Documentation for Optic Flow Project

This project is divided into two main modules, the Edge Filter and the Optic Flow. The data (image) comes in as the payload of a UDP packet. If the packet is the proper size (0408 hex for a 32x32 image), then the packet is processed by the Edge Filter module, which outputs a UDP packet (keeping the same header and trailer information), with a smaller (1/8 size) payload of edge data, 1 bit per pixel. The edge data is then captured by the Optic Flow module as the reference frame, and then subsequent frames are compared to the reference frame (and become the next reference frame). The comparison results in 1 byte vectors, consisting of 2 4-bit one's-complement numbers. For example 01001110 would represent the vector (4,-6), which is 4 pixels to the right, and 6 pixels upwards. This makes the final payload 1/64 of the original payload size (0018 hex for the 32x32 case).

• Image Width (W) must be a multiple of 32 (data word)
• Image Width must be a multiple of Patch Size (default 8)
• Image Height (H) must be a multiple of Patch Size (default 8)
• Image Width, Image Height, and Patch Size (Patch) must be chosen before compilation (generic port map)
• UDP Packet must contain only image data (no headers) so that the packet UDP length field is H*W+8
• Image data is assumed to come in row-wise, that is that the first pixel input is the upper-left corner, and that the next pixels form the first row, all the way to the upper-right corner of the image

• Edge Filter
• Optic Flow

1. Hex Version
   (N/A)	(N/A)	(N/A)	(N/A)
   (2,-1)	(2,-1)	(2,-1)	(N/A)
   (2,-1)	(2,-1)	(2,-1)	(N/A)
   (2,-1)	(2,-1)	(2,-1)	(N/A)

• Captured Images

• A. Bab-Hadiashar and D. Suter, "Robust optic flow computation," Int. J. Comput. Vis. 29#1#, 59--77 #1998#.
  
  
• B. F. Buxton and H. Buxton. Computation of optic flow from the motion of edge features in image sequences. Image and Vision Computing, 2(2):59--75, 1984.
  
  
• A. Dev, B. J. A. Krse and F. C. A. Groen, "Navigation of a mobile robot on the temporal development of the optic flow", Proceedings of the 1997 IEEE/RSJ/GI International conference on Intelligent Robots and Systems (IROS'97), Grenoble, France, September 1997, pp. 558-563.
  
  
• N. Kanopoulos, N. Vasanthavada, and R. Baker, "Design of an edge detection filter using the Sobel operator," IEEE J. Solid-State Circuits 23 (1988).
  
  
• D.N. Lee. The optic flow field: The foundation of vision. Philosophical Transactions of the Royal Society of London, Series B, 290:169--179, 1980.