How to Improve OCR Accuracy

Summary

The video script provides a detailed tutorial on using Scan2CAD's OCR functionality to convert raster images with text into editable vector text. It emphasizes the importance of high-quality, clear images for optimal OCR results and discusses the manual cleanup of small or touching text elements that may hinder recognition. The script guides viewers through setting OCR parameters, including character size and confidence levels, and offers tips for handling intersecting lines and vertical text. It concludes with manual editing techniques for perfecting the conversion and exporting the final vectorized document.

Takeaways

• 🔍 Use high-quality, high-resolution images with minimal pixelation and blurriness for the best OCR results.
• 📚 Ensure the text in the image is clear and easy to read for effective text recognition.
• 🖋 Handwritten or stylized text may not be recognized well by OCR, especially if it's not clear.
• 🔍 Small text details, like holes in letters, can be lost, affecting OCR accuracy.
• 👀 Scan2CAD's OCR functionality can convert raster text into editable vector text.
• ✂️ Manual editing may be necessary for text that is too close or has intersecting lines.
• 🔢 Set the maximum character size using 'Select from Image' for accurate OCR settings.
• 📏 The minimum character size is usually calculated automatically but can be adjusted manually.
• 📈 Minimum confidence level determines the display of text objects based on their recognition certainty.
• 🔄 Character rotation options should be used based on the presence of vertical or angular text.
• 🌐 Choose the appropriate language and document type for more accurate OCR results.
• 🖼️ After conversion, manually adjust and clean up the text for optimal results.
• 🖊️ Use the 'Draw Text' tool to replace or complete text that wasn't converted properly.
• 🖼️ The 'Highlight Vectors' feature helps in identifying and editing different vector elements.

Q & A

What is the main issue with converting raster text to vector lines using automatic conversion programs?

The main issue is that the text often ends up as vector polylines, which are not editable as true type vector text.

What does OCR stand for and what does it do in the context of Scan2CAD?

OCR stands for Optical Character Recognition. In Scan2CAD, it recognizes the text objects in a raster image and converts them into editable true type vector text.

What are the key factors to consider when choosing a document for automatic conversion?

The document should have good quality, high resolution, minimal pixelation, and blurriness, with clear and easy-to-read text.

Why might Scan2CAD struggle with converting handwritten text?

Handwritten text, especially if stylized, may not be recognized well by Scan2CAD due to variations in handwriting and potential pixelation issues.

What can be done if the text in the image is too small or lacks fine details?

For smaller text, you can manually erase parts that are touching or add details like holes in letters to make them more recognizable for Scan2CAD.

What is the purpose of setting the maximum character size in Scan2CAD's OCR functionality?

Setting the maximum character size helps Scan2CAD to identify and convert the largest characters in the image, which in turn assists in automatically calculating the minimum character size.

What does the minimum confidence level in OCR represent?

The minimum confidence level represents the certainty of Scan2CAD in converting text objects. Text objects below this level may not be displayed if they do not meet the set confidence threshold.

Why is it important to consider character rotation settings in OCR?

Character rotation settings are important to ensure that Scan2CAD can recognize and convert text that is not only horizontal but also vertical or at an angle.

What should be the default document type setting for technical drawings in Scan2CAD?

The default document type setting for technical drawings should be 'technical' to optimize the OCR conversion process.

How can you manually correct the converted text in Scan2CAD if it's not accurate?

You can use the 'Highlight Vectors' feature to see the converted vectors, then use the erase tool to remove inaccurate parts and the text tool to manually add or correct the text.

What is the recommended minimum confidence level setting for most conversions in Scan2CAD?

The recommended default minimum confidence level is 60, which should be used unless there is a specific reason to adjust it.

Outlines

00:00

📚 Advanced OCR Functionality in Scan2CAD

This paragraph discusses the process of converting raster text to editable vector text using Scan2CAD's Optical Character Recognition (OCR) feature. It emphasizes the importance of high-quality, high-resolution images with clear, easily readable text for optimal OCR results. The script mentions challenges with stylized handwriting and small text details, such as holes in letters, that may not be recognized properly. It provides tips for cleaning up images, such as using the erase tool to separate touching letters and restore details. The paragraph also explains the OCR settings, including character size selection, minimum character size, confidence level, character rotation, and language options, to ensure the best conversion results.

05:08

🖌️ Post-OCR Image Cleanup and Finalization

The second paragraph focuses on the post-OCR conversion steps in Scan2CAD. It describes generating a preview to assess the OCR results, noting issues with letters that were too close together and not converting well. The script suggests adjusting the minimum confidence level setting and provides a caution against setting it too high, which could result in the disappearance of text objects. It also touches on the option to handle intersecting lines and the manual correction process, including using the 'Highlight Vectors' feature, erasing incorrect vector lines, and adding true type text. The paragraph concludes with the finalization of conversions, adjusting text placement, and exporting the cleaned-up file in the desired format, showcasing Scan2CAD's capability to convert technical images with text effectively.

