Hey I'm looking for a good ocr api/tool that I could build into my react-native app.
The tool should be able to scan bills and I want to retrieve the name of the place and the total price from the bill.Please help me out in finding a suitable tool or api.
I'm a college student so I'm not able to get my hands onto those commercials tools available online.
Thanks
I did something similar in this blog using React Native Text Detector hope this can help you. It reads the business card and returns all the text on it.
What you can do is to add Regular Expressions for detecting different attributes. Let's say billing amount will have currency logo before or after the text or for more accuracy add expected region where the billing amount should be. RNTextDetector also returns the bounding of the text on the image. There is also a mapping function that translates boundings on an image to boundings on screen for better user experience. This tutorial can be a good point to start from and then keep on adding things on top of it.
Related
I am new in reactjs and I have a task in hand. I need to be build and application which is capable to scan a mykad(Malaysian ID card) through camera. Details like name, address, image can be extracted. I googled a bit about open source tesseract but it is not giving me the right information and also some of the informations are misspelled. If anyone can guide me in the right direction.
Eventually I will develop a PWA and deploy in mobiles as well
If you're looking for a free solution, Tesseract.js is your way to go: https://tesseract.projectnaptha.com
You need to be aware that reading data from MyKads will not only require OCR component, but also specifying semantics for the document. Meaning, you'll need to tell tesseract where the name is, where the address is, etc.
Also, tesseract will not be able to detect the document on the image. For this you'll need to use a different tool.
Disclaimer: I'm working at Microblink where we develop commercial OCR products, including one for reading data from IDs. For PWAs we have an JavaScript / TypeScript component which uses WASM to process the IDs. It supports not only MyKads but more than 500 document types in the world.
Github link: https://github.com/BlinkID/blinkid-in-browser
This is my first time getting into drones.
I am looking at DJI drones, currently as it seems most promising from a documentation and reviews point of view.
Basically, I would like to program a drone(s) to fly a certain pattern and take pictures when a certain criteria is met. For example, I would like the drone to take off and fly around a small park, stopping to take a picture of each tree it encounters, automatically (auto-piloted / driven by some "AI").
Now I glanced thru the DJI SDK documentation, and so far it SEEMS this is possible (via FlightControl class). But im not sure.
Question:
Can my requirements be met with current drone SDK technologies?
Yes, the correct SDK, 4.11.1 will do everything you mentioned. You will need to do some location calculations but that's about it.
The sample will almost do everything you want as-is, with minor changes.
With the DJI Mobile SDK you can use the Mission classes to automatically fly a given set of coordinates (waypoints) and do some actions once you arrive at a waypoint, e.g. take a picture.
However the SDK has limitations:
The SDK is unable to detects objects in the video stream. Therefore it is needed to use your own code to detect objects yourself.
The way the drone flies to the waypoint is quite limited, e.g. the drone will always face the camera in the direction of flight.
When using the DJI Mission classes, a change of the route during execution is only possible with the use of timeline Missions by adding timeline elements to the list.
As you already assumed in the comment: Yes, the Mobile SDK is more advanced than Windows SDK.
I am trying to evaluate the training function of the Watson visual Recognition API.
Has anyone some experience with costumizing classifers for Visual Recognition?
I have some expierence myself with training the classifier and found some infomation in this blog:
http://christopher5106.github.io/computer/vision/2016/12/23/ibm-watson-bluemix-visual-api-to-create-custom-classifier.html
What I really would like to know is how much pictures do I need of an object to classify it with an accuracy of 75%?
How long does it take to get such a result?
Thank you in advance for your help.
The number of pictures you need depends on how unique the object is, how many distinct image features a picture with it has, etc.
To give you a few examples from my own experience:
Logo detection: one image of the logo can be used to create several samples by adding noise, changing contrast, making small distortions and rotations, etc. If the logo is detailed and has good contrast, you should easily get 75%.
Cat detection using Haar wavelets: 100 images with data augmentation can yield around 75%
Human ear detection: about 300 images could get me to around 80%. This detector is being used in an iPhone app for virtual-trying eyeglasses.
