Font Size: a A A

The Infrared Imaging Processing System Based On Directshow

Posted on:2014-07-17Degree:MasterType:Thesis
Country:ChinaCandidate:L F YanFull Text:PDF
GTID:2268330422463299Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
111recent years, with the tremendous development of the inrfared thermal imagingtechnology, it lias been widely used in various fields and has broad prospect. Limitationsof infrared device can bring in interfering noise and the imaging machanism can cause thelow contrast, the quality of inrfared imaging is not such satisfactory. Extensive reseach oninfrared image process has been eanred out all over the world to solve the problem. In thisarticle,we adopt DirectShow as a platform to design inrfared image process system. It hasadvantage such as strong expandability and high efficiency.Noniuiifoiinity and bad pixel are key factors iiifluciiig the quality of inrfared images.Accordingly, this paper analyses the production mechanism of nonunifomiity and badpixel, and explored response model of probes. For nonunifomiity correction,the paperdiscusses two methods of noiiuiiifoiiiiity correction,two-point linear noniuiifoiinityconectioiiand and polynomial least-squares fitting method. Meanwhile it realizesalgoiithnis in DirectShow filter and analyses their advantages and applying situation. Forbad pixels detections, the paper discusses a few conmion methods and realizes algorithmsin DirectShow filter.Because infrared images have features of high noise and low contrast, histogramequalization is amialysed in the paper. And a new algoritliin combining bi-platfoimhistogram equalization with image segement is proposed. Experimental results show thatthe algoritliin can enhance the object and restrain noise of infrared image, achieving bettervisual effect.Atfer the configuration of DirectShow platform and the compilation of filters, througliconnecting Source Filter, Transfomi Filter and Renderer Filter, the leal-time infrared videoprocessing can be realized.Noniuiifoiinity corrections, bad pixel detection and compensation and imageenhancement are the key technology in infrared image process field. The subject has aprofound significance to the development of PC-based real-time processing sytem oninfrared images processing.
Keywords/Search Tags:Infrared image processing system, System architecture based on DirectShow, Image enhancement, Noil uniformity correction, Bad pixels correction
PDF Full Text Request
Related items