This paper focuses on two aspects of infrared polarization imaging detection. The first is to study the infrared polarization characteristics and its influencing factors , clarify the influencing factors of the infrared polarization characteristics of the surface and their action rules, establish the calculation model of the infrared polarization degree under the influence of the environment, and lay a theoretical foundation for infrared polarization detection. The second is to study the mechanism, method and performance characterization of microgrid array design, imaging non-uniformity and the evaluation of the system's ability to detect polarization. Content and results include: 1. The infrared polarization calculation model of rough surface under surrounding irrandiance. Besides the known factors such as material complex refractive index, surface roughness and observation angle, the infrared polarization degree was found to be affected by the ambient temperature difference and surrounding radiation sources in the field. Based on the concept of emissivity vector and polarization bidirectional reflectance distribution function, the calculation method of hybrid reflected radiation and the emitted radiation was proposed, and an infrared DoLP computational model on rough surface was established. The emission-incidence ratio (EIR) is the key factor affecting the degree of infrared polarization. The correctness of the model was verified by setting up an infrared polarization measurement environment and carrying out polarization measurement with different environmental temperature differences and external strong heat sources. 2. Design method of microgrid array and evaluation method of the system detecting-polarization performance. Microgrid array is the core component of polarization imaging system, the parameters of metal wire grid and microgrid parameters is simulated and calculated based on the vector diffraction theory. The effect of crosstalk introduced in the integration of microgrid array and focal plane is analyzed under different installation error. The relationship between measured polarization degree and microgrid array parameters was established. An method to evoulate the detect-polarization ability of the microgrid imaging system is proposed. The performance parameters of the microgrid array imaging system were tested, and the test results were consistent with the design simulation. 3. Mechanism and correction method of nonuniformity of microgrid array infrared imaging system. The micro-polarizers parameters deviation, installation error, micro-poalrizer thermal radiation, focal plane response non-uniformity, lens non-uniformity and other multi-factor coupling non-uniformity are analyzed. Based on 2×2 super-pixel structure, a model of polarization pixel is set up by using the incident light of Stokes vector as input. With the multiple input response equations, the system calibration parameters are obtained by reverse solving equations.An infrared polarization correction environment was set up, and a non-uniformity correction method for micro-array polarization imaging was proposed. The experiment was carried out for the medium-wave infrared system, and the non-uniformity was reduced by 5%-20% compared with the two-point method, and the imaging quality was significantly improved. 4. The Infrared anti-intereference application of infrared polarization imaging detection for target detection. Based on the infrared DoLP computational model, the influence of the strong radiation source on the DoLP of the surface is analyzed, and it showed that the external strong radiation source will enhance the polarization degree of the target. Therefore the strong radiation interference source is beneficial to the polarization detection of the target by improving the signal-to-clutter ratio. This paper also studies the pseudo-color display of infrared polarization image, analyzes the physical significance and application value of fusion, and proposes a polarization pseudo-color display method combining brightness and polarization parameter, which improves the visual perception effect of infrared polarization image.