Research On Depth Measurement Of Monocular Vision Image Based On Shadow Cues And Parallax Method

In order to solve the problem that the scene depth information of the measured object is lost in the process of converting the real scene to a two-dimensional image,this paper proposes a monocular visual image depth measurement scheme based on shadow clues and parallax method.The shadow clues and parallax method are combined to design and study the image acquisition system,image processing system and depth measurement system of the scheme respectively,and finally the establishment and perfection of the image depth measurement scheme are realized.Firstly,on the basis of in-depth study of image depth measurement schemes at home and abroad,a monocular image depth measurement scheme combining shadow clues and parallax method is proposed.According to the measurement scheme,the experimental equipment is selected.Through analyzing the applicability of different lighting schemes,the design adopts the lower light source dark field lighting scheme for lighting.The optical simulation software trace Pro is used to design and simulate uniform illumination of light source.According to the design requirements of the acquisition scheme and lighting scheme,the mechanical structure of the image acquisition system is designed and built.Secondly,the image processing link is studied and a fast and efficient image processing system is established.In the process of image processing,aiming at the characteristic that the noise in the collected image is located in the high frequency and low amplitude region,the soft threshold wavelet algorithm is adopted to denoise the image.Aiming at the problem of defocus blur degradation in the image,Richardson-Lucy algorithm is adopted to restore the defocus blur area to improve the accuracy of experimental data.In view of the obvious difference in color characteristics between the object to be measured and the background,as well as between regions with different tilt angles inside the model,Marr operator is used for edge recognition and segmentation of the object to be measured.When collecting the features of the object to be measured,the brightness feature is taken as the reference quantity for scene depth measurement,and the color feature of the object to be measured is converted into the surface brightness feature value in HSB space.At last,through the collection and analysis of experimental data,the scene depth measurement scheme is improved and verified by experiments.Through the image acquisition and processing of the measured object under different experimental variables,the influence of each experimental variable on the surface brightness characteristics and brightness gradient coefficient is analyzed,and the relative depth measurement method of the scene is studied.The parallax method is used to analyze the change rule of the shape characteristics of the bottom grid under different lifting heights.The absolute depth measurement method of the scene is studied,and the monocular visual image depth measurement scheme based on shadow clues and parallax method is improved.The reliability and applicability of the scheme are verified through the measurement experiment of the actual model.The experimental results show that the measurement scheme greatly improves the efficiency and accuracy of the scene depth measurement of the object to be measured,the measurement error is less than 3.5mm with good effect of scene depth measurement.