Keywords

💡Automatic Conversion Programs

Automatic conversion programs refer to software tools that convert documents or images into different formats, often from raster to vector. In the context of the video, these programs are used to transform documents with raster text into editable vector text. The script mentions that using such programs can result in vector polylines instead of clean text, which is not always desirable.

💡Raster Text

Raster text is a type of text representation in images where the text is composed of pixels, similar to how photographs are made up. The video script discusses the challenges of converting raster text into vector text, emphasizing the need for high-quality images to ensure accurate conversion.

💡Vector Text

Vector text is text that is represented by mathematical formulas, allowing it to be scaled infinitely without losing quality. The script explains that the goal of using Scan2CAD's OCR functionality is to convert raster text into editable vector text, which is more desirable for editing and manipulation in technical drawings.

💡Optical Character Recognition (OCR)

OCR is a technology that enables the conversion of various types of documents, such as scanned paper documents, PDF files, or images captured by a digital camera, into editable and searchable data. In the video, OCR is highlighted as Scan2CAD's advanced feature for recognizing text objects in raster images and converting them into editable vector text.

💡High Resolution

High resolution refers to the quality of an image, where more pixels are packed within a certain area, resulting in a clearer and more detailed image. The script stresses the importance of using high-resolution images for conversion to minimize pixelation and blurriness, which can affect the OCR process.

💡Pixelation

Pixelation is the result of an image being composed of large, visible pixels, often occurring in low-resolution images. The video script mentions that minimal pixelation is desired for the OCR process to work effectively, as pixelation can obscure the details of the text and hinder accurate recognition.

💡Handwritten Text

Handwritten text refers to text that is written by a person, as opposed to printed or typed text. The script notes that handwritten text may not be well-suited for OCR conversion, especially if the handwriting is stylized or unclear, which can complicate the recognition process.

💡Minimum Confidence Level

The minimum confidence level is a measure of how sure the OCR software is about the accuracy of the text it has recognized. The script explains that Scan2CAD assigns a confidence level to each text object, and a higher level indicates a more accurate conversion. The user can set a minimum confidence level to filter out text that is below a certain threshold of certainty.

💡Character Rotation

Character rotation in the context of OCR refers to the ability of the software to recognize and correctly orient text that is not horizontal. The script mentions that Scan2CAD allows users to select whether to include recognition for horizontal, vertical, or angular text, which can be important for certain types of documents.

💡Vectorizing OCR

Vectorizing OCR is the process of using OCR technology to convert raster images of text into vector format. The script describes the steps to initiate this process in Scan2CAD, including setting the maximum character size and using the OCR tab to configure settings for optimal text recognition and conversion.

💡Intersecting Lines

Intersecting lines refer to lines in a drawing or image that cross each other. The script mentions an option in Scan2CAD for handling intersecting lines, which can be useful in niche situations where text or other elements are obscured or incorrectly recognized due to lines crossing over them.

Highlights

Automatic conversion programs often create vector polylines from raster text, which is not always desirable.

Scan2CAD has advanced OCR functionality to convert raster text into editable vector text.

High-resolution, clear images with minimal pixelation and blurriness are recommended for conversion.

Handwritten, stylized text may not work well with OCR due to recognition challenges.

Small text details like holes in letters can be lost, affecting OCR accuracy.

Letters touching each other may be recognized as a single pixel cluster, not individual text.

Manually editing the image to separate touching letters can improve OCR results.

Clicking 'Convert Raster Image' and selecting 'Vectorizing OCR' initiates the OCR process.

Setting the maximum character size using 'Select from Image' helps in accurate OCR.

Minimum character size is automatically calculated or can be set manually.

Minimum confidence level determines the display of text objects based on OCR certainty.

High confidence levels are assigned to clear, minimally pixelated text.

Adjusting the minimum confidence level can affect the visibility of converted text.

Character rotation settings can be customized for horizontal, vertical, or angular text.

Language and document type selections are crucial for accurate OCR results.

Running the OCR generates a preview of the converted text.

Post-OCR, manual adjustments can be made to clean up the converted text.

Using 'Highlight Vectors' helps in identifying and correcting converted vector objects.

Finalizing conversions and exporting the file completes the OCR process.

Scan2CAD's OCR can effectively convert various technical images with text.

Transcripts

[music]

Usually, when you use automatic  conversion programs to convert  

documents like this one that have raster text,  you're gonna end up with a lot of vector lines,  

and even the text ends up as vector polylines.  And sometimes you don't want that, you want to  

convert them into actual editable two-type  vector text. Luckily enough Scan2CAD has  

really advanced OCR functionality, which just  means Optical Character Recognition, which should  

recognize the text objects in this raster image  and convert them into editable two-type text.  

Now with any file that you want to convert  automatically, you're gonna want to get the file  

that has good quality, high resolution, minimal  pixelation, and blurriness. As for the text in the  

image, you're gonna want to make sure that  they're really clear and easy to read.  

Sometimes, you're gonna have a  document with handwritten text.  

And that doesn't really work well, especially if  the handwriting is a little bit more stylized.

