Attention Makers


Real Time Motion Magnification

MAKERS: Reem Javed

1. Abstract: We will be using image processing to magnify the significant invisible changes and motions around us for analysis and use in various fields such as military, medical, chemical etc. A camera captures the video at 25 frames / second. After extremely efficient quality improving processing, the motion magnified output is displayed simultaneously on the LCD screen with the original video of the object placed in front of the camera. 2. Approach: To magnify small motion in each extracted region of interest, input region with RGB (Red Blue Green) intensities will be decomposed into multiple sub-bands at different spatial frequencies. Modified magnification factor will be then multiplied with filtered sub-bands and the magnified signal is added back into the original signal to obtain the final magnified output. This whole procedure will be carried out in a raspberry pi kit which is famous for its effectiveness and robustness. The system thus established will be a combination of hardware as well as software. This system is used as a standalone system in all its applications. There is no need of an external source. Raspberry Pi board works on its own. 3. Objectives: The objective of this project is to implement motion magnification in real-time for analysis of useful subtle motions that fall below human visible spectrum. Some of the highlighted objectives of this project are: Development of motion magnification code using Digital Image Processing Implementation in real-time Developing multiple applications of motion magnification Implementation on portable hardware kit to develop a stand-alone system High quality output video Efficient code Cost effective hardware 4. Design Specifications: Embedded system kit : Raspberry Pi 2 model B OS for Raspberry Pi kit: Raspbian wheezy 2015 Magnification technique: Eulerian Magnification Image processing software: opencv Programming language: python 5. Benefits of the Project: Video Magnification will be useful in a variety of applications. For example, no-contact pulse counting and vessel extraction for the patients having fatal diseases whose germs can be spread by contact. The most recent example is the outbreak of Ebola in the year 2014 when many people were infected with this virus due to contact with Ebola patients and lost their lives. Moreover this technique can also be used for detecting the spreading venom in the body if the patient is bitten by snake. This project has vast applications in military as well. This can be used for intrusion detection by enhancing small movements of hidden enemies. And it?ll be also helpful in detecting earthquakes or small movements of the surface of earth. This project finds similar applications in agriculture (time lapse videos), chemical, microscopy and microbiological applications

Home Previous Next

Vote Share Comment