| Image Fusion techniques have many useful applications in target detection, medicine, homeland security, mine detection, automotive, road following, etc. Real time implementation of the video images has been long sought, but due to unavailability of suitable technique for the dynamically varying scenes, better techniques are required. We have explored fuzzy and neuro fuzzy logic concept for real time video data fusion.; In this thesis, we have also developed alternate approaches for image fusion using subtractive cluster neuro fuzzy algorithm. But the main objective of the whole thesis is to give algorithm for the video image fusion. According to the algorithm, we first register the video in appropriate size buffers. Video is then disintegrated into frames and suitability of the image fusion is checked. Then if required, preprocessing algorithm is applied on the frames of the Video. After preprocessing algorithm, we apply pixel wise fusion using fuzzy and neuro fuzzy concept. Frames are reassembled back for the video display. We have tried out certain algorithms for the preprocessing of the video frames according to the requirement of our test case.; The algorithms have been implemented using Matlab and it can be easily extended to C language for embedded system implementation. Real time implementation issues have been addressed and future work has been envisaged for efficient use of the resources. |
