| Abstract: | Infrared imaging systems have excellent night visual ability, good anti-interference ability, and camouflage recognition ability, being widely used in the military field. The visible light imaging system is vulnerable to the impact of light intensity, and cannot work in closed space or at night under weak light conditions, while the infrared radiation energy of an object is only related to the temperature and material properties of the object, so infrared imaging does not need to consider the light intensity, and can work in all-weather time. Infrared target detection can be carried out in single-band and multi-band ways. Due to the limited target information, single-band detection often has problems such as the small, weak and dark targets, which will lead to unsatisfactory detection ability. However, multi-band detection greatly improves the probability of target detection and recognition, and improves the ability of recognition and camouflage by utilizing the redundancy and complementarity of information in different bands. However, there are many technical difficulties in dual-band target detection. First, the classification of infrared small target detection technology is carried out, and the technical difficulties are summarized and described according to different methods. Second, the dual-band target detection technology is described in detail, and some common algorithms are selected for performance comparison and principle analysis, highlighting the advantages of dual-band target detection. |
