Debug Entity

https://eslamyounis1.github.io/Juno_Cam_Debug_Entity/

https://youtu.be/x1EsLI66xQ8

Another "Demo" for our project in the form of a slide presentation:

https://docs.google.com/presentation/d/1p6rnTpymK_SEGzuvE8YHflIu4dLVBIQH/edit?usp=sharing&ouid=104560781257247814054&rtpof=true&sd=true

Source Code:

GitHub - eslamyounis1/Juno_Cam_Debug_Entity

Objectives(Purpose) of our project :

Our vision is to create a solution that could be reliable and usable to provide an accurate image processing for the raw image data which we retrieve from the JunoCam website (Amateur astronomers are invited to submit images of Jupiter from their own telescopes); we are using GAN algorithms to analysis and generate auto edited image converting the raw image source that can be used for scientific, artistic, or other fun activities. Some of these images may even lead to new scientific discoveries we allowed the user to manually Edit the raw images with editing tools controlling the (Saturation, Contrast, Brightness, Hue, Grey scale and more tools), the combination between Deep Learning Algorithm such as GAN with the ability of manually editing the images will allow users to customize the image for the best result they could achieve correcting the lose that resulting from the GAN model.

How does it works:

we will discuss how the GAN algorithms work to have better understanding of the mechanism of the workflow but for the simplicity before we dig into more technical details it is basically taking the images of the raw data from JunoCam create a fake images from these sources then discriminator generates new enhanced output.

here is how the GAN algorithm works in general

A generative adversarial network (GAN) has two parts:

• The generator learns to generate plausible data. The generated instances become negative training examples for the discriminator.
• The discriminator learns to distinguish the generator's fake data from real data. The discriminator penalizes the generator for producing implausible results.

When training begins, the generator produces obviously fake data, and the discriminator quickly learns to tell that it's fake:

As training progresses, the generator gets closer to producing output that can fool the discriminator:

Finally, if generator training goes well, the discriminator gets worse at telling the difference between real and fake. It starts to classify fake data as real, and its accuracy decreases.

Here's a picture of the whole system:

Both the generator and the discriminator are neural networks. The generator output is connected directly to the discriminator input. Through backpropagation, the discriminator's classification provides a signal that the generator uses to update its weights.

backpropagation: The primary algorithm for performing gradient descent on neural networks. First, the output values of each node are calculated (and cached) in a forward pass. Then, the partial derivative of the error with respect to each parameter is calculated in a backward pass through the graph.

back to our project we used the raw images and cut the images to create sampled to discriminator then generate the enhanced images

we used a web scraping script to fitch the data from the : https://www.nasa.gov/solve/feature/junocam/

parsing over 13000 image source and having them in a separate folder then running our GAN algorithm on these data and getting out the images with enhanced quality and better details shown on the images

here is a before and after image:

all of what we have explained now is the automotive deep learning algorithm that working on the images but later we have introduced you for a manual editing app and we have create a website to do so

with the tool box on the left of the screen user could manually modify images with specific customization

GAN algorithm feature implemented to the website then apply the changes to the image if the users wants to make more enhances they could use the design tools provided by the website here is the GAN feature(under development phase)

we hope to do

image processing to generate images that can be used for scientific, artistic, or other fun activities we hope these images may even lead to new scientific discoveries!

Benefits:

Images acquired by satellites are useful in tracking of earth resources, geographical mapping, and prediction of agricultural crops, urban population, weather forecasting, flood and fire control So..,  

we work on image processing to perform these tasks well, and image processing is used in some operations on an image, in order to get an enhanced image or to extract some useful information from it and to find out various patterns and aspects in images. Pattern Recognition is used for Handwriting analysis, Image recognition, Computer-aided medical diagnosis.

tools, coding languages:

• Python for deep learning algorithm (using TensorFlow, Keras and beautifulsoup for web scraping)
• HTML,CSS and JavaScript for the visual solution

We used NASA junocam website https://www.nasa.gov/solve/feature/junocam

We accessed JunoCam website and access the Image Processing section and downloaded the raw images.

The experience with NASA was full of adventure, enthusiasm and encouragement

We all worked as a collaborating team. We thought together on choosing an idea that would benefit society and space in particular.

We learned to work together as a team and cooperate in solving problems and uniting and developing ideas,

We chose the idea of the project from the challenges on the NASA website, where we liked the idea because it is useful for the space world and works to clarify things that were not clear, which will make us able to take advantage of them and work on solving them using artificial intelligence and deep learning to solve and clarify complex and ambiguous processes in the space world,

We worked on solving the problems we encountered using constant and constant research and learning new things for us,

In the end, we would like to thank NASA and all the assistants and organizers of the hackathon, and we thank our people for giving us the opportunity to participate.

1. https://www.nasa.gov/solve/feature/junocam
2. https://www.nasa.gov/mission_pages/juno/overview/index.html