Guide to using IronOCR Filters

IronOCR provides the tools you need to read images that may need preprocessing in the form of filters. You can choose from a wide array of filters that can manipulate your images to become processable.

Get started with IronOCR

Start using IronOCR in your project today with a free trial.

First Step:
green arrow pointer


List of OCR Image Filters

The following Image filters can really improve performance:

  • Filters to change the Image Orientation
    • Rotate - Rotates images by a number of degrees clockwise. For anti-clockwise, use negative numbers.
    • Deskew - Rotates an image so it is the right way up and orthogonal. This is very useful for OCR because Tesseract's tolerance for skewed scans can be as low as 5 degrees.
    • Scale - Scales OCR input pages proportionally.
  • Filters to manipulate Image Colors
    • Binarize - This image filter turns every pixel black or white with no middle ground. This may improve OCR performance in cases of very low text-to-background contrast.
    • ToGrayScale - This image filter turns every pixel into shades of gray. Unlikely to improve OCR accuracy but may improve speed.
    • Invert - Inverts every color. E.g. White becomes black and vice versa.
    • ReplaceColor - Replaces a color in an image with another color, within a certain threshold.
  • Filters to improve Contrast in an Image
    • Contrast - Increases contrast automatically. This filter often improves OCR speed and accuracy in low-contrast scans.
    • Dilate - Advanced Morphology. Dilation adds pixels to the boundaries of objects in an image. Opposite of Erode.
    • Erode - Advanced Morphology. Erosion removes pixels from object boundaries. Opposite of Dilate.
  • Filters to reduce Image Noise
    • Sharpen - Sharpens blurred OCR Documents and flattens alpha channels to white.
    • DeNoise - Removes digital noise. This filter should only be used in scenarios where noise is expected.
    • DeepCleanBackgroundNoise - Heavy background noise removal. Only use this filter if extreme document background noise is known, as it might reduce OCR accuracy of clean documents and is CPU intensive.
    • EnhanceResolution - Enhances the resolution of low-quality images. This filter is not often needed because OcrInput.MinimumDPI and OcrInput.TargetDPI will automatically catch and resolve low-resolution inputs.

Filter Example and Usage

In the following example, we demonstrate how to apply filters within your code.

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-1.cs
using IronOcr;
using System;

var ocr = new IronTesseract();
using var input = new OcrInput();
input.LoadImage("my_image.png");
input.Deskew();

var result = ocr.Read(input);
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private ocr = New IronTesseract()
Private input = New OcrInput()
input.LoadImage("my_image.png")
input.Deskew()

Dim result = ocr.Read(input)
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel

Debug Filter / What is the filter doing?

If you are having difficulty with reading images or barcodes within your program, there is a way to save an image of a filtered result. This way, you can debug and see exactly what each filter does and how it is manipulating your image.

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-2.cs
using IronOcr;
using System;

var file = "skewed_image.tiff";
var ocr = new IronTesseract();
using var input = new OcrInput();
var pageindices = new int[] { 1, 2 };
input.LoadImageFrames(file, pageindices);
// Here we apply the filter: Deskew
input.Deskew();

// Save the input with filter(s) applied
input.SaveAsImages("my_deskewed");

// We read, then print the text to the console
var result = ocr.Read(input);
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private file = "skewed_image.tiff"
Private ocr = New IronTesseract()
Private input = New OcrInput()
Private pageindices = New Integer() { 1, 2 }
input.LoadImageFrames(file, pageindices)
' Here we apply the filter: Deskew
input.Deskew()

' Save the input with filter(s) applied
input.SaveAsImages("my_deskewed")

' We read, then print the text to the console
Dim result = ocr.Read(input)
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel

Filter Use Cases

Rotate

API Reference

Filter Explanation

Rotate is a filter used to manually set a known rotation to an image to get it closest to being straight. IronOCR has functionality to run Deskew(), however, the degree of tolerance for this is rather narrow and is best used for images that are almost perfectly straight (within 15 degrees or so). For input images that are 90 degrees off, or upside down, we should call Rotate().

