How to use Image Correction Filters

Use Image Filters to Improve Read Example

To apply the filter, instantiate the ImageFilterCollection class and create instances of each filter individually. Then assign the object to the ImageFilters property of the BarcodeReaderOptions object. Pass the options object into the Read method along with the sample image.

Let's use the image below as our sample image.

Sample image

From the initial look of the image, it appears to be quite blurry. However, the brightness is acceptable, and the white and black colors are distinguishable. Therefore, we need to apply at least the SharpenFilter and ContrastFilter to improve barcode readability. Refer to the code snippet below to apply the filters to the image, read it, and display it on the console.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-apply-filter.cs

using IronBarCode;
using System;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection()
    {
        new SharpenFilter(3.5f),
        new ContrastFilter(2)
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Write the result value to console
foreach (BarcodeResult result in results)
{
    Console.WriteLine(result.Text);
}

Imports IronBarCode
Imports System

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection() From {
		New SharpenFilter(3.5F),
		New ContrastFilter(2)
	}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Write the result value to console
For Each result As BarcodeResult In results
	Console.WriteLine(result.Text)
Next result

From the code snippet above, apart from applying filters and reading the barcode, we also exported the filtered image to disk. The comparison between the sample and filtered images can be seen below.

Sample image

Filtered sample

Explore Image Correction Filters

IronBarcode offers multiple image filters specifically designed for image correction. These filters can assist in reading imperfect barcode images and enhancing reading accuracy. However, users will need to understand how these filters work in order to select suitable filter and avoid performance issues due to using too many filters or using the wrong filter. Below are all filters available:

• AdaptiveThresholdFilter
• BinaryThresholdFilter
• BrightnessFilter
• ContrastFilter
• InvertFilter
• SharpenFilter
• ErodeFilter
• DilateFilter
• HistogramEqualizationFilter
• Blur Filters
  • GaussianBlurFilter
  • BilateralFilter
  • MedianBlurFilter

The order in which these filters are applied is based on their placement inside the ImageFilterCollection.

Adaptive Threshold Filter

AdaptiveThresholdFilter is one of the filters available in IronBarcode that applies Bradley Adaptive Threshold technique to the image, which automatically determines the threshold for binarizing an image. This filter is ideal for images with non-uniform illumination and varying background intensity levels.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-adaptive-threshold.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new AdaptiveThresholdFilter(0.9f),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("adaptiveThreshold_0.9.png");

Imports IronBarCode

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New AdaptiveThresholdFilter(0.9F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("adaptiveThreshold_0.9.png")

Below are the outputs of applying the filter using different values.

Default value

0.9 value

The constructor also accepts additional parameters for configuration:

• Upper: Upper (white) color for thresholding.
• Lower: Lower (black) color for thresholding.
• Threshold: Threshold limit (0.0-1.0) for binarization.
• Rectangle: Rectangle region to apply the processor on.

As seen in the output image above, the image is binarized to only have black and white colors. While it still does not seem to be ideal for barcode reading as filters need to be used in combinations. Users will need to experiment with the parameter sensitivity to achieve the best results.

Binary Threshold Filter

The BinaryThresholdFilter filters an image by splitting the pixels at the given threshold, where it is used to compare the luminance of a color component. Similar to the AdaptiveThresholdFilter, this filter can introduce new or unwanted noise if it is not used correctly. However, IronBarcode has set default values for the properties of the filter.

Similar to the AdaptiveThresholdFilter, the BinaryThresholdFilter accepts the same additional parameters for configuration:

• Upper: Upper (white) color for thresholding.
• Lower: Lower (black) color for thresholding.
• Threshold: Threshold limit (0.0-1.0) for binarization.
• Rectangle: Rectangle region to apply the processor on.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-binary-threshold.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new BinaryThresholdFilter(0.9f)
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("binaryThreshold_0.9.png");

Imports IronBarCode

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New BinaryThresholdFilter(0.9F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("binaryThreshold_0.9.png")

Below is the sample output of applying filters to the sample image.

