Development of web based image processing package

[Neha Chhattani]

hello
i want to create a website that will do image processing in real time. i want to load large tiff images i.e. images used in remote sensing and satellite images. I want to do processing on satellite images. is it possible to do this processing in website using django? If yes, how to start with? what are the requirements for this?

[James Schneider]

On Thu, Mar 3, 2016 at 6:25 AM, Neha Chhattani <neha.chh...@gmail.com> wrote:

hello
i want to create a website that will do image processing in real time. i want to load large tiff images i.e. images used in remote sensing and satellite images. I want to do processing on satellite images. is it possible to do this processing in website using django? If yes, how to start with? what are the requirements for this?

Not sure what you are intending with 'real-time'. Django can easily handle the front-end operations for collecting the images, etc., but I doubt you would use it directly for the image manipulation. That would be better suited for a set of purpose-built libraries and a background batch processor such as Celery. A user would submit a set of images to Django, who would store them and pass the job information off to Celery for processing. Once the job is finished, Django can manage the other tasks associated with the processed images.

https://realpython.com/blog/python/asynchronous-tasks-with-django-and-celery/

http://docs.celeryproject.org/en/latest/django/first-steps-with-django.html

You may not need a background processor if the image manipulation can be done quickly within Django (sub-second), but if it takes any longer than that, you'll likely need to implement something like Celery, otherwise your site will crumble if you have any sort of substantial traffic to your site. If Django is directly responsible for manipulating the images, the threads/processes used by Django will be held in use and will not be available for processing other incoming requests. Celery will keep the jobs in a queue and is not dependent on the request/response cycle and end-user expectations for responsiveness. ;-)

-James

[Xristos Xristoou]

@james celeny support .grid raster image type  ?

[James Schneider]

On Mar 3, 2016 3:54 PM, "Xristos Xristoou" <sax...@gmail.com> wrote:
>
> @james celeny support .grid raster image type  ?
>
>

Celery only handles the background task management, not the task itself. It would be responsible for firing another script that would handle the image manipulation, and reporting back to Django when the job is complete for the user to view the results at a later time.

-James

[Xristos Xristoou]

if i want only raster processing for remote sensing and satellite images  ?how can i do with python ?python library for that ?

Τη Πέμπτη, 3 Μαρτίου 2016 - 7:47:56 μ.μ. UTC+2, ο χρήστης Neha Chhattani έγραψε:

[Derek]

Hi

Please realise that satellite image processing and analysis is a whole specialised field on its own; relying on the user's deep understanding of how and why the image was created, and also assumes they already are well-trained and experienced in all things geospatial.

Having said that, if you are an analyst looking for tools, then gdal and numpy would both be very useful - see this post as a practical example: http://geoexamples.blogspot.co.za/2012/12/raster-calculations-with-gdal-and-numpy.html

[Suresh Saini]

Install OpenCV and build with python , then you can do real time image processing using UI , You want to real time image processing on web server then you have to need setup all configuration of OpenCV on web server and have to need web consol

[Mario R. Osorio]

I know nothing about the subject, but I stumbled upon this article I think might be useful in your endeavor:

Fundamentals of Image Processing - behind the scenes