Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 6 years ago.
Improve this question
What is the best OCR algorithm used to extract text from an image taken with a mobile camera?
That depends on your application requirements.
Usually OCR engine can only return characters, font, word, line or region information. You need add many other modules and tune them to get the best results.
First, image preprocessing is necessary for camera OCR application. Background, noise removal, binarization, resize ... should be added to clean image as possible as you can.
Secondly, you need locate the text region. The algorithm not only depends on image features but also on your OCR objective. For example, if you need to OCR plate number, you can use the plate number's length, width or height for more accurate location.
After OCR, you can add some post processing to correct some OCR errors. Common OCR engines such as Abyy, ExperVision, Omnipage, GOCR, Tesseract are all trained according to common documents, magazine or office paper. If your documents are special, you can establish your high-level data extraction or analysis logic over OCR layer.
You can try the OCR API at http://www.wisetrend.com/wisetrend_ocr_cloud.shtml - it's a REST API that should be very simple to use from any mobile platform, and it's based on the ABBYY OCR engine which is great for low-quality images such as those from mobile phone cameras. Disclaimer: WiseTrend is my company's customer.
Related
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 2 years ago.
Improve this question
I'm wanting to submit an app to the store and I was wondering what pngs I'm meant to submit? I'm confused as there at least 60 different types of scales, for icons, splash screens, large tiles, wide tiles, small tiles. I'm almost overwhelmed with this information. Could someone give me a run through of what is required? for my app, I'm not wanting a small tile, or large tile, or any tile at all for that matter. So I'm seeking some answers as I am struggling to find resources outlining this specific case. I hope I've been clear enough; thanks.
Is this all mandatory? What must I fill out?
enter image description here
Is this all mandatory for a splash screen?
enter image description here
Is this mandatory to fill out? What if I don't want any tiles, can I leave it as is?
strong text
Please refer to the docs for an example of a minimal required package manifest file for a desktop application.
You only need to reference a 150x150 (large) and 44x44 (small) icon image and I think you may use the same image if you want to.
Otherwise, there is a tool in Visual Studio that creates images of different sizes for you based on a single file.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
This is sort of a two part question.
First off, how can I render my own maps? Google, Bing, etc. seem to provide their own renderings that are effectively images, or so I understand. However, my objective is to be able to just get the data for streets and create my own representation from that data.
The second part is where and how can I get this information? Everywhere I look I find information on how to embed a map from Google, Bing, or whomever else.
If there is a resource that answers one, the other, or both parts, that would be awesome.
Two quick answers: OpenStreetMap and Mapnik.
OpenStreetMap is the only significant worldwide source of openly licensed street data. Some countries have their own sources - for example, in the US you can use the US Census Bureau's TIGER data (which is of very variable quality), or in the UK you can use Ordnance Survey OpenData - and if your needs are restricted to one country, that might be fine. Even so, OSM's community-created data tends to be much richer than that of national mapping agencies, though in some areas it might lack completeness.
OSM's full-planet data is a vast file and you'll need fairly serious hardware to process it, but country and regional extracts are available (for free) from third-party providers such as Geofabrik.
Mapnik is the standard rendering software, capable of 2D rendering the equal of major webmapping sites (Google etc.). It's often used as part of a package called TileMill which provides a CSS-like styling language.
You can find out more about the full toolchain, and the basics of OSM rendering, at the switch2osm.org tutorial site.
Have you checked out OpenStreetMap? They're mostly a repository for open-source map data (street layouts etc.), but the project links to renderers for their mapping data which might give you a start towards creating your own.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
What is the best way to design a website and integrate web design for mobile devices within that website?
I heard the best solution is to simply create a separate CSS file...or redirect to another page that is designed for mobiles...
And are there any tools out there that make this task easier?
A lot depends on the complexity of the site. Many times your best option is to redirect to a mobile specific site. Frameworks like jQuery Mobile can be very useful.
There are multiple ways to achieve this.
You can use responsive CSS media queries to make the same page look different on different device sizes.
In case of standard CMSs like Wordpress/Drupal, there are plugins and themes which automate this for you.
Or you can use a third-party service like MobStac for Developers or Mobify to create a mobile version of your site.
Responsive comes with its own set of limitations; for ex. the markup and content being served to all devices is the same. It does not result in a lighter or faster site by itself. You should look at adaptive instead of plain responsive if you want an optimized experience across devices for your site visitors.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Do you know of any open source (open core) implementations of an OCR for FPGA either in C or in HDL? Where can I find them?
Thanks
As Philippe says above, you will need to find an algorithm and then port that to your FPGA.
I have never heard of an open source OCR engine for specialized hardware and I have been using OCR since 1997 when there were some European machines (CGK) with OCR in hardware. About that time they moved the OCR back into software.
You might find something here.
http://www.codeproject.com/KB/cs/neural_network_ocr.aspx
http://www.codeproject.com/KB/dotnet/simple_ocr.aspx
You also need to find an algorithm that suits the types of images you want to read. Some questions you need to ask :
Are you reading a fixed size font ?
Are you reading a fixed pitch or proportional font ?
Are you processing B/W or color images ?
What resolution or DPI images are you working with ?
Do you need to remove background noise or color or perform thresholding ?
Do the images need deskewing ?
How many characters do you need ? 0-9 only or the whole alphabet ?
How fast does it need to be ?
How accurate does it need to be ?
Is the text in the same place or randomly located ? If random, how do you find the text zones ?
If your needs are fairly simple then you may get away with some if the idea in the CodeProject link. Writing a good commercial OCR engine has taken many companies years of work and they are still fine tuning to improve accuracy vs speed.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 4 years ago.
Improve this question
I'm going to create mobile version of website...
What width should I choose for design? I know that every device have it's own screen width and it's really difficult to fit all devices...
I'm really confused (pretty new to mobile websites world), please help.
Thank you.
Your approach will depend on how much effort you want to (or can) put into this and what range of mobile users you want to target. For minimal effort, simply use 100% as KennyTM points out and expect your website to be browsed on screen widths from 128px to 480px . With this approach, you'll also need to avoid 'advanced' xhtml directives such as css and divs (and javascript) because most mobile browsers cannot handle some aspects of these. Plan to use tables to manage layout.
For a more advanced approach, you can use an open source project called WURFL (http://wurfl.sourceforge.net/) which is a database of mobile browsers and their capabilities (screen width, support for div, css, images etc...). It has a jsp tag library called WNG where you write once and the tag lib will render the most appropriate html to match the user's device. I believe there is also a PHP library for this.
Rgds, Kevin.
For this sort of question, I can highly recommend reading the MobiForge developers' guide. Lots of useful advice in there, including how to pick the compromises that inevitably result from mobile development.