Default value

0.9 value

Observing the output image above, we can see that the sample has been binarized into black and white color. However, it can be seen that this filter is clearly not suitable for this image due to barcode bars being eliminated as well as some new noise being introduced.

Brightness Filter

BrightnessFilter is another essential filter in the image filter collection in IronBarcode. As the name suggests, this filter adjusts the brightness of the barcode image. The input to this constructor can vary the Amount of brightness in the output image. The default value is 1, which leaves the image unchanged. A value of 0 will create a completely black image, while values above 1 will make the image brighter.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-brightness.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new BrightnessFilter(1.5f),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("brightness_1.5.png");

Imports IronBarCode

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New BrightnessFilter(1.5F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("brightness_1.5.png")

Below is the output image after applying this filter to the sample input.

Default value

1.5 value

Contrast Filter

The ContrastFilter is used to adjust the level of contrast in an image. Image contrast refers to the difference in color intensity between various elements in an image. Increasing the level of contrast enhances the visibility of details, making the image appear vivid and striking, while reducing contrast makes the image appear softer and more subdued.

The default value is 1, which leaves the image unchanged. A value of 0 will create a completely gray image, while values above 1 will increase the image contrast.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-contrast.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new ContrastFilter(1.5f),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("contrast_1.5.png");

Imports IronBarCode

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New ContrastFilter(1.5F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("contrast_1.5.png")

Applying this filter to the sample input will produce the image below.

Default value

1.5 value

Invert Filter

This filter is used to invert the colors inside an image, making the opposite colors, such as white becomes black and black becomes white. It's particularly useful when users are trying to read a barcode image with a background color. Unlike the BinaryThresholdFilter, this filter inverts the colors directly without the need to specify sensitivity. Moreover, this filter can be used with a CropRectangle to specify the location in the image that needs the color to be inverted, instead of inverting the colors for the entire image.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-invert.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new InvertFilter(),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("invert.png");

Imports IronBarCode

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New InvertFilter()}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("invert.png")

Output image below is the result of applying this filter to the sample input image.

Original image

Inverted

Sharpen Filter

We also provide a sharpening filter in IronBarcode. This filter enhances the sharpness of an image and is very useful when dealing with blurry images. Users can manipulate this filter to adjust the sharpness of an image by adjusting the Sigma value when instantiating the filter object. The default value is 3. Increase the sigma value to increase the image sharpness.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-sharpen.cs

using IronBarCode;
using System;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new SharpenFilter(0.5f),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("sharpen_0.5.png");

Imports IronBarCode
Imports System

Private options As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New SharpenFilter(0.5F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.png", options)

' Export file to disk
results.ExportFilterImagesToDisk("sharpen_0.5.png")

Image below is the sharpened version of the sample input image.

Default value

0.5 value

Comparing the image above with the original image, it seems sharper and would definitely help in barcode reading using IronBarcode. In most cases, SharpenFilter is always applied together with other filters in the ImageFilterCollection class.

Erode Filter

The ErodeFilter removes tiny white noise and thickens barcode bars by removing the pixels near the edge of the shapes. This filter is best used in scenarios where the barcode background has lots of white speckles or if the barcode image itself is too low resolution or blurred, resulting in some bars that are merged. The ErodeFilter has the effect of making the bars thicker while also removing the white specks in the background.

Users can also increase the erosion's effect by inputting an integer representing kernelSize for the filter. The bigger the kernel size, the stronger the effect on the input image. Also do note that the kernelSize is a square and in this example, it would be a 5x5 kernel.

As an example, let's use the ErodeFilter with a larger kernel size to showcase the filter's effects.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-erode.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new ErodeFilter(5),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("erodeFilter.jpg");

IRON VB CONVERTER ERROR developers@ironsoftware.com

Original image

Erode Filter Applied

Comparing the input and output images above, we can see that some bars are visibly thicker due to the more aggressive nature of inputting a larger kernel size to filter. However, the white speckles in the overall picture have decreased. By nature of the erosion filter, the larger the kernel size, you might run into issues erasing thin bars if applied too aggressively, as shown in the picture above, as such developers show test and refine the effect they want by changing the kernel size value they input to the ErodeFilter.