Use-Case Code Example

This is an example of calling Rotate to correct an upside-down image:

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-3.cs
using IronOcr;
using System;

var image = "screenshot.png";
var ocr = new IronTesseract();
using var input = new OcrInput();
// Load at least one image
input.LoadImage(image);

// Rotate 180 degrees because image is upside-down
input.Rotate(180);

// Read image into variable: result
var result = ocr.Read(input);

// Example print to console
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private image = "screenshot.png"
Private ocr = New IronTesseract()
Private input = New OcrInput()
' Load at least one image
input.LoadImage(image)

' Rotate 180 degrees because image is upside-down
input.Rotate(180)

' Read image into variable: result
Dim result = ocr.Read(input)

' Example print to console
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel
BeforeInput.Rotate(180)``AfterInput.Rotate(180)``
Screenshot related to Use-Case Code ExampleScreenshot Rotated related to Use-Case Code Example

Deskew

API Reference

Filter Explanation

Uses a Hough Transform to attempt to straighten an image within certain degrees of tolerance. This is important for images that are not completely straight because a tilted document may result in a misread.

Please note
This method returns a boolean, which is true if the filter was applied, and false if it failed to apply due to not being able to detect image orientation. This will fail if the page has no contents to define orientation.

Use-Case Code Example

This is an example of calling Deskew to correct a skewed image:

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-4.cs
using IronOcr;
using System;

var image = @"paragraph_skewed.png";
var ocr = new IronTesseract();
using var input = new OcrInput();
// Load at least one image
input.LoadImage(image);

// Apply deskew with 15 degree snap
bool didDeskew = input.Deskew(15);
if (didDeskew)
{
    // Read image into variable: result
    var result = ocr.Read(input);
    Console.WriteLine(result.Text);
}
else
{
    Console.WriteLine("Deskew not applied because Image Orientation could not be determined.");
}
Imports IronOcr
Imports System

Private image = "paragraph_skewed.png"
Private ocr = New IronTesseract()
Private input = New OcrInput()
' Load at least one image
input.LoadImage(image)

' Apply deskew with 15 degree snap
Dim didDeskew As Boolean = input.Deskew(15)
If didDeskew Then
	' Read image into variable: result
	Dim result = ocr.Read(input)
	Console.WriteLine(result.Text)
Else
	Console.WriteLine("Deskew not applied because Image Orientation could not be determined.")
End If
$vbLabelText   $csharpLabel
BeforeDeskew()``AfterDeskew()``
Paragraph Skewed related to Use-Case Code ExampleParagraph Deskewed related to Use-Case Code Example

Scale

API Reference

Filter Explanation

Scale is a useful image manipulation filter that helps to resize an image using the pixels it already has. This can be used when a barcode is not being scanned because the image is only tens of pixels wide, with each bar as one pixel, or if text is too small with no anti-aliasing.

Please note
There is a sweet-spot for barcode sizes of 1000px x 1000px where barcodes can be read well, which should be considered if your barcode is not being found.

Use-Case Code Example

This is an example of calling Scale to enlarge the gaps between bars in a barcode for scanning:

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-5.cs
using IronOcr;
using System;

var image = @"small_barcode.png";
var ocr = new IronTesseract();

// Optional: This example uses a barcode
ocr.Configuration.ReadBarCodes = true;

using var input = new OcrInput();
// Load at least one image
input.LoadImage(image);

// Apply scale
input.Scale(400); // 400% is 4 times larger

// Read image into variable: result
var result = ocr.Read(input);

// Example print to console
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private image = "small_barcode.png"
Private ocr = New IronTesseract()

' Optional: This example uses a barcode
ocr.Configuration.ReadBarCodes = True

Dim input = New OcrInput()
' Load at least one image
input.LoadImage(image)

' Apply scale
input.Scale(400) ' 400% is 4 times larger

' Read image into variable: result
Dim result = ocr.Read(input)

' Example print to console
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel
BeforeScale()``AfterScale()``
C Sharp Ocr Image Filters 1 related to Use-Case Code ExampleC Sharp Ocr Image Filters 2 related to Use-Case Code Example

Binarize

API Reference

Filter Explanation

The Binarize filter classifies all pixels in an image as either black or white, depending on an adaptive algorithm. This removes all colors and separates the background into a flat white, with anything recognized as text colored a full black for easy reading.

