Software Defined Smart Camera

Image acquisition is an integral part of digital image processing and computer vision applications. The accuracy of their algorithms is highly dependent on the quality of the input image as greater the image quality better the result of the algorithms. Usually, conventional cameras are avail

Project Title

Software Defined Smart Camera

Project Area of Specialization

Electrical/Electronic Engineering

Project Summary

Image acquisition is an integral part of digital image processing and computer vision applications. The accuracy of their algorithms is highly dependent on the quality of the input image as greater the image quality better the result of the algorithms.

Usually, conventional cameras are available with fixed functionality and image quality. They are not aware of their surrounding environmental conditions and non-adaptive to fulfil contemporary needs for producing high quality images. Because of these limitations, high quality output in uncontrolled outdoor environment conditions (regarding brightness, contrast, sharpness, saturation and intensity) is compromised. Additionally, these traditional cameras are also not automatically self-controlled. If image acquisition is performed by using these cameras, then additional steps of pre-processing will become essential for making the output images acceptable for further processing and decision making.

A more efficient technique is to replace these conventional cameras with an adaptable smart camera so that pre-processing can be performed for accomplishing the desired constant and high quality output in different environmental condition. These can be achieved through a SOFTWARE DEFINED SMART CAMERA. It acquires raw images and environment profile through sensors. Parameters for environment profiling include brightness, contrast, intensity, saturation, and sharpness. These parameters can be used to adjust the quality of the image according to weather and lighting conditions. Thereafter, a high standard output, for instance, present full-color images of day and night, can be generated at the output.

Project Objectives

To develop an efficient software-defined smart camera that can produce high quality images in uncontrolled outdoor environment conditions by maintaining standardized parameters.

• Studying and determining exact parameters which can be manipulated to produce images of high quality. 

• Use of gathered data to profile the environment. 

• Integrating models of the environment for an efficient solution. 

• Exploiting the internal architecture of the image sensor.

• Implementation of solution on FPGA.

Project Implementation Method

In the first phase, identify the most appropriate parameters and their ranges that will be used to perceive the image quality. Then, on the basis of these image attributes, a brief study and analysis will be performed for choosing the required sensors. These sensors will help in profiling. The information gathered from these sensors in real-time will help in controlling image processing techniques to make them adaptive and be responsive to the environment. This selection of algorithms, that are to be used, will be done after extensive research about computer vision techniques and its current trends towards image enhancement. It might be imperative to combine several existing computer vision algorithms alongside making necessary modifications in order to improve the image quality with consideration of all parameters. This will then lead to an adaptable program that can allow the camera to make decisions intelligently and provide a high quality image that can be used for further processing.

Benefits of the Project

This software-defined smart camera can be used in almost all computer vision applications by providing them with high quality images enhanced according to their specific environments especially for outdoor surveillance where the environment is highly unpredictable.

Other applications might include military (to predict the battle field high in the sky used in fighter jets, bomb detection and many more); medical (3D-Medical imaging used in MRI, endoscopes, tooth-scanning, inspecting eye retina, skin tissues etc.); industries (fruit sorting, pharmaceuticals, automotive mining) ; traffic control systems (to enforce law and avoid accidents). Hence, it is a ubiquitous technology. These all applications depend on the image quality of the camera. Higher the image quality better the processing. Less noise and data loss, greater accuracy. Hence, image quality is the essential element for all these applications.

Technical Details of Final Deliverable

A software-defined smart camera that has an image sensor controlled by the FPGA board enabling it to configure the sensor before the images are captured . Multiple sensors attached with the board allow it to profile the enviornment to make the system aware and hence produce high quality images which are to be be used by computer vision applications' algorithms for better calculations.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Transportation

Core Technology

Others

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com