Dilate Filter

The DilateFilter functions as the inverse of the ErodeFilter, operating by expanding bright regions—typically the background—through the addition of pixels to object boundaries. While this filter is useful for repairing damaged or faint barcodes by filling in small gaps or enhancing low-contrast areas, it’s important to note that its effect on barcode bars differs from intuition. Since dilation enlarges bright spaces, it indirectly thins out dark elements such as black barcode bars (assuming it's a white background). This makes the filter particularly effective in scenarios where barcode bars appear overly thick or merged, but excessive use can degrade scan accuracy by excessively narrowing the bars.

Similar to above, users can increase the filter's effect by inputting an integer representing the kernelSize for the filter.

For the example below, we'll use a larger kernel size to showcase the effects of the DilateFilter.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-dilate.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new DilateFilter(5),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("dilateFilter.jpg");

IRON VB CONVERTER ERROR developers@ironsoftware.com

Original image

Dilation Filter Applied

As you can see from the image above, the aggressive use of DilateFilter can potentially destroy the barcode structure, merging closely spaced bars and creating quiet zones in the barcodes. Users should test and refine the effects they want on the image by changing the kernel size value to be larger or smaller, depending on the input image.

HistogramEqualization Filter

The HistogramEqualizationFilter is a filter that enhances the image contrast by redistributing the pixel intensities to improve clarity. It is most commonly used when the barcode is either low contrast, such as faded or washed out images, or images with uneven lighting, such as dark shadows or a bright glare. By analyzing the image histogram, which is the distribution of pixel brightness, it redistributes the pixel values by boosting contrast by stretching the intensity range, where dark pixels become darker and light pixels become lighter.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-histogram-equalization-filter.cs

using IronBarCode;

BarcodeReaderOptions options = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new HistogramEqualizationFilter(),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.png", options);

// Export file to disk
results.ExportFilterImagesToDisk("histogramEqualizationFilter.jpg");

IRON VB CONVERTER ERROR developers@ironsoftware.com

Original image

Histogram Equalization Filter Applied

As you can see from the image above, the black bars are visibly darker, and the spaces are visibly brighter compared to the original image.

Blur Filters

GaussianBlur Filters

The GaussianBlurFilter is used to apply a Gaussian blur to an image. This filter is commonly used to reduce noise in an image.

The filter works by averaging neighboring pixel values in the image using a Gaussian function. The method relies on two adjustable factors:

• Kernel: a matrix used for averaging the pixels.
• Sigma: a value that controls the intensity of the blur.

The default kernel size is 3x3 pixels, and the default Sigma value is 3.0, which produces a moderate blur. Increasing the Sigma value will result in a stronger blur effect. You can also customize the kernel to control the size of the neighborhood that the blur filter averages.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-gaussianblur.cs

using IronBarCode;

BarcodeReaderOptions myOptionsExample = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new GaussianBlurFilter(3, 3, 3.0f),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sharpen.webp", myOptionsExample);

// Export file to disk
results.ExportFilterImagesToDisk("gaussianBlur.png");

Imports IronBarCode

Private myOptionsExample As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New GaussianBlurFilter(3, 3, 3.0F)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sharpen.webp", myOptionsExample)

' Export file to disk
results.ExportFilterImagesToDisk("gaussianBlur.png")

$vbLabelText   $csharpLabel

Applying this filter to the sample input will produce the image below.

Sharpen image

GaussianBlur image

Bilateral Filters

The BilateralFilter is used to smooth images while preserving edges. Unlike simple blur techniques, which affect all pixels uniformly, the Bilateral Filter takes both color differences and pixel distance into account, making it effective for edge-preserving smoothing.