Use-Case Code Example

This is an example of calling Binarize to align colored text and remove background colors and noise:

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-6.cs
using IronOcr;
using System;

var image = @"no-binarize.jpg";
var ocr = new IronTesseract();

using var input = new OcrInput();
// Load at least one image
input.LoadImage(image);

// Apply Binarize
input.Binarize();

// Read image into variable: result
var result = ocr.Read(input);

// Example print to console
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private image = "no-binarize.jpg"
Private ocr = New IronTesseract()

Private input = New OcrInput()
' Load at least one image
input.LoadImage(image)

' Apply Binarize
input.Binarize()

' Read image into variable: result
Dim result = ocr.Read(input)

' Example print to console
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel
BeforeBinarize()``AfterBinarize()``
No Binarize related to Use-Case Code ExampleAfter Binarize related to Use-Case Code Example

Invert

API Reference

Filter Explanation

IronOCR reads best when the image is black text on a white background. The Invert filter is used to achieve this by inverting all colors on an image.

Use-Case Code Example

This is an example of calling Invert to turn white on black into black on white:

:path=/static-assets/ocr/content-code-examples/tutorials/c-sharp-ocr-image-filters-7.cs
using IronOcr;
using System;

var image = @"before-invert.png";
var ocr = new IronTesseract();

using var input = new OcrInput();
// Load at least one image
input.LoadImage(image);

// Apply Invert
input.Invert(true);

// Read image into variable: result
var result = ocr.Read(input);

// Example print to console
Console.WriteLine(result.Text);
Imports IronOcr
Imports System

Private image = "before-invert.png"
Private ocr = New IronTesseract()

Private input = New OcrInput()
' Load at least one image
input.LoadImage(image)

' Apply Invert
input.Invert(True)

' Read image into variable: result
Dim result = ocr.Read(input)

' Example print to console
Console.WriteLine(result.Text)
$vbLabelText   $csharpLabel
BeforeAfter
Before Invert related to Use-Case Code ExampleAfter Invert related to Use-Case Code Example

Frequently Asked Questions

What are OCR image filters?

OCR image filters in IronOCR are tools that allow you to preprocess images to improve their quality for Optical Character Recognition. These filters can manipulate image orientation, colors, contrast, and noise.

How can image orientation be adjusted?

Image orientation can be adjusted using the Rotate and Deskew filters in IronOCR. Rotate allows you to rotate images by a specified number of degrees, while Deskew automatically straightens images that are slightly tilted.

What is the purpose of a Binarize filter?

The Binarize filter in IronOCR converts all pixels in an image to either black or white, removing colors and flattening the background to improve OCR performance, especially in cases of low text-to-background contrast.

When should a DeepCleanBackgroundNoise filter be used?

The DeepCleanBackgroundNoise filter in IronOCR should be used in scenarios where there is extreme document background noise. It is CPU intensive and may reduce OCR accuracy on clean documents.

Can a Scale filter improve barcode scanning?

Yes, the Scale filter in IronOCR can resize an image, making it useful for enlarging barcodes that are too small to scan effectively. It helps increase the gaps between bars for better scanning accuracy.

What does an Invert filter do?

The Invert filter in IronOCR reverses the colors in an image, turning white into black and vice versa. This is useful for converting images to have black text on a white background, which is optimal for OCR.

How can you save a filtered image for debugging?

You can save a filtered image for debugging in IronOCR by using the SaveAsImages function after applying any desired filters. This allows you to visually inspect the effects of the filters on the image.

What is the function of a Contrast filter?

The Contrast filter in IronOCR automatically increases the contrast of an image, improving OCR speed and accuracy in scans with low contrast between the text and background.

What is the difference between Dilate and Erode filters?

Dilate adds pixels to the boundaries of objects in an image, whereas Erode removes pixels from object boundaries. Both are advanced morphology filters in IronOCR used to enhance the clarity of image features.

How does an EnhanceResolution filter work?

The EnhanceResolution filter in IronOCR improves the resolution of low-quality images, though it is not often needed because IronOCR's OcrInput.MinimumDPI and TargetDPI settings automatically handle low-resolution inputs.

Chaknith Bin
Software Engineer
Chaknith works on IronXL and IronBarcode. He has deep expertise in C# and .NET, helping improve the software and support customers. His insights from user interactions contribute to better products, documentation, and overall experience.
Reviewed by
Jeff Fritz
Jeffrey T. Fritz
Principal Program Manager - .NET Community Team
Jeff is also a Principal Program Manager for the .NET and Visual Studio teams. He is the executive producer of the .NET Conf virtual conference series and hosts 'Fritz and Friends' a live stream for developers that airs twice weekly where he talks tech and writes code together with viewers. Jeff writes workshops, presentations, and plans content for the largest Microsoft developer events including Microsoft Build, Microsoft Ignite, .NET Conf, and the Microsoft MVP Summit