Dear Vladimir Farber,

Look the link, maybe useful.

http://phillipmfeldman.org/Python/Advantages_of_Python_Over_Matlab.html

http://moderndata.plot.ly/eight-advantages-of-python-over-matlab/

Regards, Shafagat

One important thing to mention is that Python is free and MATLAB is paid. A lot of people use MATLAB without paying (cracking, or whatever), but doing this is a crime. If you're going to publish your results or offer services using MATLAB, you better have a license.

The community around Python is huge and is only growing more and more, apart from being really active. Besides, you have a lot of people from universities and companies which, summed up, contribute to the huge and complete ecosystem Python offers. You have libraries for virtually all fields, ready to use in your application. The language is pretty straightforward regarding syntax and there is plenty of documentation available. The learning curve is pretty good, in a way that you can learn it fast, mostly if you already know another programming language. Just to give you an idea, Python is the most popular introductory teaching language at top universities (1).

Other links that you may find useful:

https://medium.com/gradbunker/matlab-vs-python-for-scientific-computing-a-beginners-guide-a27f4dcbbc81

https://pyzo.org/python_vs_matlab.html

1 - https://cacm.acm.org/blogs/blog-cacm/176450-python-is-now-the-most-popular-introductory-teaching-language-at-top-u-s-universities/fulltext

Recently people are using Python over matlab specially when they train deep neural networks. Its quite easy to use and lots of builtin functions , sometimes hardly need to code. here you can lean python from scratch to advance. https://www.learnpython.org/

MATLAB is powerful and rich and one of the best known research tool !!!!

Python has libraries , not as rich (I presume)

Python has a library for any problem you have, I assure you. Python is free and open source, has a huge user and developer community. You can learn python easily, not only because it is easy to learn but it has lots of sources to learn from on the internet. The sources are not limited to python in general, you can easily find tutorials focused on a specific topic. Python is a full fledged, general purpose, multi-paradigm programming language that you can use to write any program you need - from simple scripts to full desktop applications, from embedded systems to server side scripts.

I recommend you to learn the latest version.

I recommend you to learn the latest version because Python is a full fledged, general purpose, multi-paradigm programming language that you can use to write any program you need

Python advantageous: Ease of use and read.

The syntax in Python helps the programmers to do coding in fewer steps as compared to Java, C++, etc. Python's syntax is easy to learn, so both non-programmers and programmers can start programming right away. Python looks more like a readable, human language than like a low-level language. This gives you the ability to program at a faster rate than a low-level language will allow you.

It depends on the problem.

Christian Bauckhage that is such a concrete and specific response to the question. I feel that the versatility of Python, beyond numerical computing makes it such a vital skill.

MATLab is awesome for math and algorithms!

Python is very good for everything, from Web to Robotics through networking, but will be less good than MATLab for maths.

Python is easy to deploy, install or bundle, MATLab will come with the licenses and you cannot just deploy a library, without third parties having all the licenses required.

Matlab is suitable software for solving Mathematic and algorithm and python software high level programming language for general purpose and wide applications.

Since MATLAB is a multi-paradigm numerical computing environment and proprietary programming language and Python is an interpreted high-level programming language for general-purpose programming. The practical advantage have Python over the MATLAB is python contains various libraries which contains various algorithms and mathematical computations which can be modified according to the condition like for imagine processing we can use various algorithms which works on assumptions which can be modified according to our condition while MATLAB has also the mathematical computations but has fixed values and cannot be modified since it is a proprietary programming language. This is the advantage of Python over MATLAB.

Python is an generic programming language and open-source platform to create various applications, it has almost all kind of libraries like machine learning, linear algebra, deep learning, signal processing and many more from web development to GUIs.

Whereas MATLAB which is not open-source is a numerical computing software mostly used for solving mathematical problems and creating applications but you cannot create executable files from the MATLAB scripts. It a good tool for analysis and that is why very famous in the academia because it provides very rich tools for plotting your functions.

OOP (Object-oriented programming) in Python is simple and elegant, while still managing to offer power and flexibility comparable to that of C++. Matlab's object-oriented programming scheme is unwieldy and overly complex.

Again, for a mathematical problem, almost every programming language other than Matlab, Python uses zero-based indexing, like, if x is a one-dimensional array, the first element is x[0]—not x[1]. Because virtually every signal processing algorithm in the literature uses summations in which indices start at zero, Matlab’s one-based indexing forces implementer to translate the indices, which is both inconvenient and a source of bugs.

Koral Pietrak Ok, But I think we need to purchase the another license for this. Am I right?

The main difference between MATLAB and Python is the price to pay.

Moreover, basically with any programming or scripting language one can do what he or she wants. It just depends on how good does this person know the language and available tools, IDEs etc.

In my opinion Python is the go-to language for a simple reason: Productification is far easier.

If you consider only the research aspect, MATLAB is better in many cases, it has a full-fledged suite that helps development and data analysis and doesn't require much development knowledge. But as soon as you need to package your algorithm into a solution, deploy to a server, deal with a bunch of other software stacks and other developers, your life will be much easier if you stick to Python

I started to use Python for image processing about four years ago after coding with fortran, MatLab, etc..

As far as I'm concerned, it is the most powerfull, complete and intuitive existing language. Nowadays it is embedded in several softwares (GIS, DEM, FEM, web applications, etc.). The other side of the medal is that it is not extremely fast compared with others like C++, but it is an issue which can be easily adressed.

Don't forget, it is distributed in a compatible General Public Licence.

Whether you are a beginner with Python, I strongly suggest you to work with "spyder", a powerful environment for Python, designed for scientists, engineers and data analysts.

@ Christian Bauckhage

Your paper seems very useful. I started read it with a pleasure.

Can I Inspire you to write some NumPy / SciPy recipe which shows how to do simple image projections?

Tank you in advance!

@ Christian Bauckhage

By "image projections" I actually mean perspective projection, dependence of image on virtual camera location for example two stereo cameras with epipolar geometry.