Research Of Obstacle Distance Detection Algorithm Based On Monocular Vision

China has become the largest number of countries in the world of blind.Due to the limitation of visual impairment,walking has become the biggest problem to the blind.Now,many cities had built blind road to help blind people to walk and many guiding equipment can be used to help.In the process of guiding blind,it is important for developing guiding equipment to obtain information on the head of road accurately and measure the distance of obstacles,the main contents of this paper are as follows:First,the research of camera calibration algorithm: Traditional camera calibration uses a linear model.Although the calculation is simple,as it did not consider the impact of lens distortion,the calibration results is not very accurate.Introducing a nonlinear lens distortion model,although the calibration results are accurate,solving nonlinear equations is very complicated,it cannot meet the real-time requirement.In this paper,an improved RAC two-step method based on the linear imaging model is used to calibrate.It transforms the mixed model into a linear one.So it can avoid solving the complex nonlinear equations and takes the distortion into account.Second,the research of obstacle detection algorithm: firstly the image has some pretreatment operations including graying,median filtering,binarization and corrosion.Then sobel edge detection operator is used to analysis the edge of obstacle and it is marked by a rectangle,taking the shadow,water and another non-obstacle factors into account in order to obtain the target point coordinates.Third,the study on the imaging model of road: Based on the set of ideal road,the system uses a single-frame static imaging model to measure the distance between the obstacle and camera.The innovation lies in the establishment and calculation of detours imaging model,and the use of local parameter extraction road curves to detect,thereby improving the accuracy and applicability of the algorithmThe experimental results show that the proposed algorithm of obstacles distance detection uses a camera lens distortion model which makes the calibration results more accurate.It eliminates the interference to obstacles that from the non-obstacles during the stage of image processing,and solves the problem of bend road.The system measures the distance accurately and guides the blind to walk better.