High Accuracy Structured Light Vision Measurement Method

Structured light vision measurement system uses a projector to project initiative light pattern to measured objects,the modulated pattern is known.After capturing the pattern projected object image using camera,the modulated pattern can be decoded and correspond information between the images and pixels can be obtained.Then uses image processing and three-dimensional model transformation method are used to calculate the object 3D information.Structured light vision measurement is considered one of the most reliable three-dimensional measurement technologies.In this article we use the sinusoidal grating stripe and three-step phase shift grating method,the grating stripe is projected horizontal and vertical directions separately.The modulated image of the scene is captured by the camera and the wrapped phase image is calculated.Unwrap the wrapped phase image and the absolute phase image is obtained,which can be used to find corresponding points with its phase feature.In practical projecting process,the sinusoidal fringe image will be distorted due to various reasons,and then the captured sinusoidal fringe image will contain noise and other interferences apart from the 3D information of the measured object,so the measurement accuracy will be affected.In this article,the compensation and correction methods of sinusoidal fringe images are researched to improve the accuracy of the sinusoidal fringe measurement.The main contents include:First,an overview of the methods and principles used to measure the three-dimensional objects,highlights to the structured light sinusoidal fringe measurement.The feature of sinusoidal fringe and main elements affecting the accuracy of the method are summarized.Secondly,the gamma value calculation and compensation method is researched.Gamma model is established.Sinusoidal fringe patterns are projected to a standard flat surface,through capturing the distorted image;the gamma value can be calculated and compensated.Finally the compensated phase can be obtained.Again,the gamma compensation method based on look up table is researched.The sinusoidal fringe patterns are projected to the standard flat surface,the actual sinusoidal fringe patterns and the ideal sinusoidal fringe patterns are compared.Then the fringe distortion level is calibrated and stored into a look up table.It can be used to compensate the measurement error caused by the gamma distortion in practical measurement.Then,the intensity saturation compensation method based on additional image is researched.By adding three additional sinusoidal fringe patterns with inverse intensity,the platform phenomenon occurred by the too strong intensity can be compensated.Finally,the 3D reconstruction theory and method is discussed.According to the 3D measurement method,the measured object is reconstructed by the uncompensated method and compensated method,separately.Then the obtained 3D point cloud images are compared.The effectiveness of proposed methods are proved.