Based On The Video Image Fire Detection Design And Realization

Based on digital image processing technology and the analysis of fire flames characteristics, the thesis puts forward fire flames detection method through the validation fire color, image size threshold, similarity calculation. First, it discusses the development of traditional fire detection technology. Compared with the traditional detection technology, image-based fire detection method possesses more advantages in the large space fire detection. Then it examines the characteristics of fire development in different periods and analyzes the fire flames image which are characteristic of color and area change, edge change, shape similarity, hierarchical changes, the overall movement, flashing law and smoke. It puts forward the method of flames image processing and flames feature extraction, and verifies the fire color image based on the features of flames model and smoke color. The thesis calculates the flame area through counting pixels of flame image connected to flames region, and gets the outline of the target object's images through establishing the list. To further improve the accuracy of fire flame detection, it also employs the similarity calculation of consecutive frames each frame difference method to verify fire occurrence. Experiments show that the proposed fire flame detection method can effectively eliminate noise, accurately identify the image of fire flames, possess high detection rate, rule out other confounding factors, and enjoy huge practical value. Finally, this thesis carefully analyzes the problems concerning video sequences flame image detection and search plan in next step.