image to enhance barcode reading performance in C#?

How to process image to enhance barcode reading in C# ASP.NET Core, Windows application?

Using Free C# Souce Code to read, decode barcodes from processed image for ASP.NET Web Application & IIS Projects

This page will show you how to process the barcode images to improve barcode reading performance in C# program.

How to process image to enhance barcode reading in C#

Download C# Barcode Reader Library
Install C# library to process, read barcode from images in .NET apps
Step by Step Tutorial

Introduction

The main conditions for the accurate recognition of the barcode in the image are as follows: after the image is converted to a black-and-white binary image, the boundaries between the bars and spaces in the barcode must be clear, the shape (boundaries) of the bars must not be distorted (e.g., enlarged, reduced, warped, etc.), and the image must be free of noise.

When the quality of the barcode in the image is poor, the algorithms provided in the SDK can only support certain fixed processing methods to improve the recognition rate. To recognize barcodes from low-quality images, more complex and targeted processing is required.

Image processing is generally carried out in the following ways:

Converting color images/grayscale images to black-and-white images

Different methods used to convert images to black-and-white may result in different widths of the bars and spaces in the barcode. Trying various conversion methods may help find more accurate boundaries between bars and spaces.

Adjusting brightness/contrast

Enhance boundaries in the image (e.g., the junction between bars and spaces), so that the bar areas and space areas can be distinguished more correctly when the image is converted to black-and-white.

Using filters to remove noise points

The less noise there is, the less interference there will be to barcode recognition.

Applying operations such as blurring/sharpening

Modify boundary characteristics to increase the chance of locating accurate boundaries.

How to process barcode images in C#?

Install OnBarcode Imaging library

You need install OnBarcode C# imaging library to process images in your .NET projects.

Search OnBarcode.Image package from "Manage NuGet Packages..S" in Visual Studio
Install the latest version

Read, save processing images

Here we will explain how to apply the image processing in your C# applications.

Create a ImageProcessor object with an image loaded.
Apply one or multiple imaging process functions on the image
Call method GetResult to get the processed image in Bitmap object

Bitmap bitmap = new Bitmap("C:\\input\\sample-image.png");
using (ImageProcessor obj = ImageProcessor.FromBitmap(bitmap))
{
    // To apply one or more imaging processes here. 
    obj.Bilevel(128);
    Bitmap result = obj.GetResult();
}

Using OnBarcode image library, you can read, load an image from file path, Stream or Bitmap object. And you will get the processed image data from Bitmap, Stream object or file path on disc also.

// Read an image
public static ImageProcessor FromFile(String imageFilePath)
public static ImageProcessor FromStream(Stream imageStream)
public static ImageProcessor FromBitmap(Bitmap bitmap)

// Get the processed image
public Bitmap GetResult()
public void SaveToStreamAsPNG(Stream stream)
public void SaveToFileAsPNG(String filePath)

Explore the image processing details

OnBarcode C# image processing library includes the following categories

Image Enhancement
Image Segmentation
Image Filtering
Morphological Processing

Image Enhancement

Image enhancement includes the following functions

Grayscale
Negate
Gamma
Brightness and Contrast
Equalize

Grayscale

Convert the color image to a grayscale image using the specified conversion method.

GrayscaleMethod The difference between the various conversion methods lies in the weights used for the different components (Red, Green, Blue) during conversion.
Average: Uses equal weights for R, G, and B.
Rec601: Based on the color standard Rec.601 (or BT.601).
Rec709: Based on the color standard Rec.709 (or BT.709).

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Grayscale(GrayscaleMethod.Average); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	GrayscaleMethod.Average

GrayscaleMethod.Rec601	GrayscaleMethod.Rec709

Why Grayscale will improve barcode reading performance?

This image filter applies to images that are overall too dark or too bright. Manually selecting different thresholds for black and white pixels allows for more accurate definition of the boundaries between bars and spaces. Generally, a smaller threshold should be used for darker images, and a larger threshold for brighter ones.

