Research On Video Fire Bright Light Suppression Algorithm And Realization Of Detection And Alarm System

The discovery of fire played an important role in the development of human civilization.Fire reconstituted the human diet and promoted human development.But the use of fire is fraught with danger.When the fire out of human control,it formed a fire,its harm is self-evident.In order to prevent the huge casualties caused by fire,people have done a lot of research and experiments in fire prediction and monitoring.Traditional fire monitoring methods usually detect smoke or temperature to determine whether there is a fire,but such detection is limited.In addition,with the rapid development of computer vision in recent years,image processing technology has become more and more mature,which provides a lot of technical support for the development of image-based fire detection technology.Therefore,fire video detection technology based on image processing has become one of the important means in the field of fire monitoring and control,and has become a hot spot in current research.This paper designs a video fire detection and alarm system based on the study of fire video in visible light and infrared scene.The system is composed of video acquisition module,image processing module and fire alarm module.The video acquisition module mainly collects flame and smoke images.Image processing module mainly includes image preprocessing,image segmentation,detection and recognition of flame and smoke area,and study of infrared image temperature characteristics of fire.The fire alarm module gives hierarchical alarm prompts in the system interface according to the results obtained from the image processing module,and starts the sound alarm to remind the staff to make corresponding preparations.In the video acquisition module,in order to obtain fire image information and temperature information in the monitoring area at the same time,a binocular camera is used in this paper.One eye is the visible light camera,the other eye is the infrared camera.In addition to acquiring fire video,the former also has video monitoring function.The latter is mainly used to obtain infrared images of fire and study fire temperature information.This can meet the demand of obtaining both visible and infrared images in the same scene.In the video acquisition module,in order to obtain fire image information and temperature information in the monitoring area at the same time,binocular camera is adopted in this paper.One is a visible light camera and the other is an infrared camera.In addition to the former access to fire video,also has a video monitoring function.The latter mainly obtains the infrared image of fire,which is used to study the fire temperature information.In this way,both visible image and infrared image can be obtained in the same scene.In the image processing module,considering that the images collected in the monitoring area are often interfered by bright light in the scene and even overexposed,the detection and recognition of fire images are seriously affected.In view of this situation,this paper proposes an adaptive bright light suppression algorithm for fire image.First,the V component in the HSV color space is processed,and the incident light component is extracted by means of the median filter,and then the gamma adaptive brightness correction is carried out to adjust the brightness and contrast of the image.Then,MSRCR algorithm is integrated on this basis to restore the color of the areas interfered by strong light,so as to reduce the image color distortion caused by overexposure.In the process of smoke and flame detection and recognition,the static and dynamic characteristics of smoke and flame are used to identify and judge.In addition,according to the relationship between temperature and pixels in the infrared fire image,the temperature information of the object measured can be obtained by measuring the pixel value of the infrared image.In the fire alarm module,this paper designs a fire alarm mechanism.If the flame and smoke are detected simultaneously in the video image,and the temperature also exceeds the set threshold,the first-level fire alarm will be given.If any two kinds of smoke,flame and temperature exceed the set threshold are detected at the same time,a second-level fire warning is given.If only smoke is detected or only temperature exceeds the set threshold,a level 3 suspected fire warning is given.Relevant personnel can take corresponding actions according to the alarm prompt given by the interface.The ignition test of the system under simulated strong light interference shows that the fire identification algorithm can identify the fire effectively and accurately,and the system can run stably to meet the design requirements.