First post on SO; hopefully I am doing it right :-)
Have a situation where users need to upload and view very high resolution files (they need to pan, tilt, zoom, and annotate images). A single file sometimes crosses 1 GB so loading complete file on client side is not an option.
We are thinking about letting the users upload files to the server (like everyone does), then apply some encryption on server side creating multiple, relatively small low resolution images with varying sizes. We then give users thumbnails with canvas size option on the webpage for them to pick and start their work.
Lets assume a user opens low grade image with 1280 x 1028 canvas size. Image will be broken into tiles before display, and when user clicks on a title it will be like zooming in to a specific tile. Client will send request to the server asking for higher resolution image for the title. Server will send the image which will be broken into titles again for the user to click and get another higher resolution image from server and so on ... Having multiple images with varying resolution will help us break images into tiles and serve user needs ('keep zooming in' or out using tiles).
Has anyone dealt with humongous image files? Is there a preferred technical design you can suggest? How to handle areas that have been split across tiles is bothering me a lot so not sure how above approach can be modified to address this issue.
We need to plan for 100 to 200 users connected to the website simultaneously, and ours is .NET environment if it matters
Thanks!
The question is a little vague. I assume you are looking for hints, so here are a few:
I see uploading the images is a problem in the firstplace. Where I come from, upload-speeds are way slower than download speeds. (But there is litte you can do if you need your user to upload gigabytes...) Perhaps offer some more stable upload than web. FTP if you must.
Converting in smaller pieces should be no big problem. Use one of the availabe tools. Perhaps imagemagick. I see there is a .net wrapper out: https://magick.codeplex.com/
More than converting alone I think it is important not to do it everytime on the fly (you would need a realy big machine) but only once the image is uploaded. If you want to scale you can outsource this to another box in the network.
For the viewer. This is the interessting part. There are some ready to use ones. Google has one. It's called 'Maps' :). But there is a free alternative: OpenLayers from the OpenStreetmap Project: http://wiki.openstreetmap.org/wiki/OpenLayers All you have to do is naming your generated files in the right way and a litte configuration.
Even if you must for some reasons create the tiles on the fly or can't use something like OpenLayers I would try to stick to its naming scheme. Having something working to start with is never a bad idea.
Related
First, you need to know that I'm a beginner in this subject. Initially, I'm an Embedded System Developpers but I never worked with image recognition.
Let me expose my main goal:
I would like to create my own database of Logos and be able to
recognize them in a larger image. Typical application would be, for
example, to make a database of pepsi logos and coca-cola logos and
when I take a photo of a bottle of Soda, it tells me if it one of
them or an another.
So, here is my problem:
I first wanted to use the Auto ML Kit of Google. I gave him my
databases so it could train itself on it. My first attempt was to
take photos of bottle entirely and then compare. It was ok but not
too efficient. I then tried to give him only logos but after
training, it couldnt recognize anything in the whole image of a
bottle.
I think I didn't give enough images in the first case. But I'd prefer to use the second case (by giving only logo) so that the machine would search something similar in the image.
Finally, my questions:
If you've worked with ML Kit from Google, were you able to train a
model by giving images that should be recognized in a larger image?
If yes, do you have any hints to give me?
Do you know reliable software that could help me to perform tests of this kind? I thought about Azure Machine Learning Studio from
Microsoft (since I develop on Visual Studio).
In a first time, I'd like to code as few as I can just for testing. Maybe later I could try to code my own Machine Learning System but I think it's a big challenge.
I also thought that I would need to split my image in smaller image and then send each of this images into the Machine but it would be time consuming and I need a fast reaction (like < 2 seconds).
Thanks in advance for your answer. I don't need complete answer with full tutorial (Stack Overflow is not intended for that anyway ^^) but just some advices would already be good.
Have a good day!
Azure’s Custom Vision is great for this: https://www.customvision.ai
Let’s say you want to detect a pepsi logo. Upload 70 images of products with the logo on them. Use Custom Vision to draw a box around the logo for each photo. Click “train”, and you get a tensorflow model with code.
Look up any tutorial for it, it’s pretty incredible and really easy to use.
I have about 2,000 images of cars, most pointing right, but some pointing left.
I'd like to find a way of automatically tagging a car with it's direction (new images will be coming in continually).
I'm struggling to get started and wondered if this kind of image detection problem has a name that may help my searches. Is object orientation detection a thing?