This function works with color images and internally converts them to grayscale using the default method.

Negate

Inverts the bits of all pixels in the image. It is mainly applicable to grayscale images and can also be applied to color images.

Suitable for images with inverted black and white.

In some cases, the bars in the image are white (or bright), while the spaces are black (or dark). In such situations, the image must be color-inverted before recognition.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Negate(); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Result

Why Negate image filter will improve barcode reading performance?

Suitable for images with inverted black and white. In some cases, the bars in the image are white (or brighter colors) and the spaces are black (or darker colors). In such situations, the image must be color-inverted before barcode recognition.

Gamma

Applies Gamma correction (grayscale correction) to the image. Supports using independent Gamma values for each color component (Red, Green, and Blue).

gamma When the Gamma value is less than 1, the image becomes darker; when it is greater than 1, the image becomes brighter.
Common range: 0.8 ~ 2.3

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float gamma = 0.8F; 
    obj.Gamma(gamma); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Gamma: 0.8

Gamma: 1.0	Gamma: 2.3

Applies Gamma correction for each color component (Red, Green, and Blue).

gammaRed Gamma for Red component in image.
gammaGreen Gamma for Green component in image.
gammaBlue Gamma for Blue component in image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float gammaRed = 1.0F;
    float gammaGreen = 2.0F; 
    float gammaBlue = 1.0F; 
    obj.Gamma(gammaRed, gammaGreen, gammaBlue); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Gamma Red: 1; Gamma Green: 2; Gamma Blue: 1

Why Gamma image process will improve barcode scanning performance?

Gamma correction is mainly used to adjust the brightness and contrast of an image. When the barcode image is overall too bright or too dark, you can apply correction first to make the boundaries between bars and spaces in the image clearer, thereby improving recognition performance.

When the image is generally dark, select a Gamma value greater than 1; when bright overall, select a Gamma value less than 1.

Supports using different correction values for different color components separately.

Brightness Contrast

Adjust the brightness and contrast of the image.

brightness Adjust the image brightness. The smaller the value, the lower the image brightness; the larger the value, the higher the image brightness.
Valid range: -100 ~ 100 (0 means no change in brightness).
contrast Adjust the image contrast. The smaller the value, the lower the image contrast; the larger the value, the higher the image contrast.
Valid range: -100 ~ 100 (0 means no change in contrast).

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float brightness = -50F; 
    float contrast = 0F; 
    obj.BrightnessContrast(brightness, contrast); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Brightness: -50; Contrast: 0

Brightness: 50; Contrast: 0	Brightness: 0; Contrast: -50

Brightness: 0; Contrast: 50

Why Brightness Contrast image process will improve barcode scanning performance?

Directly adjust the brightness/contrast of the image to make the boundaries between the Bar and Space more distinct, thereby improving the barcode recognition effect.

Equalize

Use Histogram Equalization to adjust the distribution of brightness values of image pixels, thereby enhancing the contrast of the image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Equalize(); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png"));
}

Source image	Brightness: -50; Contrast: 0

Why Equalize image filter will improve barcode scanning performance?

Enhancing the contrast can highlight the boundaries between Bar and Space, thereby improving the barcode recognition effect.

Image Segmentation

Image Segmentation includes the following functions

AdaptiveThreshold
Bilevel

Adaptive Threshold

Take each pixel as the center, calculate the grayscale average of all pixels within a region of the specified size, and use this average as the threshold to define black and white pixels. This method is mainly applicable to grayscale images.

width, height: Width and height of the specified region (in pixels). The larger the value, the more pixels are included in the mean calculation. In terms of processing effect, a smaller region can better highlight details.
offset Define the difference between the mean value and the threshold. The larger the value, the higher the threshold, which will result in more pixels being defined as black.

Note: Setting the offset to a negative number has no effect.

