Font Size: a A A

Research On Video Fusion Technology Of Fully Mechanized Mining Face In Coal Mining

Posted on:2019-05-05Degree:DoctorType:Dissertation
Country:ChinaCandidate:H J ZhaoFull Text:PDF
GTID:1361330596956030Subject:Computer application technology
Abstract/Summary:
Cyclonic static micro-bubble flotation is a new column separation method and device with China self-owned intellectual property.The successful application of this equipment in coal preparation along with its special cyclonic field structure has laid a solid base for the further application of column flotation in mineral processing.The unmanned mining of coal mines is currently the goal of all coal mines in the world.The centralized monitoring platform plays an important role in the unattended work site.The premise of centralized monitoring is that the system can fuse the videos captured by multiple cameras to form a panoramic image of a coal mine,in which video fusion technology is required.The foundation of video fusion technology is image fusion.Image fusion is the process of combining information from two or more images of a scene into a single composite image,which is more informative and more suitable for visual perception or computer processing.The goal of image fusion is to reduce uncertainty and reduce redundancy in the output while maximizing application-or task-specific information.This article focuses on the research of how to centralize the monitoring platform in the field of unattended work,and puts forward a series of video fusion technologies for the coal mine comprehensive excavation work surface.The main contents include:(1)Aiming at the problem of parallax in multi-view in the actual application process,this paper studies the video fusion method for multi-view driving face.For the multi-view video in the actual application process of parallax,the traditional video fusion technology can not accurately determine the position of the coal mine comprehensive excavation work surface,the location and size of the items.In order to obtain a more accurate scene of fully mechanized coal mining face,this paper needs a video fusion method for multi-angle driving face.The idea of this method is to extract the image features first,and combine the multi-view image matching technology to complete the image registration and matching.Finally,a locally-approximated projection transformation model is used to deal with the larger parallax.(2)Aiming at the problem that the working platform of the camera is unstable due to the jitter of the machine on the mining work surface,the video fusion adaptive method of jitter adaptive is studied.Due to the mine’s special working environment,the shaking of the mining face will cause the camera working platform to be unstable,which will cause motion blur on the imaging plane.In this paper,we propose a real-time video fusion algorithm for dynamic video sequences.The algorithm is mainly composed of two parts: Video Registration and Video Fusion For video registration,the geometric transformation between images is estimated by extracting feature points from each video frame and matching and then the features in the image are extracted at regular intervals to avoid camera translation as much as possible.At the same time,in the image fusion stage,the corresponding image transformation model is combined to avoid the appearance of distortion after image fusion.(3)Aiming at the high complexity of video fusion algorithm,this paper studies the fast face video fusion method.At present,the panoramic video fusion system often has the problem of image misalignment and ghosting in the process of rapid image fusion.At the same time,the weak light and large amount of noise in the mine pose new challenges to the performance of the panoramic fusion system.In order to solve the problem of poor real-time image fusion,this paper considers the use of a powerful GPU combined with the CUDA platform(Compute Unified Device Architecture,unified computing device architecture)to achieve rapid video image fusion.(4)On the basis of the above research work and the need of safe production in Zhaojiazhai,a dynamic video image fusion system for comprehensive working face was developed.The system completes the video image clearing process,the dynamic video image fusion of the integrated mining face,the dynamic video viewing of the working face and the development and testing of the storage function.Through on-site operation,the system is stable and reliable,and it has produced good safety,economic and social benefits.
Keywords/Search Tags:Centralized monitoring, video fusion, feature extraction, comprehensive excavation work surface
Related items