Study On 3-D Reconstruction And Particle Tracking Velocimetry In Flame Flow

Particle tracking velocimetry(PTV)is a new particle image velocimetry technique which is developed on the basis of traditional image matching technology.The PTV particle tracking velocimetry technique is a non-invasive and non-intrusive real-time technique for particle flow velocity measurement,which has been successfully applied in the field of aerodynamics,flow field mechanics and so on.The two-dimensional particle tracking velocimetry has been widely used,but the three-dimensional particle tracking velocimetry needs to be solved urgently.Velocimetry.The two-dimensional particle tracking velocimetry can express three-dimensional flow movement trend more intuitive,which also makes the 3D image particle image tracking technology has become the key problem to study and solve.According to the characteristics of particles such as the laws of movements and so on,the binocular vision particle tracking velocity measurement technology based on PTV has been optimized and analyzed.This paper mainly includes:(1)This paper proposes an optimization method of difference: First,according to the the distribution structure characteristics and movement trend of the particles,this method makes pixel brightness of particles for a certain range weighted average,and formulate the different peak threshold of pixel brightness based on these data;Secondly,this method identify the particles and store them respectively.Through these,the particle has more observable space volume and lower time complexity in the process of binocular vision reconstruction and particle tracking velocimetry.Thus this method effectively reduce particle overlap and occlusion problem in binocular vision 3D reconstruction processing.This paper can provide a strong guarantee for the accuracy of binocular vision3 D reconstruction.(2)A PTV optimization algorithm based on feature matching is proposed.In the particle matching process,a matching matrix is built between the target particle and the candidate particle which based on proximity principle and exclusion principle,the similarity index between particles is calculated which based on characteristic of particle distance,and find the matching with similar index maximum as the matching results.This paper proposes a PTV optimization algorithm based on feature matching,in the matching process,the singular value can be iterative detection,effectively handle suddenly appearance and disappearance of particle.Removing abnormal value during the matchingprocess,the mis-matching problem is reduced and effectively improve the correct rate of matching.(3)this paper presents a PTV optimization method based on order of magnitude.This method is used to analyze the particles at short intervals,and divides the problem into some sub problem so that can reduce the time complexity of PTV matching.Since the particle itself has a strong identifiable information,the classification of different particle in the process of recognition and matching,can solve the non-contact particle mismatching problem to a great extent.If there is one to many,many to one problem in the matching process,the algorithm can make a reasonable choice.The algorithm has achieved good results in the non-intrusive and non-intrusive 3D flow field,which effectively reduces the time complexity and improves the accuracy in the PTV matching processing.Many research and analysis has been achieved for optimize particle tracking velocimetry algorithms in this paper,the comparison of the result data in the experimental processing shows that the optimization method is feasible,the flow trend of3 D PTV get good progress.