The C# source code below explains how to apply the Adaptive Threshold image process in C# program.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    int width = 10; 
    int height = 10; 
    obj.AdaptiveThreshold(width, height, 0); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Width: 10; Height: 10; Offset: 0

Width: 50; Height: 50; Offset: 0	Width: 500; Height: 500; Offset: 0

Why Adaptive Threshold will improve barcode scanning performance?

This is applicable to situations where there is a significant difference in brightness between different areas of an image. If a uniform threshold is used, the bars in darker areas will appear wider, while those in brighter areas will appear narrower, resulting in incorrect width ratios of bars within the same barcode. Evaluating the threshold for black and white pixels based on adjacent pixels (rather than the entire image) can mitigate such issues.

This function is only applicable to grayscale images. Applying it directly to color images yields poor results. For optimal outcomes, you need to first convert the image to grayscale (using the Grayscale function) before invoking this function.

The selection of Width/Height values is primarily determined by the size of regions in the image where significant brightness variations occur. If such regions are small, smaller Width/Height values should be chosen.

Offset can be used to adjust the width of bars. A larger Offset value means more pixels will be classified as bars, resulting in wider bar widths.

Bilevel

Use the given grayscale value as the threshold to define black and white pixels. This method is applicable to Color or Grayscale images, and Color images will be automatically converted to grayscale images internally.

threshold Define the threshold. A larger value results in more pixels being defined as black; conversely, a smaller value results in more pixels being defined as white.
Valid range: 0 ~ 255.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    int threshold = 128; 
    obj.Bilevel(threshold); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Threshold: 96

Why Bilevel will improve barcode scanning performance?

Suitable for images that are overall too dark or too bright. By manually selecting different thresholds for black and white pixels, you can more accurately define the boundaries between Bars and Spaces. Generally, a smaller threshold should be used if the image is dark overall; conversely, a larger threshold should be used if bright.

This function supports color images and will internally convert the image to grayscale using the default method.

Image Filtering

Image Filtering includes the following functions

Gaussian Blur
Adaptive Blur
Selective Blur
Sharpen
Unsharp Mask
Despeckle
Edge

Gaussian Blur

Apply blur to the image (convolution with a Gaussian distribution).

radius Blur radius, in pixels. The larger the value, the more blurred the image.
Recommended value: 1 ~ 10.
0 means the function will automatically select an appropriate value.
sigma Standard deviation, in pixels. The larger the value, the more blurred the image.
Recommended value: 1/3 of the radius.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float radius = 6; 
    float sigma = 2; 
    obj.GaussianBlur(radius, sigma); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 6; Sigma: 2

Radius: 12; Sigma: 4	Radius: 24; Sigma: 8

Why Gaussian Blur will improve barcode reading performance?

Blurring the image can reduce noise in the image (but also weakens the effective signal). Since the degree of reduction for the two may not be the same, it is beneficial for barcode recognition when noise is reduced more significantly than the signal.

Adaptive Blur

Based on Gaussian blur, reduce the blur intensity in edge regions of the image to preserve edge sharpness.

radius Blur radius, in pixels. The larger the value, the more blurred the image.
Recommended value: 1 ~ 10.
0 means the function will automatically select an appropriate value.
sigma Standard deviation, in pixels. The larger the value, the more blurred the image.
Recommended value: 1/3 of the radius.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float radius = 6; 
    float sigma = 2; 
    obj.AdaptiveBlur(radius, sigma); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 6; Sigma: 2

Radius: 12; Sigma: 4	Radius: 24; Sigma: 8

Why Adaptive Blur will improve barcode reading performance?

Similar to Gaussian blur, but reduces the blur intensity at edge regions (i.e., the boundaries between Bars and Spaces), making it more helpful for locating the boundaries between Bars and Spaces.

