A Research Of Detection And Recognition Technology Of The Plane Tail Flame Based On Image Processing

One of the most widely electromagnetic radiation existed in the nature is the infrared radiation, all objects will radiate infrared, and the infrared radiation detector is a kind of devices which receiving the infrared signal and outputting an infrared image. The infrared thermal imaging recognition technology which doing a processing with these images for the purpose of detecting and identifying targets is an advanced technology. It is widely used not only in the military field but also in the field of optics and image field. The detection and recognition system of the aircraft plume uses low-altitude satellites that carrying an infrared camera to detect targets, and detect and identify the coming aircraft depending on the different characteristics of the plume presented at various stages, then issues early warning signals timely, track the signals at the same time, and be ready to take the form of active attacks.This thesis uses infrared images of the aircraft plume, Firstly, do an in-depth analysis of the infrared radiation characteristics, containing the plume, the tailpipe, the aircraft skin and the context, at the same time, do a study of atmospheric transmission characteristics, both of which provide a theoretical basis for the choice of the band of the infrared detector. Secondly, do pro-processing of the plume infrared image with the infrared image characteristics and particularities in infrared images of the plume. On the base of researching, comparing and improving existing algorithms, this thesis proposes the method of plume infrared image processing, drawing that with the texture segmentation method the aircraft plume can be well separated from the background and body. Thirdly, the thesis do a depth study of the plume image in different stages, when the target is far away from the imaging plane, the target performs a point with weak signals, which can be hardly submerged from the background. This thesis proposes a multi-stage point target detection in a single image, and a time-domain detection based on multi-frame image; In the face of the target phase, the plume infrared image has some useful features contour, texture, area, etc., so at this stage this thesis proposes the recognition technology based on the extraction of the value of the feature extracting of the target.In recognition, the thesis extracts the characteristics from the following several unrelated and feasibility aspects: firstly, the physical characteristics, mainly containing complexity and external aspect ratio rectangle; Secondly, the temperature characteristic, mainly embodied the average luminance; Thirdly, the target rotational invariance, including compactness and geometric moment invariants. According to the characteristics of infrared image, this thesis proposes the target recognition method based on the Euclidean distance using the composition of the above features of the plume. Experiment shows that this method has a high recognition rate and a better prospect.Finally, the thesis gives the software implementation of the system. Use Matlab graphical user interface(GUI) as a development environment, and give out the software design and interface layout, then base on the theory and the results of the proposed algorithm which is used in target detection and recognition, improve the algorithm without compromising performance with easy to implement software, and make these algorithm be used in the platform of the detection and identification.