Based On Infrared Images Of The Ship Target Detection Technology

Target detection and recognition technology on infrared image is an important part in modern photoelectron technology field. Infrared imaging system can detect target through such obstacles as smoke, dust, fog, etc. and realize continuous passive detection day and night. Also it can observe the target detail, to recognize, locate and track target accurately. Infrared imaging system is good at concealment, has strong detection ability and long distance detection advantage. So, infrared detection and recognition system can serve as a supplement to radar and other detection system and it will play an irreplaceable role in ship collision prevention, ship navigation safety, and Vessel Traffic Service. As time goes on, development of science and technology, infrared detection and recognition system will play an important role in traffic system on the sea.Considering the characteristics of infrared image and moving ship target, this dissertation presented and designed an automatic detection system for ship target. The main task included ship target infrared characteristic analysis, infrared image preprocession, target detection algorithm research, target segmentation algorithm research, target track and forecast algorithm research, system design and several other aspects.This dissertation focused on the ship target detection and track at the port and in narrow channel. Because infrared image of ship target has characteristics of poor background contrast, edge blur, high noise, this dissertation compared many preprocession methods and put forward a new method based on wavelet. Given complex background, slight target, and no shape outline, this dissertation put forward a difference detection method based on background prediction. As to ship target in medium distance, segmentation method based on maximum entropy and improved Canny algorithm was researched. On target track and forecast aspect, considering ship moving character, Kalman filter was researched.This dissertation has finished algorithm analysis, experiment, testing, and research work on the main process part, and has designed the ship target detection system. Algorithm in the main process part has been realized on thesoftware, and ship target can be detected accurately. So the validity and feasibility of the algorithm has been verified.This dissertation analyzed and researched several main algorithm in infrared ship target detection. Because of the particularity of application, we also paid special attention to the robust adaptation and real-time of the algorithm during research. So this dissertation has certain meanings in constructing the ship target real-time detection system. In regard to the application aspect, various kinds of algorithm, which is put forward in this dissertation, can also be integrated with small-scale and portable embedded system, in order to develop the real-time target detection system of even more practical value.Chen Ren (Telecommunication and Information System )Directed by Xu Kaiyu...