Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device

Research of Deinterlacing Algorithm of Video Format Conversion

Author YueLeiFei
Tutor JiangJianGuoï¼›QiMeiBin
School Hefei University of Technology
Course Signal and Information Processing
Keywords Video format conversion Deinterlacing Motion Detection Motion Estimation
CLC TP391.41
Type Master's thesis
Year 2010
Downloads 94
Quotes 1
Download Dissertation

Flashing detail with respect to , the human visual system is more sensitive to large area flicker . TV display interlaced scanning method is due to this effect , while also initially broadcast format is designed with the TV scan format to match the display , whereby the result is a video chain trace interlaced everywhere . The main disadvantage of interlaced scanning is that it makes a lot of image processing work becomes complicated, in particular, it allows the scanning format conversion becomes complicated. With HDTV, video telephony , networking, and PC video appeared in large numbers , many of scanning formats need to be converted into a broadcast format , display technology also makes defects such as line interlaced crawling, interline flicker more and more obvious , go interlaced become more important. For embedded systems and advanced de-interlacing systems for different applications , we were proposed based motion adaptive de-interlacing algorithm based on motion compensation and adaptive de-interlacing method , the paper 's main work and contributions are as follows : 1. At the traditional field-based differential motion detection technology , based on an improved four- field motion detection, using the current point eight neighborhood pixel motion to vote on the current point the way to judge the state of motion , reduce the test results of noise interference , thus reducing the probability of missed detection and false detection , so that the test results are more precise and smoother. 2 In the conventional edge average interpolation algorithm (ELA: Edge-based Line Average) , based on the relevant principles under the adjacent , an improved hierarchical edge average (RELA) interpolation algorithm , to overcome the inherent ELA algorithm disadvantages , and there is no significant increase in computational complexity. 3 In D recursive search block matching motion estimation algorithm based on motion compensation is proposed adaptive de-interlacing algorithm, which effectively avoids errors caused by motion vector steep drop in image quality , and the computational complexity is not high, ideally suited for de-interlacing applications .

Related Dissertations
More Dissertations