The method relies on three adjustable factors:

• NeighborhoodDiameter: Specifies the diameter of the pixel neighborhood used for filtering. A larger diameter includes more surrounding pixels in the filter. The default value is 5.
• SigmaColor: This represents the color influence. It determines how much the color difference between neighboring pixels affects the filtering. A higher value means that pixels with different colors will influence each other more. The default value is 75.0.
• SigmaSpace: This represents the spatial influence. It determines how much the distance between pixels affects the filtering. A higher value means that pixels that are farther apart will influence each other more. The default value is 75.0.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-bilateral.cs

using IronBarCode;

BarcodeReaderOptions myOptionsExample = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new BilateralFilter(5, 75, 75),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sharpen.webp", myOptionsExample);

// Export file to disk
results.ExportFilterImagesToDisk("bilateral.png");

Imports IronBarCode

Private myOptionsExample As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New BilateralFilter(5, 75, 75)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sharpen.webp", myOptionsExample)

' Export file to disk
results.ExportFilterImagesToDisk("bilateral.png")

$vbLabelText   $csharpLabel

Applying this filter to the sample input produces the image below.

Sharpen image

Bilateral image

MedianBlur Filters

The MedianBlurFilter is a filter used for reducing noise in an image by replacing each pixel’s value with the median value of the surrounding pixels. This filter is particularly effective at preserving edges while removing noise.

• KernelSize: Defines the size of the neighborhood around each pixel used to calculate the median. The value must be an odd number greater than 0. The default value is 5.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-medianblur.cs

using IronBarCode;

BarcodeReaderOptions myOptionsExample = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new MedianBlurFilter(5),
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sharpen.webp", myOptionsExample);

// Export file to disk
results.ExportFilterImagesToDisk("medianBlur.png");

Imports IronBarCode

Private myOptionsExample As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {New MedianBlurFilter(5)}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sharpen.webp", myOptionsExample)

' Export file to disk
results.ExportFilterImagesToDisk("medianBlur.png")

$vbLabelText   $csharpLabel

Applying this filter to the sample input produces the image below.

Sharpen image

MedianBlur image

Save Iterations

When applying multiple filters to the barcode, it can be difficult to view the output after each filter method. This feature allows saving the filtered image after each filter is applied, in the order they are processed. To enable this feature, first pass true to the ImageFilterCollection constructor. Then, use the ExportFilterImagesToDisk method to provide the path and name for the output images.

:path=/static-assets/barcode/content-code-examples/how-to/image-correction-save-iterations.cs

using IronBarCode;

BarcodeReaderOptions myOptionsExample = new BarcodeReaderOptions()
{
    // Choose which filters are to be applied (in order)
    ImageFilters = new ImageFilterCollection(true) {
        new SharpenFilter(3.5f),
        new AdaptiveThresholdFilter(0.5f),
        new ContrastFilter(2)
    },
};

// Apply options and read the barcode
BarcodeResults results = BarcodeReader.Read("sample.webp", myOptionsExample);

// Export file to disk
results.ExportFilterImagesToDisk("filteredImage.png");

Imports IronBarCode

Private myOptionsExample As New BarcodeReaderOptions() With {
	.ImageFilters = New ImageFilterCollection(True) From {
		New SharpenFilter(3.5F),
		New AdaptiveThresholdFilter(0.5F),
		New ContrastFilter(2)
	}
}

' Apply options and read the barcode
Private results As BarcodeResults = BarcodeReader.Read("sample.webp", myOptionsExample)

' Export file to disk
results.ExportFilterImagesToDisk("filteredImage.png")

The filters are applied in the order of the code, and the output images reflect the results of each iteration:

• Sharpen -> After Sharpen
• Sharpen + Adaptive Threshold -> After Adaptive Threshold
• Sharpen + Adaptive Threshold + Contrast -> After Contrast

Sample image

After Sharpen

After Adaptive Threshold

After Contrast

Apart from the ImageFilters properties, users can also add other properties to BarcodeReaderOptions for more accurate reading; 请参阅此文章以获取更多信息。