For the smaller text, you can see  that some of them don't really have  

fine details, the holes in the A's are erased,  for example. And some of the left letters are  

a bit too close to each other so much so  that the pixels are actually touching,  

Scan2CAD might not recognize this very readily.  If these letters are touching, Scan2CAD recognizes  

as one single pixel cluster, so they won't  be recognized as individual text. If you can  

get images that don't have these types of texts,  or you can manually erase the parts that are  

touching by clicking on the erase tool here, and  just going over and making gaps between these two  

letters. You can even put some holes in the A's  here. In any case, if you have a better quality  

document, you shouldn't be  having those problems here.  

Once you've had some time to clean up the image,  you can click on "Convert Raster Image". Click on  

"Vectorizing OCR". Once you click that, there's  an OCR tab that pops up here, click on that.

You wanna set the maximum character size, you can  click on "Select from Image", look for the largest  

characters here, which I think might be the  room label. So let's zoom in on any one of them.  

Click and drag from the top to the  bottom or from the bottom to the top.  

Once you let go, it automatically  sets the maximum character size here.  

And by default, the minimum character size  will also be calculated automatically from the  

maximum character size. You can set it up  manually by un-ticking this one and just  

typing it in. But usually, you're gonna wanna  keep that ticked to get the best results.

Another thing I want to talk about  is the minimum confidence level.  

Basically based on the quality of the  image, based on the pixelation, based on the  

quality of the individual raster text, Scan2CAD  assigns all of these text objects a certain  

confidence level in percent. If it's sure that it  converted it properly, because it's a clear word,  

and has minimal pixelation, it has very clear  letters, then it's gonna assign it a high  

confidence level. If it's a little bit more  blurry, if it's not super clear, it's gonna  

assign it a lower confidence level. This minimum  percentage here just makes it so that any text  

object that's below this minimum confidence level  won't be displayed. The default value is 60,  

if you set it to 90 for example, once  you convert the file, a lot of these  

text objects might start to disappear because  they don't reach the minimum confidence level that  

you set here. For character rotation, by default,  horizontal is ticked and vertical angular are not.  

If you have any vertical or angular  text here, then you might as well  

take them but as much as possible leave them  unticked if you don't need to use them anyway.

Because if you leave them ticked, it  might create some false positives.  

As for the language, you can choose between all  of the languages that we have here. By default is  

going to be English, and for the document type  and technical drawings such as this one should  

be set the technical and if you have a mostly  text-based raster file, then click on text here.  

With all these settings, set the default  and just put the maximum character size  

ready here, then click on "Run" to  generate a preview. See what we get.  

It's pretty clean. But like I expected,  some of these letters that were too  

close to each other didn't convert  very well. But as you can see,  

the word here that we took the time to clean up,  which is arched, converted with no problem. Again,  

it's just a matter of cleaning  up the images as much as you can.  

Like I mentioned before, if you set this to  let's say the highest possible value of 99,  

Click on "Run", you're gonna see a  lot of these words start to disappear.  

Like so, 'cause nothing can go higher than  99. So stick to the default level at 60.

Also, we have an option I forgot to mention  that has intersecting lines. I think it's  

just for these niche situations where some  of these others are kind of intersecting.  

Again, this isn't the most ideal placement  for a letter, so this should help,  

but I think for this particular  one, it's just too...  

Well, for one, the line that's intersecting is  too thick and it's intersecting where there are  

two points. So this might not be recognized,  but let's see if it helps. I think for certain  

letters that are more legible, it should help.  Yeah, no, it didn't really help that much,  

but I think for these other ones, it kind of did.  Okay. Let's click on "Okay" to finalize all of  

these conversions. Now, if you want to manually  change this, you can go to the both tab here,  

click on "Highlight Vectors" to see the  converted vectors on top of the original raster.

If you wanna turn this into a raster  object, rather no, a true type object,  

you can click on the erase tool, erase the  lines that were converted, click on the  

text tool. Actually, I wanna check how large  these characters are. So I'm gonna click on one,  

vector information says that the  size is 20. So I'm going to click on  

"Draw Text", click here, set the size to  20, and then just type in the word down.  

Let's click on the arrow tool and just drag  this to its proper placement over here.  

What else do we wanna change? Okay,  for example, we have this one where  

part of the word was converted, but  the rest wasn't. You can actually just  

erase the part of the word that wasn't converted  back to the both tab to see the reference.  

Click on the arrow tool, click on the  R, and then add the rest of the word.  

Click on okay. Move it a little bit  'cause it's intersecting the next word,  

so just move it over here like this. You can  take your time and clean up the whole image.

Also, if you click on the "Highlight Vectors"  button here, it just makes all of the  

unique vector line types into different colors.  So, polylines are red, arcs are magenta,  

text objects are also magenta, as you can see,  dashes are black, and if you're happy with this,  

un-tick the "Highlight Vectors" button, and  you can click on export here in the upper  

right to just save the file into whatever file  format it is that you wanna use. With the OCR  

functionality of Scan2CAD, you can convert  many various types of technical images that  

have text on them with no problem, you don't  have to worry about the polythyne text anymore.