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 5 years ago.
Improve this question
I'm starting with an university project and I'm looking for a tool that help me to find the coordinates(X,Y) in pixels from an specific objects in an image(I'm not talking about text). I'm trying to know if IBM Watson Visual recognition could help me out to get this achieve, or if you know any other tool that could work better.
Thank you.
You can also take a hybrid "cloud-edge" approach, as described here: https://medium.com/unsupervised-coding/dont-miss-your-target-object-detection-with-tensorflow-and-watson-488e24226ef3
It uses a TensorFlow model running locally to detect regions, then uses Watson VR to say what is in each region. This combines the flexibility of TensorFlow with the ability of Watson VR to classify many many (tens of thousands of) different types of objects.
You can "kind of" do this with Watson visual recognition. First you need to train a custom classifier to "find" the objects that you are looking for. Once you have done this, you're halfway done.
The second part involves taking the image that you want to find the object in, and splitting it up into four parts (upper left, lower left, upper right, lower right). Then you search each portion of the image for your target object. If you find it in one of those quadrants, you then take that quadrant and break it up into four parts, and search each portion of the image for the target object. If you continue and do this recursively (and keep track of the pixel boundaries of each quadrant and sub-quadrant), eventually you will narrow down on the object you are searching for.
Now you should also want to do other search algorithms. Consider the case where your target object is in the center of the image - it won't show up in ANY quadrant. If your object happens to span a quadrant boundary, you will not get an accurate location, so multiple search patterns are needed, but the strategy and approach is the same.
Related
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
For the last two weeks I have been researching the GIS world for a project I have been tasked with. The end goal is to have thousands of GPS points with information bubbles when you click on the point. This was a pretty simple task and now I'm moving onto the next part of this and I need some help to point me in the correct direction for me to research. What I'm trying to accomplish is to have an Alarm go off when I'm within a certain proximity to these points. I know the OpenGEO suite along with QGis has some GPS functions but what I'm looking for is something that can be used by someone either VIA web or a PC or mobile Application that will alarm on proximity. I have also been looking into Leaflet and OpenLayers in hopes I can do it through these map library's or in junction with them.
What are some ways I could do this? Please do not give me any code because I like to research things to learn. I just need some guidance. Thanks in advance!
Have a look at ifttt.com (if this then that). Here, is a recipe that lights a philips hue light bulb when someone enters an area with android gps app. Best part, you don't have to code anything.
If your solution has to be more inhouse developed, you definitely have to read about the new mysql gis functions. IIRC, you have to use mysql 5.6 or newer.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
Inspired by a recent Kickstarter campaign: http://www.kickstarter.com/projects/dominikmazur/camfind-a-mobile-visual-search-app?ref=category
The app uses the mobile camera to take a picture and identify virtually any object. Snapping a photo of a movie poster will recognize the movie and pull up results on the web for you about it, taking a picture of a product will show you websites that product is available for sale on.
My question is, is this realistic? I find it very intriguing, but it object detection really that simple? I'm interested in some feedback regarding resources to help someone get started in learning about this topic.
Computer vision and Pattern Recognition is not easy at all. It's an entire field related to Artificial Intelligence. It is, however, relatively straightforward to understand at a high level though. There is NO WAY they are doing this all on the client. The phones just aren't fast enough, and do not have even close to enough storage space.
What they are most likely doing is sending the image to their servers, then use some kind of nearest neighbour approximation on the image, and run the result through a decision tree look-up in a massive database on images which all have some hash. This will give a close match to an image they have (assuming they have A LOT of images in there database), even if only part of the image matches. Then, using the hash, they look up some other information about that image to send to the device.
Hope that Helps!
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 6 years ago.
Improve this question
I am wondering if there is an open source scientific library for chemistry and/or physics
for C++ (or maybe C)?
I am NOT looking for simulation models, visualization, 3-d modeling.
I am just looking for a basic toolbox that would have, for example: various constants such as Ideal Gas Constant, Avogadro constant, periodic table of the elements values; molecular weight calculation; maybe basic functions implementing equations for stoichiometry, gas laws, thermal dynamics, kinetics.
Chemistry/Physics 101 kind of stuff.