I'm a software developer (not doing much ML or Image stuff) and have a ton of azure and gcc resources available, but I can't find anything to solve this. Azure Cognitive Service can tell us it's a car in the picture, but doesn't tell us the direction.
Could just do with a good starting point to get going.
Should add, the images are quite clean on white backgrounds, examples:
Thanks to Venkata for commenting, it was a bad dataset causing our issues (too many rights vs left).
Here's what we did to get it all working:
We set up a training and prediction instance in azure (using custom vision cognitive services in our portal).
We then used https://www.customvision.ai/ to set everything up and train the model (it's super simple).
We didn't actually need any left facing images in the end, we just took all the right facing images we had (about 500 in the final instance), we uploaded them all with the tag "Right". We then mirrored all the images with a photoshop script and then uploaded them all again with a "Left" tag. It trained for about 15 minutes and we ended up with a 100% prediction score. We tested it with a load of images that weren't contained in the training set to confirm it was all working.
We then did the same for a ton of van/truck images, these were taken from a different angle (cars were all side profile shots, the vans were all front 3 quarter so we weren't sure if we'd have the same success).
Again, we flipped the images ourselves to create the left images so we only needed to source right facing vans to create the whole model.
We ended up with a 99.8% score, which is totally acceptable for our use case and we can now detect all cars and van directions and it even detects car directions that are front 3 quarters and vans that are in profile (even though we only trained cars in profile and vans in 3 quarter).
The custom vision portal gives you an API endpoint and a key, now when we detect a new image in our system it goes via the API (using the custom image sdk/nuget in our .net site) and we check the tags to see if it needs flipping. If it does, we flip it and save it back to the disk and it's then cached so it doesn't keep hitting the API.
It's pretty amazing, it took us just two days to research the options, pick a provider and then implement the solution in to a production platform. It's probably a simple use case for ML, but 10 years ago (or even 5) we couldn't have dreamed that things would have come along so far.
tldr; If you need to detect if an object in an image is pointing left or right, just grab a lot of right facing examples and then flip them yourself to create a well balanced model. Obviously, this relies on the object looking the same from one side to the other.
My team and I are building an iOS application. We allow technicians in the field to upload images for certain issues they are resolving on technical equipment. It will be important to zoom in (so keep quality relatively high) when these images are uploaded to S3.
Recently we decided to add thumbnails because it will be much faster when others browse the iOS app, rather than downloading a 1.5-2.5mb image.
My co-worker decided the best way to handle this is to generate a 200-500kb thumbnail in iOS then upload the image and the thumbnail to s3.
I voiced my concern that some of our technicians may be in some parts of the world where internet is slow and data usage is limited. So doing all this additional work on the device and uploading makes no sense to me. However the team considers this a good solution and will move forward. I've shown them easy examples of how to generate thumbnails from S3 and Lambda automatically on the server... allowing us to either upload higher fidelity images with the additional bandwith or just increase the speed of the app by uploading much less. Sometimes a user may upload as many as 100 images... meaning an additional 20-50mb...
Anyways I wanted to hear some answers about how you guys think the best way to handle this is, mainly for my own sanity check.
I don't completely comprehend the intricacies of your project, but from experience, I have one word for you - Cloudinary. As opposed to S3, which a general purpose Cloud storage solution, cloudinary is designed to handle images.
We have a 200,000 hits a day online classified app that handles tens of thousands of photos daily. And cloudinary provides an extremely mean solution for all our needs. We have uploads by users from their mobile and desktop devices, bookmarking of those images, CDN based serving, and thumbnail generation.
Did I mention they have thumbnail generation built in? They have lots of other features as well, including
Resize and Crop
Optimized JPEG Custom Crop
Face Thumbnail
Rotated Circular Thumbnail
Zoom Effects and Zoom Image Overlay
Watermark Image
Optimized WebP
Overlay, Border, Shadow Text Overlay, Border, Shadow etc.
The admin console is pretty kickass too, with all of the above features available for you to configure over the cloud. And it fits well with pretty much any application (we use it in our internal Ruby, Go, NodeJS services, our Web Application and our iOS and Android apps as well).
I'm not paid to sell Cloudinary to you, but I can vouch that if it is image based services I needed, I would go for Cloudinary any day over S3. Major players like EBay and TED etc. use it for their image requirements.
please bear with me as I'm not trying to frustrate anyone with inane questions, and I did google search this but I couldn't really find anything recent or helpful.