You can also try this out yourself using Kaggle's Dogs Vs. Cats data. Just try various classifiers on them with different amounts of data, and you will get a very good idea.
Hi guys im brand new and not a developer but I need a way for users when they go to my site they can upload there video and there would be a option for them to add there first name and email so when the video is uploaded the database can keep all the data together.
Ideally I want this as easy as possible for the user and this would just go to our youtube channel or any video platform will work.Any advice would be great!
Please provide more information like what platform are you using ?.
There's more than one way to skin a cat.
The simple way to achieve with web technologies like (Php,node,jave) is maintain the basic user information into the sessions, and whenever it's necessary use this information.
You need to get some knowledge about the system you are using. You particularly need:
access to the server
to know the server type
access to the database
to know the database type
where the relevant files are
After you have gathered all these information, you at least know what you do not know. The next step is to gather information about how you can implement the feature you need. Look at it like at a puzzle with many small pieces. If you are patient-enough, at the end you will resolve the puzzle.
First off, this is not a question about how to scrape websites. I am fully aware of the tools available to me to scrape (css_parser, nokogiri, etc. I'm using Ruby to do the scraping).
This is more of an overarching question on the best possible solution to scrape the logo of a website starting with nothing but a website address.
The two solutions I've begun to create are these:
Use Google AJAX APIs to do an image search that is scoped to the site in question, with the query "logo", and grab the first result. This gets the logo, I'd say, about 30% of the time.
The problem with the above is that Google doesn't really seem to care about CSS image replaced logos (ie. H1 text that is image replaced with the logo). The solution I've tentatively come up with is to pull down all CSS files, scan for url() declarations, and then look for the words header or logo in the file names.
Solution two is problematic because of the many idiosyncrasies of all the people who write CSS for websites. They use Header instead of logo in the file name. Sometimes the file name is random, saying nothing about a logo. Other times, it's just the wrong image.
I realize I might be able to do something with some sort of machine learning, but I'm on a bit of a deadline for a client and need something fairly capable soon.
So with all that said, if anyone has any "out of the box" thinking on this one, I'd love to hear it. If I can create a solution that works well enough, I plan on open-sourcing the library for any other interested parties :)
Thanks!
Check this API by Clearbit. It's super simple to use:
Just send a query to:
https://logo.clearbit.com/[enter-domain-here]
For example:
https://logo.clearbit.com/www.stackoverflow.com
and get back the logo image!
More about it here
I had to find logos for ~10K websites for a previous project and tried the same technique you mentioned of extracting the image with "logo" in the URL. My variation was I loaded each webpage in webkit so that all images were loaded from CSS or JavaScript. This technique gave me logos for ~40% of websites.
Then I considered creating an app like Nick suggested to manually select the logo for the remaining websites, however I realized it was more cost effective to just give these to someone cheap (who I found via Elance) to do the work manually.
So I suggest don't bother solving this properly with a fully technical solution - outsource the manual labour.
Creating an application will definetely help you, but I believe in the end there will some manual work involved. Here's what I would do.
Have your application store in a database a link to all images on a website that are larger than a specified dimension so that you can weed out small icons.
Then you can setup a form to access these results. You may want to setup the database table to store the website url and relationship between the url and image links.
Even if it we're possible to write an application to truly figure out if it was a logo or not seems like it would be a massive amount of code. In the end, it would probably weed out even more than the above, but you have to take into account it could be faster for human to visually parse the results then the time it took for you to write and test the complex code.
Yet another simple way to solve this problem is to get all leaf nodes and get the first
<a><img src="http://example.com/a/file.png" /></a>
you can lookup for projects to get html leaf nodes on the net or use regular expressions to get all html tags.
I used C# console app with HtmlAgilityPack nuget package to scrape logos from over 600+ sites.
Algorithm is that you get all images that have "logo" in url.
The challenges you will face with during such extraction are:
Relative images
Base url is CDN HTTP/HTTPS (if you don't know
protocol before you make a request)
Images have ? or & with query
string at the end
With that things in mind I got approximately 70% of success but some images were not actual logos.