We have taken efforts in this project. However, it would not have been possible without the kind support and help of many individuals and organizations. I/we would like to extend my/our sincere thanks to all of them.
We are highly indebted to Prof. Alpa Oza for their guidance and constant supervision as well as for providing necessary information regarding the project. We take this opportunity to thank all my friends and colleagues who started us out on the topic and provided extremely useful review feedback and for their all- time support and help in each and every aspect of the course of our project preparation. We are grateful to our college Gandhinagar Institute of Technology, Gandhinagar for providing us all required resources and good working environment.
We would like to express my gratitude towards Head of Department, Prof. Rahul Vaghela for their kind co-operation and encouragement which help us in this project
Abhishek Joglekar (160120116025)
Nowadays, many applications require zooming of a specific area of interest in the image wherein high resolution becomes essential, e.g., surveillance, forensic, and satellite imaging applications. And since, high- resolution cameras are not always affordable for everyone, that’s why high-resolution images are not still available all the time.
These problems can be overcome using image processing and machine learning algorithms, which are relatively inexpensive, giving rise to the concept of pixlarge. It provides an advantage as it may cost less, and the existing low-resolution imaging systems can still be utilized.
Pixlarge is the process of up scaling and improving the details within an image. By using state-of-the-art Deep Learning algorithm, we improve the clarity of the image.
Enhancing Microscopic images
Clarifying Low-resolution images from CCTV
License plate enhancement
Can be used in video calling applications
Image transfer at low bandwidth
Useful in photo editing
Many applications require zooming of a specific area of interest in the image wherein high resolution becomes essential, e.g., surveillance, forensic and satellite imaging applications.
However, high-resolution images are not always available. And since the setup for high- resolution imaging proves expensive so it may not always be feasible.
These problems can be overcome using image processing and machine learning algorithms, which are relatively inexpensive, giving rise to the concept of pixlarge.
It provides an advantage as it may cost less, and the existing low-resolution imaging systems can still be utilized.
The main aim of this project is that sometimes, a lack of clarity in image leads to a huge problem for many.
And not everyone can afford a good quality camera for capturing high-resolution images.
Pixlarge can work in many different scenarios where the facility of high-resolution cameras are not there.
Tools and Technology
We took this definition from the real-life problem faces by multiple organizations, as sometimes lack of clarity in image leads to a huge problem.
A Feasibility Study is a preliminary study undertaken before the real work of a project starts to ascertain the likelihood of the projects success. It is an analysis of all possible solutions to a problem and a recommendation on the best solution to use. It involves evaluating how the solution will fit into the corporation. There are various types of feasibility studies.
Operational Feasibility measures how well the solution will work in the organization and how end user & management will feel about the system. Proposed system is helpful for the scientists, researchers, developers, geologists, engineers, etc. It will allow them to upscale the low-resolution images from which they can find appropriate information. Here are the questions that will help test the operational feasibility of the project.
Is there enough support for the project from the management? From the users? If the current a system is well applied and used to the extent that person will not be able to see for a change, there may be resistance.
Answer: -Yes there is an enough and more support and motivation from the management. The existing system has got series of limitations and the quality of output of existing system is also not significant.
Will the proposed system cause harm? Will it produce poor result in any respect or area? Will loss of controlled result in any area? Will accessibility of information be lost? Will individual performance be poor after implementation, then before? Will users be affected in undesirable way? Will the systems slow performance in any area?
Answer: -No. the proposed system will not produce wrong results; it will all together wont produce wrong outputs but is also less likely to produce error. No, the accessibility to information will not be lost. Yes, if the attached file is larger than it does take bit of more time to process, due to huge amount of byte size, for sending it to the model which converts the image.
Operationally the application will be most feasible due to its strong requirement.
Due to good accuracy percentage of application, it can be trusted.
Due to its easy functionality for operators, who are not from IT background easily, can use the application.
Technical feasibility determines whether the technology needed for the proposed system is available and how it can be integrated within the organization. Technical evaluation must also assess whether the existing system can be upgraded to use the new technology and whether the organization has the expertise to use it.
The Technical feasibility in the proposed system deals with the technology used in the system. It deals with the hardware and software used in the system whether they are of latest technology or not. It happens that after a system is prepared a new technology arises and the user wants the system based on that technology. Thus, it is important to check the system to be technically feasible.
The minimum memory requirement is 8GB of RAM while 12GB is better to have for better performance. As far as software is concerned, Tensor flow and Python should be installed on the server. There should be printer attached to the network incase if the output needs to be printed.
Another thing to be considered during the feasibility study is the time limit: 4 months. Again, the main concern for me was any unexpected problems that the Python module might present to us: The part regarding Python was clearly laid out, and so it was relatively easier for us to, make out an approximate break-up of the no. of days that would take. In the end, we thought that the amount of time in hand after the Python section was completed would be enough for completion of the more challenging part.
Implementation feasibility covers two aspects. One is the technical performance aspect and other is the acceptance within the organization. Implementation feasibility determines how the proposed system will fit in the current operations and what, if any job restructuring and retraining may be needed to implement the system.
In the system implementation feasibility checks, whether the user who is going to use the system can work with the software with which the system is coded and also the mind of the user going to use the system. If the user does not understand or is able to work on the system further development is of waste.
Project Development Approach
Iterative and Incremental Model : The model used is Iterative and Incremental model. It is used because Iterative process starts with a simple implementation of a subset of the software requirements and iteratively enhances the evolving versions until the full system is implemented. At each iteration, design modifications are made and new functional capabilities are added. The basic idea behind this method is to develop a system through repeated cycles (iterative) and in smaller portions at a time (incremental).
The low resolution images will get converted to high resolution image. It will help in multiple ways. When we share images over the internet some application reduce the resolution of images to save data which can sometime result into bad result of image with this application the user can convert the low resolution image back to the high resolution image.
Currently, our model cannot enhance satellite images, but it’s beneficial. Thus, if time and resources permit, then such enhancements can be made into the system to make it robust and highly effective as well as efficient.