I am a novice programmer and I am using a classic asp web application. I just enabled the users to upload and download images, but I'm quickly regretting it as it's eating up all of the router bandwidth. I am finding my solution inadequate, so I wanted to start over.
My desire is threefold with this functionality:
Compression. I understand that this is impossible to do BEFORE uploading without some kind of Java/Silverlight/Flash portion of the application to handle uploads, correct? What is the common way most places go about this? Just allow regular file uploads and compress once they are on the server?
Resizing. I want to resize all images before they are uploaded to a reasonable size, instead of just telling users that try and upload huge camera images that they can't upload. I figure I just want to let them upload and have it resize for them before uploading. Does this functionality exist already?
Changing filetype. I want to allow users to upload all image file types but make them .jpg on the server after the upload.
With these three requirements, how hard is it to implement something like this in just pure code and libraries? Would it be better to just use a 3rd party plugin, such as ASPjpeg or ASPupload? Have you encountered something similar, and what was your solution?
Thanks.
Take a look at ASPJpeg and ASPUpload from Persits. We use these components to upload a full size image (can be png even though the library is "ASPJpeg"), resize it to several different sizes we need on our site, then store the resized images on the server in a variety of folders. The ASPUpload component is a little tricky but if you follow their sample code you'll be fine.
I never found a good component for decompressing uploaded zip files and had to write my own, which I've since abandoned. In the end with upload speeds increasing and storage getting so cheap, it started to matter less and less that the files were compressed before being uploaded.
EDIT: Just noticed you mentioned these components in your question. Consider this an endorsement of your idea to use them. :-)
I was searching on Google and StackOverflow to see if anyone have solution for my problem, but didn't found anyone with same problems.
So, currently I'm running Debian machine with Mapserver installed on it. The server also run webserver for displaying map data over the browser. The generation of map is dynamic, based on layers definition in database I built mapfile in PHP and based on that generated PHP the map is shown to user. The data is defined in database and as a SHP files (both combined in single mapfile).
It is fully dynamic, what I mean with that is that user can enable/disable any of layers or click inside polygon (select some points on map) it color the selection (generate new mapfile based on selection and re-generate tiles).
So the execution of all that code from selecting some area to coloring selected items somtimes take too much time for good user experience.
For solution I'd like to use some kind of temporary tiles cache, that can be used for single user, and to be able to delete it's content when user select some items on map or enable/disable one of the layers.
P.S. I already did all the optimizations provided from Mapserver documentation.
Thanks for any help.
It sounds to me like your problem is not going to be helped by server-side caching. If all of the tiles depend on user selections, then you're going to be generating a bunch of new tiles every time there's an interaction.
I've been using MapCache to solve a similar problem, where I am rendering a tileset in response to a user query. But I've broken up my tiles into multiple logical layers, and I do the compositing on the browser side. This lets me cache, server-side, the tiles for various queries, and sped up performance immensely. I did seed the cache down to zoom level 12, and I needed to use the BerkeleyDB cache type to keep from running out of inodes.
I'm using Leaflet.js for the browser-side rendering, but you should also consider OpenLayers.
After looking at the source code, I have some other ideas.
It looks like you're drawing each layer the same way each time. Is that right? That is, the style and predicate of a particular layer never change. Each user sees the image for that layer the same way, if they have selected the layer. But the combination of layers you show does change, based on OpenLayers control? If that's the case, you don't need per-user caching on the server. Instead, use per-layer caching, and let the user's browser figure out the client side caching.
A quick technique for finding slow layers is to turn them all of. Then reenable them one by one to find the culprit. Invoke Mapserver from the command line, and time the runs, for greater precision than you'll get by running it from your webserver.
You mentioned you're serving the images in Google 3857 while the layers are in Gauss-Kruger/EPSG 3912. Reprojecting this on the fly is expensive. Reprojecting the rasters on the fly is very expensive. If you can, you should reproject them ahead of time, and store them in 3857 (add an additional geometry column).
I don't know what a DOF file is--maybe Digital Obstacle File? Perhaps preload the DOF file into PostGIS too? That would eliminate the two pieces you think are problematic.
Take a look at the SQL queries that PostGIS is performing, and make sure those are using indexes
In any case, these individual layers should go into MapCache, in my opinion. Here is a video of a September 2014 talk by the MapCache project leader.