I have found on SourceForge:
Christoph Steinbeck's The Chemical Development Kit (Java) on SourceForge.
NIST-const
I would take a look at the GNU Scientific Library:
http://www.gnu.org/software/gsl/
It should have most of the physical constants you're looking for. As far as the chemistry related stuff, I don't know of any basic packages that do all of those calculations right out of the box.
Update:
I forgot about another project that would have all of the periodic table stuff (and probably more):
http://openbabel.org
specifically check out:
http://openbabel.org/api/2.2.0/classOpenBabel_1_1OBElementTable.shtml#_details
Since you excluded simulations, PyQuante probably isn't what you're looking for, but you might find kinpy and thermopy interesting.
I really like this book for it:
http://www.amazon.com/Physics-Game-Programmers-Grant-Palmer/dp/159059472X
I has the physics formulas in there and the source code is available online:
http://apress.com/book/view/159059472X
I haven't seen much in the way of chemistry. Physics is more popular because it has direct affect on gaming, of course. :-)
Update: a few on chemistry:
http://www.iupac.org/inchi/
http://code.google.com/p/simsoup/
http://gchemutils.nongnu.org/
http://www.simsoup.info/SimSoup/Design/Chemistry_Subsystem.html
Kalzium, part of the KDE tools. Periodic table plus (based on the screenshots) an equation solver and more.
In High Energy Physics, two common c++ libraries/frameworks are CLHEP and ROOT. Both of these contain units and physical constants - see e.g. CLHEP Units. CLHEP also contains a lot of libraries for linear algebra, while ROOT has a lot of tools for data analysis.
Probably not entirely what you are looking for, but may be useful to someone
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
Been looking through a number of other related posts relating to flowchart software.
I have been asked to put together a document outlining some of the typical problems our users encounter with our software product.
What I would like to do, is create an interactive/online flowchart that lets users choose from 4-5 overall headings on whats wrong. Then for this to dynamically expand more choices on pinpointing the problem, and so on and so on, until they can get a resolution to their problem.
The key thing that I have not been able to find in some of the flowchart software out there, is having the click + expand element.
- I dont want all options to appear to the end-user in a huge flow chart as it will distract away from their specific issue.
- I want them to be able to click away and go down a specific avenue that will end up giving them some good things to try, based on their decisions/clicks.
I was originally thinking of perhaps putting something in Flex or Silverlight (ideally someone would have a template out there) but am now thinking of taking advantage of 3rd party (ideally free) software.
This will need to be hosted in a browser.
Any ideas?
Check out FreeMind. It's mind mapping software, so not necessarily a flowcharting tool, but you can use it for what you describe.
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 4 years ago.
Improve this question
I am looking for an open source image processing libraries or APIs those provide me to create a program which captures the license plate of a car in a parking system and gives me the number plate as a text . It would be great if I can have it in managed code (c# or java).
Any help would be greatly appreciated.
Thanks in advance
There are no libraries that are going to explicitly give you the license plate from an image if the image is passed as an argument, at least open source. You have to use multiple functions within the library to transform, manipulate, and extract the information you want. This is considered a 'solved' problem within Computer Vision. If there were in fact an open source library to do what you want, I'm sure a lot of Companies selling LPR technologies would stop selling it because it wouldn't be economically feasible.
The other problems you will have is the vast difference in license plate designs and styles. Your algorithm will have to be tweaked and tweaked constantly for ever changing license plates. For instance, in the US, the State of Florida has hundreds of license plates. It has been stated in the past that performing LPR on Florida license plates is one of the most difficult tasks.
OpenCV is the closest you will get. However, you will need to understand Fourier transforms and other advanced mathematical algorithms to derive the information you want.
This esnips site has various zipped solutions that other people have come up with. Some may or may not work.
You can also take a look at this CodeProject article on Image Recognition with Neural Networks
if it was c++
imageMagicK for image processing (good)
openCV for video recording (fair)
Magick++ (http://www.imagemagick.org/Magick++/) is great for the job you want but it's in C++.
In the VM world there you can use the Java Advanced Imaging API LeadTools for jvm imaging. I don't
know any decent open source library in .NET but there is a Ruby port of Camellia if you are interested
in (http://camellia.sourceforge.net/).