Selective Blur

On the basis of applying Gaussian blur, set a contrast threshold and only blur pixels below the threshold.

radius Blur radius, in pixels. The larger the value, the more blurred the image.
Recommended value: 1 ~ 10.
0 means the function will automatically select an appropriate value.
sigma Standard deviation, in pixels. The larger the value, the more blurred the image.
Recommended value: 1/3 of the radius.
threshold Pixels with contrast below the threshold are blurred. The higher the threshold, the larger the area that will be blurred.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float radius = 12; 
    float sigma = 4; 
    float threshold = 80; 
    obj.SelectiveBlur(radius, sigma, threshold); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 12; Sigma: 4; Threshold: 0

Radius: 12; Sigma: 4; Threshold: 50	Radius: 12; Sigma: 4; Threshold: 100

Why Selective Blur will improve barcode reading performance?

Similar to Gaussian blur, but performs selective blurring on image pixels, blurring only regions with high contrast. This helps highlight the boundaries between Bars and Spaces and reduces the impact of the blurring operation on the barcode signal.

Sharpen

Sharpen the image.

radius Gaussian radius, in pixels. The larger the value, the sharper the edges in the image.
0 means the function will automatically select an appropriate value.
sigma Gaussian standard deviation, in pixels. The larger the value, the sharper the edges in the image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float radius = 0; 
    float sigma = 2; 
    obj.Sharpen(radius, sigma); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 0; Sigma: 2

Radius: 0; Sigma: 4	Radius: 0; Sigma: 8

Why Sharpen will improve barcode reading performance?

Sharpening the edges can highlight the boundaries between Bars and Spaces, improving recognition performance.

Sharpening also increases noise, so an appropriate value should be chosen to balance noise and the effective signal (bar boundaries).

Unsharp Mask

Sharpen the image using the Unsharp Mask algorithm.

radius Gaussian radius, in pixels. 0 means the function automatically selects an appropriate value.
sigma Gaussian standard deviation, in pixels.
amount The larger the sharpening amount value, the more obvious the edge areas in the image.
Recommended value: 0.5 ~ 1.5
threshold Defines a threshold for pixel difference. Pixels below the threshold will be ignored. The smaller this value is, the more pixels will be sharpened.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    float radius = 0; 
    float sigma = 1; 
    float amount = 1; 
    float threshold = 0; 
    obj.UnsharpMask(radius, sigma, amount, threshold); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 0; Sigma: 1; Amount: 1; Threshold: 0

Why Unsharp Mask will improve barcode reading performance?

Sharpening the edges in the image can highlight the boundaries between Bars and Spaces, improving the recognition effect.

Sharpening also increases noise at the same time. An appropriate value should be selected to balance noise and the effective signal (Bar boundaries).

Despeckle

Reduce speckle noise in the image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Despeckle();
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Result

Why Despeckle will improve barcode reading performance?

Reduce speckle noise in the image.

Edge

Highlight the edges of shapes in the image. Mainly suitable for grayscale images.

radius Filter radius, in pixels. Valid values: greater than 0.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    int radius = 5; 
    obj.Edge(radius); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Radius: 5

Radius: 10	Radius: 20

Why Edge will improve barcode reading performance?

Use different radius values to detect edges in the barcode image. Choosing an appropriate radius value can reduce noise while highlighting the boundaries between Bars and Spaces.

Morphological Processing

Morphological Processing includes the following functions

Erode
Dilate

Erode

Apply morphological erosion to the image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Erode(); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Result

Why Erode will improve barcode reading performance?

Mainly used to improve the situation where bars in the image are contaminated by surrounding spaces. Expanding the bar areas can correct the width ratio of bars and spaces, thereby improving barcode recognition performance.

Dilate

Apply morphological dilation to the image.

using (ImageProcessor obj = ImageProcessor.FromFile(Path.Combine(Root, "Source.png"))) 
{ 
    obj.Dilate(); 
    obj.SaveToFileAsPNG(Path.Combine(Root, "Result.png")); 
}

Source image	Result

Why Dilate will improve barcode reading performance?

Mainly used to improve the situation where the Bars in the image have invaded the surrounding Spaces.

Shrinking the Bar areas can correct the width ratio between Bars and Spaces, thus improving the recognition effect.

In this page

Start with free trial package: