Research On Fire Risk Assessment Based On Video Image

Fire detection based on video image not only has the advantages of wide detection range,but also could quickly respond to the fire.To judge the risk level by using digital image processing technology in the building fire,according to the development status of the fire the right measures and effective rescue strategy could be taken.This technology has the very important value of use and research.The fire risk assessment based on video is to research the flame and smoke in the view of image processing,by extracting the shape,color,contour and motion feature of the flame and smoke images,to identify the fire.And it could be explore the rules of combustion and spread in video images,building the fire risk evaluation index based on video,to judge the current fire risk level.This thesis focuses on the research of smoke image segmentation algorithm,evaluation of fire risk based on video,including the following aspects:(1)Smoke image segmentation.In view of the problem existed in image fire smoke segmentation algorithm about extracting white,gray and black smoke at the same time,a new smoke image segmentation algorithm based on rough set and region growing method is proposed.It constructed the roughness histogram according to the statistical histogram of R component,extracted from RGB color space of image.And it selected appropriate wave valleys as the segmentation threshold to segment the image in a coarse way.As smoke belongs to motion information compare with background image,the frame difference method was used to extract moving region,excluding static interference.A smoke color model in the RGB color space was established with the unique color characteristics of smoke,to remove the movement interference and obtain the candidate smoke region.The seed points were selected inthe region and region growing method was used to segment image on the results of the rough set,extracted the smoke region.The results showed that the algorithm could segment white,gray and black smoke at the same time.(2)Fire risk assessment based on video.Analysis the relationship between smoke,the heat release rate and the characteristics of smoke,flame in the image.The smoke layer height,heat release rate,smoke concentration and visibility are selected as the evaluation index.The fire risk assessment index system based on video images is established.The weights of evaluation indexes is determined by using fuzzy AHP,the fire risk with fuzzy comprehensive evaluation model is evaluated based on video.It could be applied the evaluation of fire risk grade,putting out the fire and rescue strategy.The experimental result show that the fire risk assessment based on video image can objectively reflect the risk level of fire and has great application value.